Try to use this articles each with a DIFFERENT study design. (i.e. cross-sectional, case-control, etc.)  Do not use case-control all three times.
answer these questions; 

2 pages

1. What is the study design?
2. Justify why you believe it is the study design you stated in #1 above. Use information you learned in this course to explain the study design.
3. What data set or data source was used? Did the authors collect their own data (primary data collection) or use pre-existing data (secondary data analysis)?

In order to receive full credit, your responses to questions #2 and #3 should be detailed and show that you have read and critically thought through the epidemiological articles you have selected.

RESEARCH ARTICLE Open Access

New parathyroid function index for the
differentiation of primary and secondary
hyperparathyroidism: a case-control study
Yanhong Guo1,2†, Qin Wang1†, Chunyan Lu1, Pianpian Fan1, Jing Li1, Ximing Luo1 and Decai Chen1*

Abstract

Background

: Patients with primary hyperparathyroidism (PHPT) may be asymptomatic, and some may present with
normocalcemic PHPT (NPHPT). Patients with vitamin D deficiency may also be asymptomatic, with normal calcium
and elevated PTH concentrations. These latter patients are usually diagnosed with vitamin D deficiency-induced
secondary hyperparathyroidism (VD-SHPT). Therefore, it is very difficult to distinguish PHPT and NPHPT from VD-
SHPT based on calcium or PTH concentrations in clinical settings. In this case-control study, we aimed to verify the
diagnostic power of a new parathyroid function index (PFindex = Ca*PTH/P).

Methods

: This study enrolled 128 patients with surgically and pathologically confirmed PHPT, including 36 with
NPHPT, at a hospital in West China between January 2009 and September 2017. Thirty-seven patients with VD-SHPT
and 45 healthy controls were selected from the population of a cross-sectional epidemiological study as the SHPT
and healthy groups, respectively. We used the PFindex to describe the characteristics of PHPT, NPHPT, and VD-
SHPT.. Differences between the four groups were compared, and a receiver operating characteristic (ROC) curve
analysis was used to evaluate the diagnostic power of PFindex.

Results

: The PHPT group had the highest PFindex (454 ± 430), compared to the other three groups (NPHPT: 101 ±
111; SHPT: 21.7 ± 6.38; healthy: 12.2 ± 2.98, all p < 0.001). A PFindex cut-off value of 34 yielded sensitivity and specificity rates of 96.9 and 97.6% and of 94.4 and 94.6% for the diagnoses of PHPT and NPHPT, respectively. The use of a PFindex > 34 to differentiate NPHPT from VD-SHPT yielded the highest positive likelihood ratio and lowest
negative likelihood ratio.

Conclusion

: The PFindex provided excellent diagnostic power for the differentiation of NPHPT from VD-SHPT. This
simple tool may be useful for guiding timely decision-making processes regarding the initiation of vitamin D
treatment or surgery for PHPT.

Keywords: Hyperparathyroidism, Parathyroid function index, Vitamin D deficiency

Background
Parathyroid hormone (PTH) is one of the most import-
ant hormones required for the maintenance of calcium
and phosphate homeostasis. This hormone induces the
1α-hydroxylation of 25(OH) D to 1,25(OH)2D, which
promotes intestinal absorption and the release of cal-
cium and phosphate from the bone [1–3], while regulat-
ing mineral reabsorption in the renal tubules. The

intrinsic abnormal excretion or extrinsic abnormal
stimulation of PTH production leads to primary, second-
ary, or tertiary hyperparathyroidism [4].
A diagnosis of classic primary hyperparathyroidism

(PHPT) can be made easily according to its biochemical,
skeletal, and renal manifestations. However, increases in
routine serum calcium testing, as well as the incidental
discovery of parathyroid nodules on thyroid ultrasonog-
raphy, has led to an increase in the detection frequency
of asymptomatic PHPT (including normocalcemic
PHPT, NPHPT) in recent decades. Thus, a correct clin-
ical diagnosis of this disease is important.

© The Author(s). 2020 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0
International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and
reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to
the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver
(http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated.

* Correspondence: cdc1309@163.com
†Yanhong Guo and Qin Wang contributed equally to this work.
1Endocrinology Department of West China Hospital, Sichuan University,
Chengdu, China
Full list of author information is available at the end of the article

Guo et al. BMC Endocrine Disorders (2020) 20:5
https://doi.org/10.1186/s12902-019-0487-8

http://crossmark.crossref.org/dialog/?doi=10.1186/s12902-019-0487-8&domain=pdf

http://orcid.org/0000-0002-7005-1298

http://creativecommons.org/licenses/by/4.0/

http://creativecommons.org/publicdomain/zero/1.0/

mailto:cdc1309@163.com

Chronic renal insufficiency and vitamin D deficiency are
the most common causes of secondary hyperparathyroid-
ism (SHPT). The former can be easily distinguished from
the medical history and laboratory tests. However, it is dif-
ficult to distinguish vitamin D deficiency-induced SHPT
(VD-SHPT) from PHPT. Patients diagnosed with either
VD-SHPT or PHPT would have an elevated PTH concen-
tration [5], and many presented with vitamin D deficiency
as well as a normal serum calcium concentration [6–8].
Therefore, it is difficult to distinguish between these dis-
eases based on laboratory results. Although the two dis-
eases can be distinguished when a patient’s vitamin D
storage is replete, a period of at least 8 weeks is required
to normalize the 25(OH) D concentration [9–11], leading
to delays in appropriate treatment. Therefore, a conveni-
ent clinical tool to differentiate PHPT from VD-SHPT.
According to the pathogeneses of these two diseases,

PHPT is associated with a relatively higher serum calcium
but lower phosphate concentration, while VD-SHPT is
characterized by relatively lower serum calcium and phos-
phate concentrations. Using this information, we created a
parathyroid function index (PFindex) to magnify the bio-
chemical differences between these diseases. This equation
multiplies the serum PTH (pmol/L) by the albumin-
corrected serum calcium concentration (mmol/L), and
then divides this value by the serum phosphate concentra-
tion (mmol/L). In this case-control study, we aimed to
verify the diagnostic power of the PFindex in subjects with
confirmed PHPT and SHPT, as well as healthy subjects.

Methods
Study design
This retrospective case-control study included 92 patients
with PHPT and elevated calcium levels, 36 with NPHPT,
37 with SHPT, and 45 healthy patients. Data were retrieved
from the PHPT registry of the Department of Endocrin-
ology at a single hospital in West China. Biochemical pa-
rameters were obtained from electronic medical records
and compared among different groups. Control groups
were selected from a community-based, cross-sectional
study conducted in Sichuan Province, China. This study
was approved by the Medical Ethics Committee of Sichuan
University, which agreed to waive the requirement for
informed consent.

Study population
This study included 128 cases of pathologically confirmed
PHPT, including 36 cases NPHPT, at a hospital in West
China between January 2009 to September 2016. Overall,
108, 11, 7, and 2 patients had parathyroid adenoma, para-
thyroid hyperplasia, parathyroid carcinoma, and ectopic
parathyroid, respectively. Thirty-seven age-matched cases
of VD-SHPT and 45 age-matched controls were selected
from a community-based, cross-sectional study conducted

in Sichuan Province, China. Subjects with a serum PTH
concentration > 6.9 pmol/L and 25-(OH) D concentra-
tion < 20 ng/mL were classified into the SHPT group. All healthy control subjects had a normal serum PTH con- centration (normal laboratory range: 14.5–62.7 pg/mL). In addition, subjects in the two control groups met all the following inclusion criteria: T-scores less than − 1 at the femoral neck, total hip, or lumbar spine measured by dual energy x-ray absorptiometry (DXA); no history of fracture, kidney disease, severe scoliosis, or hyperosteogeny of the lumbar vertebra; and serum calcium and phosphorus con- centrations within normal ranges.

Assessment of PTH, 25-(OH) D, serum calcium, and
phosphorous
The PTH and 25-(OH) D concentrations in the control
groups were measured by our Laboratory Department
using enzyme-linked immunosorbent assays (ELISAs;
Immunodiagnostic Systems, IDS Ltd., London, UK). The
interassay coefficients of variation (CV) were 4.7 and
4.6%, respectively. The serum calcium, phosphorous,
and albumin concentrations were measured using colori-
metric methods (CVs: 1.8, 1.5 and 1.5%, respectively).
Corrected calcium was calculated as the measured cal-
cium + (40 − measured albumin) * 0.02 [12]. The PFin-
dex was calculated as follows: PFindex = Ca*PTH/P.

Statistical analysis
SPSS statistical software was used for the data analysis
(version 18.0.2; SPSS Inc., Chicago, IL, USA). Normally
distributed continuous variables are presented as means
± standard deviations (SDs). Differences between groups
were tested using one-way ANOVAs after normal trans-
formation, and non-normally distributed data were eval-
uated with statistical disposal. A receiver operating
characteristic (ROC) curve analysis was performed to
evaluate the diagnostic ability of the PFindex, and ROC
curves were plotted to examine the balance between
sensitivity and specificity. To compare the diagnostic
value of the PFindex with the PTH and serum calcium
concentrations, the Youden index was calculated (You-
den index = sensitivity + specificity – 1). Statistical sig-
nificance was defined as a P-value < 0.05.

Results
Characteristics of the subjects
The PHPT group had the highest corrected serum cal-
cium, PTH, and PFindex values among the four groups
(all P < 0.05). The 25(OH) D level was higher in the healthy group than in the PHPT, NPHPT, and SHPT groups (Table 1, Fig. 1).

Guo et al. BMC Endocrine Disorders (2020) 20:5 Page 2 of 5

ROC curves for the corrected serum calcium, phosphorus,
PTH, calcium×PTH, and PFindex values
The plotted ROC curve for the PFindex yielded sensitiv-
ity and specificity rates of 96.9 and 97.6%, respectively,
at values > 34. The PFindex curve yielded a higher You-
den index than the serum calcium, phosphorus, PTH,
and Calcium×PTH curves (Tables 2, 3).

Discussion

The PFindex was the first index designed to quantify the
parathyroid function and differentiate PHPT from
SHPT. Based on our findings, the Youden index and
positive likelihood ratio of the PFindex were higher than
those of the Wisconsin Index. Consequently, this com-
prehensive index is superior to either single parameters
or pairs thereof, due to its ability to reflect the

interactions of serum calcium and phosphorus concen-
trations with PTH.
In a previous study, HaggiMazeh et al. designed the

Wisconsin Index [13] by multiplying the preoperative
serum calcium by the PTH concentration [13]. However,
this index was not used for differential diagnosis, but ra-
ther was used to help surgeons determine whether to ex-
plore the neck further or wait for PTH results after
minimally invasive parathyroidectomy. In another study
by Madeo et al., a Ca/P ratio of 2.71 was considered
valuable for the diagnosis of PHPT [14], consistent with
our study. However, in our study, the diagnostic power
of the PFindex was stronger than those of the Ca/P Ratio
and Wisconsin Index (Tables 2, 3). The Wisconsin Index
yielded a lower Youden index value than the PFindex for
the diagnosis of NPHPT and PHPT, partly because it is

Table 1 Characteristic of subjects in PHPT, NPHPT, SHPT and Health groups

PHPT(n = 92) NPHPT(n = 36) SHPT(n = 37) Healthy(n = 45)

Age (years) 47.7 ± 15.5 b 53.7 ± 15.9 46.0 ± 12.1 45.1 ± 12.3

Gender(M/F) 36/56 10/26 37/0 45/0

Serum calcium
(2.1–2.7 mmol/L)*

3.24 ± 0.82a 2.59 ± 0.10 a 2.19 ± 0.07 2.21 ± 0.11

Serum phosphate
(0.81–1.45 mmol/L)

0.77 ± 0.34a 0.79 ± 0.15 a 1.06 ± 0.15 1.07 ± 0.15

PTH
(14.5–62.7 pg/mL)

884 ± 863a 252 ± 239 94.8 ± 28.6 52.6 ± 9.09

25(OH)D
(19.08–57.6 ng/mL)

11.7 ± 4.0a 14.0 ± 5.02a 13.8 ± 2.63a 17.6 ± 6.74

Calcium×PTH 395 ± 402 72.4 ± 69.0 23.3 ± 48.4 12.3 ± 48.4

Lg Calcium×PTH** 2.31 ± 0.46 a 1.75 ± 0.26 a 1.37 ± 0.01 1.12 ± 0.10

PFindex 454 ± 430 101 ± 111 21.7 ± 6.38 12.2 ± 2.98

LgPFindex** 2.44 ± 0.46 a 1.87 ± 0.31 a 1.35 ± 0.12 1.10 ± 0.11

Ca/P 4.63 ± 1.40 a 3.35 ± 0.71 a 2.11 ± 0.34 2.10 ± 0.35

*: serum calcium was corrected by albumin. **: normal transformation
a: P<0.01 b: P<0.05

Fig. 1 Pfindex value in 4 groups

Guo et al. BMC Endocrine Disorders (2020) 20:5 Page 3 of 5

incorrect to assess an imbalance in calcium and phos-
phate homeostasis only according to the PTH level, es-
pecially when using non-optimal reference intervals for
serum PTH. The Youden index of the Ca/P Ratio was
lower than those of both the PFindex and Wisconsin
Index for the diagnosis of NPHPT and PHPT. Although
the diagnosis of PHPT is based generally on an elevated
PTH level combined with an elevated or normal calcium
level, the Ca/P Ratio ignores the PTH level.
The symptoms of PHPT are mainly attributable to an

elevated calcium concentration. However, many PHPT
patients are asymptomatic [15–17]. This phenomenon is
partly explained by a decrease in serum calcium concen-
trations due to vitamin D deficiency. In previous studies,
a high prevalence of vitamin D deficiency was observed
in both healthy individuals [18, 19] and PHPT patients
[6, 7]. Thus, the third International Workshop guidelines
recommended the exclusion of vitamin D deficiency-
induced SHPT prior to the diagnosis of PHPT [20]. It
would be better to make a final diagnosis after correct-
ing the vitamin D deficiency. As noted previously, vita-
min D repletion requires a period of 2–3 months [9]. A
PFindex > 34 could avoid this additional waiting time
and allow the arrangement of advanced tests for PHPT,
such as parathyroid ultrasound and radionuclide im-
aging. Patients with suspected hyperparathyroidism and
a PFindex < 34 could take vitamin D supplements first instead of having expensive parathyroid tests such as SPECT. Thus, the simple calculation required to gener- ate the PFindex might conserve medical resources dur- ing the diagnosis of hyperparathyroidism. This study was subject to several limitations. First, there

is no gold standard for the inclusion of SHPT subjects in
our study. Patients with SHPT and healthy controls were

selected from an epidemiology study of 1500 female resi-
dents of southwestern China, which has a low level of
daily solar exposure [21]. Notably, the prevalences of vita-
min D deficiency and vitamin D deficiency plus elevated
PTH (PTH>6.9 pmol/L) in this population were 72.5 and
40.2%, respectively (unpublished data). Conversely, the
prevalence of PHPT was so low (approximately 1–4 per
1000) that it was difficult to select PHPT patients [22].
Therefore, the recruited subjects with vitamin D defi-
ciency and slightly elevated PTH concentrations were
likely to be true SHPT patients. Second, the PHPT and
control groups were not balanced with respect to sex, as
only women were enrolled in the cross-sectional study.
However, no evidence suggests that the reference ranges
for serum calcium, phosphorus, and PTH differ between
males and females. Finally, this study did not include chil-
dren. Further studies are needed to determine whether the
PFindex can be used to differentiate NPHPT from VD-
SHPT in children.

Conclusion
The PFindex, which assesses instability in calcium and
phosphorus metabolism, was a more useful clinical diag-
nostic tool than serum calcium, phosphorus, or PTH
alone. Specifically, Subjects with PHPT had a signifi-
cantly higher PFindex than those with VD- SHPT, and
the PFindex yielded a higher Youden index when com-
pared with other indicators. A PFindex > 34 was identi-
fied as an appropriate differentiator between NPHPT
and VD-SHPT, and this value yielded the highest posi-
tive and lowest negative likelihood ratios. This tool could
help clinicians to differentiate PHPT from VD-SHPT.

Abbreviations

NPHPT: Normocalcemic PHPT; PFindex: Parathyroid function index;
PHPT: Primary hyperparathyroidism; PTH: Parathyroid hormone; VD-
SHPT: vitamin D deficiency induced secondary hyperparathyroidism

Acknowledgements

First, I would like to show my deepest gratitude to my supervisor Prof. Decai
Chen, a respectable, responsible and resourceful scholar, who has provided
me with valuable guidance in every stage of the writing of this artical. His
keen academic observation enlightens me not only in this reserch but also
in my future study. I shall extend my thanks to QW, CL, PF for all their
kindness and help. We would like to thank Editage (www.editage.cn) for
English language editing. I would also like to thank all authors who
participated this study with great cooperation.

Authors’ contributions

YG and QW designed the study and wrote the manuscript. DC made critical
revisions to the discussion and conclusions sections of the manuscript. CL,
PF, JL, and XL helped with the data collection and analysis. All authors have
read and approved the final manuscript.

Funding

No funding was received for this study.

Availability of data and materials

The dataset used in this study is available and can be provided upon written
request (Yanhong Guo, Email: guoyanhong198911@163.com).

Table 2 Sensitivity, Specificity and Youden index of serum
calcium, PTH, Calcium×PTH and Pfindex in diagnosis of PHPT

Sensitivity (%) Specificity (%) Youden index

PFindex> 34 96.9 97.6 0.945

Calcium> 2.51 95.3 97.6 0.929

PTH > 11.96 85.1 100 0.851

Calcium*PTH > 35.43 93.0 97.6 0.906

Ca/P > 2.71 91.4 93.9 0.853

Table 3 Sensitivity, Specificity and Youden index of serum
calcium, PTH, Calcium×PTH and Pfindex in diagnosis of NPHPT

Sensitivity (%) Specificity (%) Youden index

PFindex> 34 94.4 94.6 0.890

Calcium> 2.45 91.7 89.2 0.809

PTH > 11.71 94.4 75.7 0.701

Calcium*PTH > 31.73 91.7 86.5 0.782

Ca/P > 2.71 82.2 91.9 0.741

Guo et al. BMC Endocrine Disorders (2020) 20:5 Page 4 of 5

http://www.editage.cn

mailto:guoyanhong198911@163.com

Ethics approval and consent to participate

This study was approved by the Medical Ethics Committee of Sichuan
University, which agreed to waive the requirement for informed consent.

Consent for publication

This study was approved by the Medical Ethics Committee of Sichuan
University, which agreed to waive the requirement for informed consent.

Competing interests

This study did not have competing or any potential competing interests.

Author details

1Endocrinology Department of West China Hospital, Sichuan University,
Chengdu, China. 2Endocrinology Department, Hospital of Chengdu Office of
People’s Government of Tibetan autonomous Region, Chengdu, China.

Received: 29 January 2019 Accepted: 30 December 2019

References

1. Hoenderop JG, Nilius B, Bindels RJ. Calcium absorption across epithelia.

Physiol Rev. 2005;85(1):373–422.
2. van Abel M, Hoenderop JG, van der Kemp AW, Friedlaender MM, van

Leeuwen JP, Bindels RJ. Coordinated control of renal Ca (2+) transport
proteins by parathyroid hormone. Kidney Int. 2005;68(4):1708–21.

3. Dai JC, He P, Chen X, Greenfield EM. TNFalpha and PTH utilize distinct
mechanisms to induce IL-6 and RANKL expression with markedly different
kinetics. Bone. 2006;38(4):509–20.

4. Fraser WD. Hyperparathyroidism. Lancet. 2009;374(9684):145–58.
5. Eastell R, Brandi ML, Costa AG, et al. Diagnosis of asymptomatic primary

hyperparathyroidism: proceedings of the fourth international workshop [J]. J
Clin Endocrinol Metab. 2009;99(10):3570–9.

6. Jayasena CN, Modi M, Palazzo F, De Silva A, Donaldson M, Meeran K, et al.
Associations of serum 25-hydroxyvitamin D with circulating PTH, phosphate
and calcium in patients with primary hyperparathyroidism. Clin Endocrinol.
2013;78(6):838–43.

7. Souberbielle JC, Maury E, Friedlander G, Cormier C. Vitamin D and primary
hyperparathyroidism (PHPT). J Steroid Biochem Mol Biol. 2010;121(1–2):199–203.

8. Silverberg SJ, Lewiecki EM, Mosekilde L, Peacock M, Rubin MR. Presentation
of asymptomatic primary hyperparathyroidism: proceedings of the third
international workshop. J Clin Endocrinol Metab. 2009;94(2):351–65.

9. Holick MF, Binkley NC, Bischoff-Ferrari HA, Gordon CM, Hanley DA, Heaney
RP, et al. Evaluation, treatment, and prevention of vitamin D deficiency: an
Endocrine Society clinical practice guideline. J Clin Endocrinol Metab. 2011;
96(7):1911–30.

10. Peacey SR, Wright D, Harries MJ. Failure to normalize parathyroid hormone
during treatment of vitamin D deficiency in Asian patients. Clin Endocrinol.
2004;61(5):603–6.

11. Bonjour JP, Benoit V, Payen F, Kraenzlin M. Consumption of yogurts fortified in
vitamin D and calcium reduces serum parathyroid hormone and markers of
bone resorption: a double-blind randomized controlled trial in institutionalized
elderly women. J Clin Endocrinol Metab. 2013;98(7):2915–21.

12. Al-Azem H, Khan A. Primary hyperparathyroidism. CMAJ. 2011;183(10):E685–9.
13. Mazeh H, Chen H, Leverson G, Sippel RS. Creation of a “Wisconsin index”

nomogram to predict the likelihood of additional hyperfunctioning
parathyroid glands during parathyroidectomy. Ann Surg. 2013;257(1):138–41.

14. Madeo B, Kara E, Cioni K, Vezzani S, Trenti T, Santi D, et al. Serum calcium to
phosphorous (ca/p) ratio is a simple, inexpensive, and accurate tool in the
diagnosis of primary hyperparathyroidism. JBMR Plus. 2018;2(2):109–17.

15. Wermers RA, Khosla S, Atkinson EJ, Achenbach SJ, Oberg AL, Grant CS, et al.
Incidence of primary hyperparathyroidism in Rochester, Minnesota, 1993-
2001: an update on the changing epidemiology of the disease. J Bone
Miner Res. 2006;21(1):171–7.

16. Marcocci C, Cetani F. Clinical practice. Prim Hyperparathyroidism N Engl J
Med. 2011;365(25):2389–97.

17. Zhao L, Liu JM, He XY, Zhao HY, Sun LH, Tao B, et al. The changing clinical
patterns of primary hyperparathyroidism in Chinese patients: data from
2000 to 2010 in a single clinical center. J Clin Endocrinol Metab. 2013;98(2):
721–8.

18. Holick MF. High prevalence of vitamin D inadequacy and implications for
health. Mayo Clin Proc. 2006;81(3):353–73.

19. Shin YH, Kim KE, Lee C, Shin HJ, Kang MS, Lee HR, et al. High prevalence of
vitamin D insufficiency or deficiency in young adolescents in Korea. Eur J
Pediatr. 2012;171(10):1475–80.

20. Bilezikian JP, Khan AA, Potts JT Jr. Third international workshop on the
Management of Asymptomatic Primary H. guidelines for the management
of asymptomatic primary hyperparathyroidism: summary statement from
the third international workshop. J Clin Endocrinol Metab. 2009;94(2):335–9.

21. Wang SX, Yao Y, Zhou Y. Analysis of ecological quality of the environment
and influencing factors in China during 2005–2010. Int J Environ Res Public
Health. 2014 Jan;30:11(2).

22. Khan A, Bilezikian J. Primary hyperparathyroidism: pathophysiology and
impact on bone. Can Med Assoc J. 2000;163(2):184–7.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in
published maps and institutional affiliations.

Guo et al. BMC Endocrine Disorders (2020) 20:5 Page 5 of 5

BioMed Central publishes under the Creative Commons Attribution License (CCAL). Under
the CCAL, authors retain copyright to the article but users are allowed to download, reprint,
distribute and /or copy articles in BioMed Central journals, as long as the original work is
properly cited.

Abstract
Background
Methods
Results
Conclusion
Background
Methods
Study design
Study population
Assessment of PTH, 25-(OH) D, serum calcium, and phosphorous
Statistical analysis
Results
Characteristics of the subjects
ROC curves for the corrected serum calcium, phosphorus, PTH, calcium×PTH, and PFindex values
Discussion
Conclusion
Abbreviations
Acknowledgements
Authors’ contributions
Funding
Availability of data and materials
Ethics approval and consent to participate
Consent for publication
Competing interests
Author details
References
Publisher’s Note

RESEARCH ARTICLE

“Doctor, my back hurts and I cannot sleep.”

Depression in primary care patients: Reasons

for consultation and perceived depression

stigma

Ines HeinzID
1,2*, Sabrina Baldofski1, Katja Beesdo-Baum3,4, Susanne Knappe3,4,

Elisabeth Kohls
1☯

, Christine Rummel-Kluge
1☯

1 Department of Psychiatry and Psychotherapy, Medical Faculty, University Leipzig, Leipzig, Germany,

2 German Alliance Against Depression, Leipzig, Germany, 3 Behavioral Epidemiology, Institute of Clinical

Psychology and Psychotherapy, Technische Universität Dresden, Dresden, Germany, 4 Center for Clinical

Epidemiology and Longitudinal Studies, Technische Universität Dresden, Dresden, Germany

☯ These authors contributed equally to this work.
* ines.heinz@medizin.uni-leipzig.de

Abstract

Background

General practitioners (GPs) play a significant role in depression care. Recognition of

depression is crucial for adequate treatment but is impeded by a high portion of depressed

patients only reporting physical symptoms to their GP. Among the many reasons for this

phenomenon is mental health stigma. We investigated how patients with depression differed

from patients without depression regarding the types and number of complaints presented

to their GP, as well as their depression stigma. For the subgroup of patients with depression,

potential associations between perceived depression stigma and number and types of pre-

sented complaints were investigated to see if these might reflect the patient’s intention to

conceal mental health symptoms due to fear of being stigmatized by others. Further, we

investigated if perceived depression stigma is related to depression treatment.

Methods

Data on depressive symptoms (assessed by the Depression Screening Questionnaire;

DSQ), depression stigma (assessed by the Depressions Stigma Scale; DSS), type of com-

plaints reported to the GP and treatment-related factors were collected from 3,563 unse-

lected primary care patients of 253 GPs in a cross-sectional epidemiological study (“VERA

study”) in six different German regions. Data of a total of 3,069 patients was used for analy-

sis on complaints reported to the GP (subsample of the VERA study), and for 2,682 out of

3,069 patients data on a stigma questionnaire was available.

Results

Nearly half of the primary care patients with depression (42.2%) reported only physical com-

plaints to their GP. Compared to patients without a depression diagnosis, patients with

PLOS ONE

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 1 / 16

a1111111111

a1111111111
a1111111111
a1111111111
a1111111111

OPEN ACCESS

Citation: Heinz I, Baldofski S, Beesdo-Baum K,

Knappe S, Kohls E, Rummel-Kluge C (2021)

“Doctor, my back hurts and I cannot sleep.”

Depression in primary care patients: Reasons for

consultation and perceived depression stigma.

PLoS ONE 16(3): e0248069. https://doi.org/

10.1371/journal.pone.0248069

Editor: Kenji Hashimoto, Chiba Daigaku, JAPAN

Received: October 6, 2020

Accepted: February 18, 2021

Published: March 5, 2021

Copyright: © 2021 Heinz et al. This is an open
access article distributed under the terms of the

Creative Commons Attribution License, which

permits unrestricted use, distribution, and

reproduction in any medium, provided the original

author and source are credited.

Data Availability Statement: All relevant data are

within the manuscript and its Supporting

Information files.

Funding: Data for this study was obtained through

the VERA Project that was financially supported by

the German Federal Ministry of Health

(Bundesgesundheitsministerium) under the grant

number II A 5- 2513 FSB 011 (grant recipient

KBB). The sponsors had no influence on study

design, data collection and analysis, decision to

publish, or preparation of the manuscript.

https://orcid.org/0000-0001-7352-2422

https://doi.org/10.1371/journal.pone.0248069

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

http://crossmark.crossref.org/dialog/?doi=10.1371/journal.pone.0248069&domain=pdf&date_stamp=2021-03-05

https://doi.org/10.1371/journal.pone.0248069

https://doi.org/10.1371/journal.pone.0248069

http://creativecommons.org/licenses/by/4.0/

depression reported twice as many complaints to their GP with a mean of 2.02 (1.33) vs. 1.2

(0.69), including a more frequent combination of physical and mental symptoms (28.8% vs.

3.5%). Patients with depression showed higher total stigma compared to patients without

depression, Mdn = 48 (IQR 40–54) vs. Mdn = 46.3 (IQR 29–53), due to higher perceived

stigma, Mdn = 27 (IQR 21–32) vs. Mdn = 25.9 (IQR 20–29). Perceived stigma was associ-

ated with male gender (beta -.14, p = .005) and a lack of pharmacological treatment (beta

-.14, p = .021) in patients with a depression diagnosis.

Conclusion

The number of complaints presented to the GP might function as a marker to actively

explore depression in primary care patients, in particular when both physical and mental

symptoms are reported. Perceived depression stigma should also be addressed especially

in male patients. Further research should clarify the role of perceived stigma as a potential

inhibitor of pharmacological treatment of depression in primary care.

Introduction

Depression is among the leading causes for disability-adjusted life years [1]. Effective and evi-

dence-based treatments are available [2], nevertheless more than 50% of patients do not

receive depression-specific treatment [3, 4], or else experience delays to treatment due to a

variety of reasons [5, 6]. Among them are both structural and financial barriers, as well as a

low perceived need for professional help and fear of being stigmatized [5–7].

General practitioners (GPs) play a significant role in the detection, diagnosis, referral and

treatment of depression, which is defined by a high point-prevalence of depressive episodes in

primary care patients of 8–17% [8, 9]. This result was recently confirmed in an epidemiological

study of primary care patients in Germany with a point-prevalence of 14.3%, according to self-

reports [10]. Moreover, the majority of individuals with depression are being treated in pri-

mary care rather than in specialized care, while the treatment of depression according to

guidelines depends heavily on a correct diagnosis (e.g. [4]). Studies in primary care settings

have shown that only every second patient with depression is diagnosed correctly [10].

Several reasons for the false-negative detection of depression are discussed in the literature,

such as heterogeneous depression symptoms, lacking objective laboratory markers, time

restrictions and lacking reimbursements in primary care settings that impede the necessary

exploration and evaluation for a differential diagnosis [10, 11]. Further, studies revealed that

44–69% of patients with depression report only their physical symptoms [12–14]. Other stud-

ies have shown that both a correct GP diagnosis and further adequate treatment rely heavily

on the symptoms reported by the patient during consultation [10, 13, 15, 16]. The detection

rate for individuals that report only physical symptoms is therefore much lower than for those

that report physical and mental symptoms [15].

Different reasons for the association between depression and patients reporting mainly

physical instead of mental complaints have been discussed, among them the stigma associated

with the depression. Depression stigma might result in defense mechanisms such as masking

certain symptoms and emphasizing somatic aspects so as to disguise mental health problems

that may require treatment in anticipation of negative consequences [14, 16, 17].

Stigma is described as a discrediting attribute (e.g. physical attribute, religion, skin color,

mental health disorder) by which the carrier deviates from the expected social norm [18].

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 2 / 16

Competing interests: We declare the following

competing interests: CRK received lecture

honoraria from Recordati and Servier outside and

independent of the submitted work. This does not

alter our adherence to PLOS ONE policies on

sharing data and materials. KBB, SK, EK, SB and IH

have nothing to declare.

https://doi.org/10.1371/journal.pone.0248069

Among the leading concepts of stigma is the one developed by Link and colleagues, describing

stigma by use of the following five interrelated components: 1. labeling (creating labels for

social salient attributes), 2. stereotyping (linking those labels to undesirable characteristics), 3.

separating (the labeled, stigmatized group), 4. emotional reaction (of stigmatizers as well as

stigmatized persons influencing their subsequent behavior), and 5. status loss and discrimina-

tion (of stigmatized persons) [19]. Another often cited concept of stigma was developed by

Corrigan [6], who defined stigma with three core components of stereotypes, prejudices and

discrimination, while distinguishing between two types of stigma: Public stigma, which refers

to the (discriminating) reactions of the society towards a stigmatized person or group (e.g.

withholding a job) based on stereotypes (e.g. “Individuals with depression are incompetent.”),

and prejudices (believing in these stereotypes) [6, 20]. When public stigma is internalized by

an individual belonging to the stigmatized group, it may result in self-stigma [6, 21, 22],

thereby lowering the individual’s self-esteem [20, 23]. This process appears to be moderated by

further factors, e.g. if an individual is conscious about the public stigma [24] and in agreement

with public stigmatizing attitudes [20]. Two further related types of stigma have been found in

the literature: Personal stigma, which describes a person’s attitudes toward a mental health dis-

order regardless of whether he/she belongs to the stigmatized group [7, 25, 26], and perceived

stigma, which is related to public stigma and describes an individual’s belief about public atti-

tudes towards a mental health disorder [6, 7, 26–28]. Some authors have used public and per-

ceived stigma synonymously [7, 29]. Beside the different types and concepts of stigma,

different measures exist to assess the various stigma types [19, 21].

Several studies and theories can be found in the literature describing the association between

different stigma types and help-seeking for mental health disorders. Despite mixed results, there

is evidence that different types of stigma can influence different stages of the help-seeking process

(e.g. [7, 30, 31]). Self-stigma appears to be associated with less help-seeking intentions and behav-

iors at an early stage of the help-seeking process. Studies assessing self-stigma via the Self-Stigma

of Seeking Help (SSOSH) scale [31, 32] or by the Internalized Stigma of Mental Illness (ISMI)

scale [33] have also found that being labeled “mentally ill” may pose a potential threat to an indi-

vidual’s self-esteem. Like self-stigma, there is some evidence that personal stigma may also impact

help-seeking at an early stage, e.g. by not appraising symptoms as a mental health problem and a

perceived need for professional help. Schomerus and colleagues [30] assessed personal stigmatiz-

ing attitudes by observing participants’ attitudes of blaming mentally ill people (according to the

Self–Stigma of Mental Illness Scale (SSMIS) [34]), their discrimination of mentally ill people and

the social distance they maintain towards them (Social Distance Scale, [35]). The authors con-

cluded that participants who supported discrimination or blaming a person with a mental illness

showed lower self-identification of having a mental health problem themselves, as well as a lower

perceived need for help. Likewise, Griffith and colleagues reported that within a sample of 2,000

Australian adults, a positive association was found between personal depression stigma, assessed

with the DSS [25, 27], and the belief to deal with depression alone [26].

While personal stigma has been shown to impair early stages of help-seeking, i.e. avoiding

professional help, perceived and public stigma do not appear to have comparable effects on ini-

tial help-seeking [7, 17, 21, 22, 29, 30, 33]. Schomerus and colleagues [22] assessed a person´s

beliefs about public attitudes towards seeking psychiatric help with use of the 17-item ADSP

scale (anticipated discrimination when seeing a psychiatrist), revealing no association between

beliefs and help-seeking intentions. Meanwhile, the authors also found that participant´s per-

sonal attitudes, assessed by their desire for social distance, decreased their help seeking inten-

tions. A systematic review on active help-seeking and mental-health related stigma obtained

similar results: public stigma, mainly assessed with the Perceived Devaluation Discrimination

Scale [36] and its adaptations, was not linked to active help-seeking [7]. Similar results have

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 3 / 16

https://doi.org/10.1371/journal.pone.0248069

been reported for perceived stigma, which was assessed with the Perceived Devaluation Dis-

crimination Scale, and which was not found to be associated with barriers to depression care

such as a low perceived need, negative treatment expectations or treatment seeking attitudes

and behaviors [29, 33]. However, in these same studies, self-stigma, assessed by the 29-item

Internalized Stigma of Mental Illness (ISMI) scale [37], showed a strong association with barri-

ers to care and treatment seeking attitudes and behaviors respectively.

Some studies have indicated that perceived stigma may also be a relevant factor in later

stages of the help-seeking process by impeding engagement in and maintenance of depression

treatment as well as treatment outcomes [38–40]. However, the full scope of the influence of

perceived stigma remains unknown.

The current study investigates depression stigma in a large sample of primary care patients

in Germany with and without depression (based on self-reports) as well as their reported com-

plaints to the GP (including type and number of complaints). To our knowledge, there are

only two studies available that have investigated whether primary care patients with depression

differed from primary care patients without depression in regard to the number of reported

complaints to their GP [14, 41]. Both studies reported a higher number of physical

complaints

reported to the GP by patients with depression with a mean of 4.4 (SD = 4.2) and 4.5
(SD = 2.3) respectively, compared to patients without depression with a mean of 1.2 (SD = 1.9)
and 1.8 (SD = 1.3) respectively.

Based on previous literature, we state the following hypotheses:

1. In line with previous findings regarding reported complaints to the GP, we assume that

patients with depression will present more physical than mental complaints, as well as more

symptoms in total, in comparison to patients without depression [14, 41].

2. Patients with depression will not differ in personal depression stigma from patients without

depression, as personal depression stigma has been reported to deter help-seeking at an

early stage (i.e. the wish to deal with the problem alone instead of seeking professional help,

e.g. from a GP [26]).

3. With regard to perceived depression stigma, patients with depression will show higher stigma

scores than patients without depression, as has been found in other studies [25, 41, 42].

4. Further, for the subgroup of patients with depression, we will explore to which degree per-

ceived depression stigma is relevant at this stage of the help seeking process (consulting a

GP). This will be determined through associations between the number and types of

reported complaints for consultation (mental vs. physical), as well as the factors related to

the treatment of depression (type of treatment, help-seeking behaviors and/or referral to

specialized care).

Materials and methods

Sample and procedures

As part of a cross-sectional epidemiological study investigating the diagnosis and treatment of

depression in the primary care setting in Germany (VERA study), 269 randomly selected GPs

of six regions in Germany (Dresden, Leipzig, Frankfurt/ Kassel/ Fulda, München, Berlin,

Hamburg) stratified by location (city, town, rural) were recruited to participate in a survey at

the end of 2013 and the beginning of 2014 (response rate 5.8%). These regions were selected

based on their representativeness for the heterogeneous geographical situation within the fed-

eral territory of Germany. In total, 253 GPs and 3,563 unselected patients (response rate 55.9%

of suitable patients) took part in the survey by completing a GP and a patient questionnaire,

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 4 / 16

https://www.sciencedirect.com/topics/neuroscience/psychopathology

https://doi.org/10.1371/journal.pone.0248069

respectively. The patient questionnaire consisted of a compulsory Part A and a voluntary Part

B. Part A contained the Depression-Screening-Questionnaire (DSQ; [43]) and asked patients

to state the complaints for which they consulted their GP. Meanwhile, Part B contained the

Depression Stigma Scale (DSS; [27, 28]). As illustrated in Fig 1, 64 of the 3,563 unselected

patients were excluded (no patient questionnaire available), resulting in 3,499 patients with

questionnaire data suitable for analysis. Another 132 cases were excluded because there was no

corresponding GP questionnaire. Of the remaining 3,367 patients, 298 cases were excluded

due to missing data in the DSQ items and/or items to assess complaints to consult the GP for

this analysis. Therefore, the final sample in which we analyzed complaints reported when con-

sulting a GP with regard to depression diagnosis consisted of 3,069 patients. Of these patients,

387 cases were excluded due to missing data in the DSS. Depression stigma analysis were

therefore performed for a sub sample of 2,682 patients (see results section). A detailed descrip-

tion of the design of the VERA study can be found elsewhere [4, 10].

Written informed consent was obtained prior to study participation from all participating

GPs and patients. The study was approved by the ethics committee of the Technische Universi-

tät Dresden on 2013.10.07 under the reference number EK 392102013 and according to the

Declaration of Helsinki.

Instruments

Sociodemographic data. Sociodemographic information including age (in years), gender,

marital and occupational status was collected. Marital and occupational status was dichoto-

mized into “single” vs. “not single” and “occupied” vs. “not occupied”, respectively.

Fig 1. Flow chart of sampling process.

https://doi.org/10.1371/journal.pone.0248069.g001

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 5 / 16

https://doi.org/10.1371/journal.pone.0248069.g001

https://doi.org/10.1371/journal.pone.0248069

Depression diagnosis. Depressive symptoms were rated by patients according to ICD-10

criteria using the Depression-Screening-Questionnaire (DSQ; [43]). In total, 12 items regard-

ing depressive symptoms during the preceding 2 weeks were rated on a three-point Likert-

scale (0 = “not at all”, 1 = “sometimes”, 2 = “most of the days”). If at least 3 items were

endorsed with “most of the days” and the sum score was higher than 7 (i.e. at least 4 symptoms

were present), the study diagnosis “depression” was coded (according to [44]). Depression

severity was rated as follows: “mild depression” if at least 3 items were indicated for “most of

the days” and the sum score exceeded 7, “moderate depression” if at least 5 symptoms were

indicated for “most of the days” and the score exceeded 11, and “severe depression” if at least 7

symptoms were indicated for “most of the days” and the score exceeded 16 (according to [4]).

Depression severity was dichotomized into “mild” vs. “moderate or severe”. The DSQ demon-

strated good internal consistency (Cronbach’s alpha of 0.83) as well as high inter-rater reliabil-

ity (kappa = 0.84–0.89) in a German study with primary care patients [45].

Complaints reported to the GP. Patients provided information regarding the complaints

they reported during their consultation with a GP through use of a multiple-choice question-

naire containing seven response categories (physical complaints or illnesses; sleeping prob-

lems; pain; depression or depressiveness or desperation; anxiety problems; other mental health

problems; another reason). Multiple answers were possible. For the category “another reason”,

follow up appointment and referral due to any emergency were given without asking for the

type of complaint. “Physical complaints” were counted if a patient indicated a physical com-

plaint or illness, sleeping problems or pain as the reason they consulted the GP. “Mental com-

plaints” were counted if a patient indicated they sought help from a GP for either depression,

depressiveness or desperation, anxiety problems or other mental health problems. “Physical

and mental complaints” were counted if a patient indicated at least one reason from each of

the aforementioned categories. The items used to assess complaints reported to the GP were

developed by the researchers (psychologists, senior psychiatrists and GPs), as there was no

established instrument available.

Depression treatment and treatment-related factors. Patients who endorsed at least 2

DSQ items with either “sometimes” or “most of the days” were asked to document their cur-

rent and planned future treatment of depression. In addition, GPs provided information for

each patient regarding the type of treatment he/she had received prior to the reference date

(i.e. the day when the questionnaires were completed), as well as any other treatment of

depression. In order to maximize sensitivity in identification, treatments were assumed to be

present if they were mentioned by the patient and/or the GP. Treatment was categorized

according to Trautmann [4] into (1) psychotherapy, (2) antidepressants, (3) other treatment,

and (4) no treatment. According to the National Disease Management Guidelines, Unipolar

Depression, the evidence-based recommendations for diagnosis and treatment of unipolar

depression in Germany [46] with either psychotherapy or antidepressants are indicated to

treat mild to moderate depression, whereas a combination of both is indicated in the case of

severe depression. For a more detailed description of the categorization of treatments, please

refer to Trautmann [4] (supplementary material, eTable 1).

To assess referral to and/or help-seeking from specialized care (i.e. psychiatrist, psychother-

apist, inpatient treatment), patients were asked if their GP had referred them to specialized

care or if they had sought specialist help by themselves due to their depressive symptoms.

Depression stigma. Attitudes towards depression were assessed by the standardized

Depression Stigma Scale (DSS), a commonly used instrument to assess depression stigma in

the general public as well as depressed individuals [27, 28]. The DSS measures perceived and

personal stigma with 18 items which are scored on a five-point Likert Scale ranging from 1 =

“strongly disagree” to 5 = “strongly agree” [28]. Items cover common prejudices including

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 6 / 16

https://doi.org/10.1371/journal.pone.0248069

depression as weakness of character or personal fault, unpredictability and dangerousness,

shame, avoidance, and discrimination. For nine items the participant indicates to what extent

he/she agrees with a statement reflecting personal depression stigma (e.g. “Depression is a sign

of personal weakness.”). In the remaining nine items, the participant rates what he/she thinks

the broad public believes about these same statements (e.g. “Most people believe that depres-

sion is a sign of personal weakness”), thereby reflecting perceived depression stigma. Higher

sum scores on each subscale (range 9–45) and as a total score (range 18–90) indicate more stig-

matizing attitudes. The DSS has shown high test-retest reliability as well as moderate to high

internal consistency across various countries and in different populations (Cronbach´s alpha

ranging from .70 – .82 for subscales and total scale) [25, 28, 47–49]. We used the German ver-

sion of the DSS [49], which has been translated both forward and back from the original

English DSS in accordance with the guidelines of the World Health Organization [50] by a

native German speaker and a German mental health professional. The DSS factor-structure

depends on language, sample and cultural context and was subject to previous studies [48, 51–

54]. To date, the factor-structure of the German version of the DSS has not been investigated.

Statistical analysis

Statistical analyses were performed using IBM SPSS Statistics version 25.0. A two-tailed α =
0.05 was applied to statistical testing. First, to examine group differences between patients with

and without a depression diagnosis, χ2 tests were used for categorical variables (gender, marital
status, occupational status and reported complaints). Bonferroni correction was applied for

post-hoc analysis in case of multiple tests. Group differences for continuous variables were

analyzed using Mann-Whitney U tests, as all continuous outcome variables (age, number of
reported complaints per patient, DSS sum score and subscale scores) were non-normally dis-

tributed, as indicated by the Shapiro-Wilks test (all p < .05). Additional exploratory analyses of covariance (ANCOVAs) examined group differences in DSS sum scores and subscale

scores, respectively, when adjusting for age, gender and marital status.

Second, in order to examine whether sociodemographic and treatment-related factors pre-

dicted DSS perceived stigma scores (dependent variable) in the subgroup of patients with

depression, a multiple linear regression analysis was applied with the following predictor vari-

ables: age, gender, severity of depression (according to self-report DSQ), number and type of

complaints reported to the GP, help-seeking from specialized care and depression treatment.

Categorical variables with more than two categories (depression severity, complaints and

depression treatment) were recoded into binary dummy variables. The variable “treatment

according to guidelines”, which combines the type of depression treatment and depression

severity, was initially included in the regression analysis but did not become significant. There-

fore, depression severity and treatment were included as separate predictors since their poten-

tial association with perceived stigma is of great practical interest. All predictor variables were

entered simultaneously. The dummy variables for “no treatment” and “reporting only physical

complaints to the GP” had to be excluded from the regression analysis due to multicollinearity.

All effect sizes were interpreted as suggested by Cohen [55], i.e. 0.2 was considered a small

effect, 0.5 a medium, and 0.8 a large effect.

Results

Sample

At the reference date, 430 out of 3,069 patients (14.0%) reported a current depressive episode

according to the ICD-10 criteria, i.e. they received a depression diagnosis based on the DSQ.

Of these patients, 261 (60.7%) reported a mild depressive episode, 114 (26.5%) reported a

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 7 / 16

https://doi.org/10.1371/journal.pone.0248069

moderate depressive episode and 55 (12.8%) reported a severe depressive episode. Differences

between patients with (n = 430) and without (n = 2,639) depression with regards to sociode-
mographic information, depression stigma and complaints reported to the GP are displayed in

Table 1. All analyses of depression stigma were conducted for a reduced sample due to missing

data: 2,357 (76.8%) participants of the total sample (3,069) answered all DSS items without

missing values. For 712 patients with missing values in the subscales, data was computed by

the patient´s subscale mean score for non-missing items if the patient answered at least 7 out

Table 1. Sociodemographic characteristics, depression stigma and complaints reported to the GP in patients with and without depression.

Patients with depression
a

(n = 430)
Patients without depression

(n = 2,639)
Test Effect size

N (%) or M (SD) N (%) or M (SD)
Age (years) 50.22 (15.81) 53.9 (17.12) U = 489,994.50 r = .08

p < 0.01 Gender

b

Male 137 (31.9%) 1,061 (40.2%) χ2 (1) = 10.89 φ = .06
Female 293 (68.1%) 1,576 (59.8%) p = .001

Marital status
b

Not single 210 (49.8%) 1,561 (60.0%) χ2 (1) = 15.73 φ = .07
Single 212 (50.2%) 1,040 (40.0%) p < .001

Occupational status
b

Occupied 254 (60.2%) 1,572 (60.0%) χ2 (1) = .03 φ = .00
Not occupied 168 (39.8%) 1,020 (39.3%) p = .858

Number of reported complaints per patient 2.02 (1.33) 1.2 (0.69) U = 362,665.50 r = .25
p < .001

Type of complaints reported to the GP 430 (14%) 2,639 (86%) χ2 (4) = 449.81 φ = .38
p < .001

Number of patients only stating physical complaints 190 (42.2%) 1560 (59.1%) χ2 (1) = 33.62 φ = .11
p < .001

Number of patients only stating mental complaints 42 (9.8%) 59 (2.2%) χ2 (1) = 65.90 φ = .15
p < .001

Number of patients stating physical and mental

complaints

124 (28.8%) 93 (3.5%) χ2 (1) = 360.57 φ = .34
p < .001

Number of patients stating no complaints 19 (4.4%) 213 (8.1%) χ2 (1) = 7.06 φ = .05
p = .008

Number of patients with other reasons
c

55 (28.8%) 714 (27.1%) χ2 (1) = 40.07 φ = .11
p < .001

Depression stigma
b Median (IQR) Median (IQR)

DSS sum score 48 (40–54) 46.3 (29–53) U = 420,759.50 r = .00
p = .047

DSS personal stigma 20 (16–24) 20.25 (16.9–25) U = 441,133.00 r = .03
p = .118

DSS perceived stigma 27 (21–32) 25.9(20–29) U = 399,821.50 r = .08
p < .001

N = Number of patients; % = percent calculated for valid cases; M = mean; SD = standard deviation; DSS = Depression Stigma Scale.
a

according to DSQ self-report.
b

reduced sample size due to missing data, valid percentage are reported.
c

follow up appointments and emergency cases

https://doi.org/10.1371/journal.pone.0248069.t001

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 8 / 16

https://doi.org/10.1371/journal.pone.0248069.t001

https://doi.org/10.1371/journal.pone.0248069

of 9 (78%) subscale items according to Dardas [56]. Thus, 325 patients with missing data met

these criteria for data imputation, resulting in a total sample of 2,682 patients with DSS scores.

Patients with depression were significantly younger, more likely to be female, more fre-

quently single (small effects, all p < .01) and reported an average of 2 (vs. 1) complaints to their GP (small effect, p < .001), when compared to patients without depression. Regarding the com- plaints reported to the GP, 190 (42.2%) patients with depression reported physical complaints

only, 42 (9.8%) reported mental complaints only and 124 (28.8%) consulted the GP due to men-

tal and physical complaints. Patients with depression significantly differed in all 3 categories

from patients without depression (small to medium effects, all p < .001, see Table 1). Further, both subsamples differed in perceived depression stigma scores and depression

stigma sum scores, with higher scores in patients with depression (small effect, p < .001 and p = .047). No differences were found for occupational status and personal depression stigma scores (p > .05). Three additional exploratory ANCOVAs with DSS sum scores and subscale
scores as outcome variables did not result in changes in the significance of results when includ-

ing age, gender and marital status as covariates.

Results of an exploratory linear regression analysis for the subgroup of patients with depres-

sion are presented in Table 2.

Male gender (p = .005) and less frequent pharmacological treatment (p = .021) predicted
higher scores of perceived depression stigma in patients with depression. The other predictors

were unrelated to perceived depression stigma (all p > .05). The overall model fit was R2 = 0.06
(adjusted R2 = 0.03).

Discussion

Almost half of the primary care patients with depression reported mostly physical complaints

to their GP. Compared to patients without depression, patients with depression reported

Table 2. Linear regression for predictors of perceived depression stigma.

DSS perceived stigma score (n = 391a)
Variable Unstan-dardized β SE Standar-dized β 95% Confidence Interval (CI) t p
Age -.07 .04 -.10 -.141, .010 -1.70 .091

Gender -3.53 1.24 -.14 -5.993, -1.062 -2.81 .005

Depression severity
b

-.13 .87 -.01 -1.844, 1.582 -.15 .880

Number of reported complaints to consult the GP 1.27 .71 .14 -.127, 2.667 1.79 .075

Number of patients only reporting mental complaints 1.97 2.09 .05 -2.130, 6.072 .95 .345

Number of patients reporting physical and mental complaints .62 1.92 .02 -3.148, 4.386 .32 .75

Number of patients with different reasons
c

2.24 1.95 .06 -1.585, 6.073 1.15 .25

Number of patients reporting no complaints 3.23 3.14 .06 -2.936, 9.393 1.03 .304

Only pharmacological treatment -4.97 2.15 -.14 -9.189, -.747 -2.31 .021

Only psychotherapeutic treatment -3.65 2.02 -.11 -7.617, .314 -1.81 .071

Combination treatment -3.41 1.86 -.12 -7.086, .247 -1.83 .068

Any treatment other than pharmacological or psychotherapeutic -1.64 1.66 -.06 -4.903, 1.625 -.99 .324

Referral to and/or help-seeking from specialised care -.06 1.30 -.01 -2.613, -2.503 -.04 .966

R2 (R2 adjusted) .06 (.03)
F 1.94
p = .025

a
reduced sample size due to missing data.

b
depression severity dichotomized into “mild” vs. “moderate or severe”.

c
patients in emergency cases, coming for referral, prescription, or follow-up appointment only.

https://doi.org/10.1371/journal.pone.0248069.t002

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 9 / 16

https://doi.org/10.1371/journal.pone.0248069.t002

https://doi.org/10.1371/journal.pone.0248069

significantly more complaints, as well as a more frequent combination of physical and mental

complaints. Patients with depression did not differ from patients without depression regarding

personal depression stigma but showed higher perceived depression stigma and therefore

higher stigma scores in total.

Patients with depression presented significantly less exclusively physical complaints to their

GP than patients without depression. Nevertheless, 40% of patients with depression did not

report any mental health symptoms to their GP. Only 1 out of 10 patients with depression indi-

cated that they consulted their GP due to (only) mental complaints (according to hypotheses

1). In comparison, one third of patients with depression reported a combination of mental and

physical complaints. The large proportion of patients with depression who reported only phys-

ical complaints is aligned with findings of previous studies, including a study of German pri-

mary care patients with depression, wherein 57% of patients reported only somatic symptoms

[13], as well as an international study in which a range of 45–95% of primary care patients

with depression exclusively reported somatic symptoms to their GP [14]. Further, the combi-

nation of different complaints, i.e. mental and physical symptoms, appears to be an important

indicator for depression, as patients without depression reported this combination of symp-

toms significantly less frequently than patients with depression. In addition, depressive

patients reported twice as many complaints in total during their consultation with the GP, in

comparison to patients without depression. Our results thereby provide further evidence for

this phenomenon which has previously been reported by only a handful of studies with com-

parable ratios [14, 41].

The relevance of patients with depression reporting mainly physical symptoms, as well as

their nondisclosure of mental symptoms has been widely discussed in the literature. Contrib-

uting factors to this phenomenon may include: attributing depressive symptoms to somatic

causes [15, 57], believing somatic symptoms to be a core component of depression, a lack of

trust in primary care providers regarding the care of depression, fear of being placed on anti-

depressants, as well as stigma surrounding depression [14, 16, 58, 59].

In our sample, personal depression stigma was comparable between patients with and with-

out depression (according to hypothesis 2). Prior studies investigating path models of help-

seeking concluded that personal stigma may affect help seeking behavior at an early stage, e.g.

recognizing and appraising symptoms, a perceived need for help [7, 30, 60], and the desire to

deal with the problem alone [7, 26, 30]. This implies that the current study may have predomi-

nantly investigated patients with comparably low personal depression stigma, since these

patients may have identified or recognized their symptoms as being related to a mental illness

or else perceived a need for treatment which was followed by the intention and respective

action to seek help, which in our study was the consultation with a GP.

Perceived depression stigma was significantly higher in patients with depression as com-

pared to patients without depression (according to hypothesis 3), which is similar to the results

of previous studies measuring perceived stigma with the corresponding DSS subscale [25, 41,

42] and the Stigma Scale for Receiving Psychological Help (SSRPH) [41]. This suggests that

individuals with depression may experience stigmatization from their community due to their

diagnosis or, in case of first-time help seeking, are more sensitive to such events [22, 25].

It is also conceivable that the choice of primary care setting could have been influenced by

patients´ perceived stigma, as consulting a GP does not provide the same level of branding

someone as mentally ill as is associated with the consultation of other mental health specialists

[6, 17, 21, 61]. On the other hand, accessing a GP in Germany is much easier compared to spe-

cialized care, especially in rural areas where the density of mental health professionals is com-

parably low. Thus, an appointment in primary care can usually be arranged at short notice and

without waiting time. As we did not assess patients´ reasons to choose primary care providers,

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 10 / 16

https://doi.org/10.1371/journal.pone.0248069

no conclusions on perceived stigma as a potentially influencing factor on the choice of care

provider can be drawn from this study.

Higher scores of perceived stigma were unrelated to the type of complaints reported to the

GP in our study, i.e. fearing negative consequences of a depression diagnosis by others did not

correlate with the reporting of predominantly physical symptoms or the non-disclosure of

mental symptoms in primary care settings. Simon [14] reported that in a sample of primary

care patients with depression, the majority of patients reporting only somatic complaints

(60%) did not deny depressive symptoms when asked. The authors argued that reporting phys-

ical symptoms to a GP might function as an “admission ticket” to primary care, as GPs are not

commonly seen as the appropriate person to talk to about depression symptoms [16, 58, 62,

63]. In this case, perceived stigma plays a minor role with regards to complaints presented to

the GP.

Examining the treatment for depression within our sample, patients with depression and

higher scores of perceived stigma were less frequently treated with psychopharmacological

medication. Our findings match those of previous studies, which reported lower initiation of

pharmacological treatment and lower antidepressant medication adherence for individuals

with depression who reported higher perceived stigma [21, 38, 40], as assessed with the Per-

ceived Devaluation Discrimination Scale [36]. Since the assessment of perceived stigma was

different and given the low percentage of variance explained by the regression model in our

study, conclusions should be drawn with caution if and to what extent the fear of belonging to

a stigmatized group when receiving pharmacological treatment for depression conflicted with

the need for treatment [38]. Nevertheless, this finding is particularly important for patients

with severe depression, as antidepressant medication is typically recommended for this group,

either independently or in combination with psychotherapy.

Findings in the literature have indicated either no consistent gender differences in per-

ceived stigma (e.g. [25, 39], higher perceived stigma in females [25, 42, 47, 56], or higher per-

ceived stigma in males [64]. However, former studies on gender differences have been

heterogenic regarding samples, cultural context and measures used to investigate stigma [64].

As a result, the conclusion of our study that male primary care patients with depression are

more likely to be influenced by public attitudes than female patients requires further research.

Nevertheless, perceived depression stigma may have an effect on medication adherence [40].

Therefore, GPs should pay special attention to male patients so as to address and manage the

anticipated negative effects of treatment.

Perceived stigma is only one type of mental health stigma and help-seeking is a complex

process that is influenced by a variety of different factors, including patient and illness charac-

teristics [25]. Perceived stigma does not appear to prevent primary care patients from consult-

ing a GP, even if we cannot yet draw conclusions on whether a patient associates his/her

complaints with a mental health disorder or with depression. The question of whether per-

ceived stigma influences the types of complaints reported, as well as how many complaints are

reported by patients in primary care settings requires further research. In later steps of the

help-seeking process, perceived stigma may hamper the treatment of depression, in particular

pharmacological treatment and treatment for male patients. This should therefore be

addressed in future studies.

Conclusion

This study provides evidence for the importance of physical complaints reported by patients

with depression in primary care settings, in light of the small number of patients disclosing

mental symptoms. The study further emphasizes the importance of the number of complaints

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 11 / 16

https://doi.org/10.1371/journal.pone.0248069

reported by patients as a potential marker for a depression. GPs should take note of patients

reporting multiple complaints, particularly if this includes a combination of physical and men-

tal complaints, and consider screening these patients for an underlying depression. Further to

this, we suggest GPs should also address patients´ concerns about public attitudes towards

depression and provide support to overcome them [11], particularly as pharmacological treat-

ment may be influenced by the anticipation of negative consequences [38]. Broad awareness

campaigns should include a focus on primary care settings as appropriate sources for depres-

sion care and individuals should be encouraged to disclose mental symptoms within this set-

ting [16].

Future stigma research should focus on the different types of stigma, as well as the various

stages of the help-seeking process. These studies should also take covariates into account (e.g.

awareness of services, believing in treatment efficacy [31], and believing in a continuum of

symptoms from health to illness [65]), so as to tailor individual interventions according to the

respective stage of an individual’s help-seeking process.

Strengths and limitations

Our study provides further evidence for the importance of the number of complaints primary

care patients with depression report to their GPs. These findings are of great practical rele-

vance, as patients with and without depression differ significantly in how many complaints

they report. To date, there have been only two studies making mention of this phenomenon.

Further, to our knowledge there has been no prior research regarding the association between

the number and the type of complaints reported (physical vs. mental) and their association

with perceived depression stigma. However, this study may also have had a number of limita-

tions. Due to the cross-sectional design of the study, no causal inferences can be drawn. Fur-

ther, as participation in the study was voluntary, a selection bias may have occurred and as a

result, patients with lower stigma may have been overrepresented in our sample. Social desir-

ability bias when answering the DSS might also have been an issue, resulting in an underesti-

mation of depression stigma within our sample. Moreover, the factor structure of the German

version of the DSS has not been replicated and its psychometric properties require further

research. Depression diagnosis was based on self-report assessments with the DSQ, and a

potential bias may have occurred. We did not include GP diagnoses in our analyses, as this

could have significantly reduced our sample size. Further, from a clinical point of view, the

patients’ subjective condition was our main area of interest. Since we did not assess the specific

type of complaints for patients who self-referred to the GP for immediate care (i.e. emergency

case) or for follow-up appointment, a potential bias cannot be ruled out. Finally, when predict-

ing perceived stigma we were not able to control for all potential influencing factors (e.g. men-

tal health literacy) in a systematic way.

Supporting information

S1 Dataset. Study data to reproduce the results.

(XLSX)

Acknowledgments

We thank Melissa-Claire Daugelat of the Technische Universität Dresden for language editing.

We acknowledge support from Leipzig University for Open Access Publishing.

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 12 / 16

http://www.plosone.org/article/fetchSingleRepresentation.action?uri=info:doi/10.1371/journal.pone.0248069.s001

https://doi.org/10.1371/journal.pone.0248069

Author Contributions

Conceptualization: Ines Heinz, Katja Beesdo-Baum, Susanne Knappe, Christine Rummel-

Kluge.

Data curation: Ines Heinz.

Formal analysis: Ines Heinz, Sabrina Baldofski.

Funding acquisition: Katja Beesdo-Baum.

Investigation: Ines Heinz, Katja Beesdo-Baum, Susanne Knappe, Christine Rummel-Kluge.

Methodology: Ines Heinz, Christine Rummel-Kluge.

Project administration: Katja Beesdo-Baum, Susanne Knappe.

Supervision: Elisabeth Kohls, Christine Rummel-Kluge.

Validation: Sabrina Baldofski, Elisabeth Kohls, Christine Rummel-Kluge.

Visualization: Ines Heinz.

Writing – original draft: Ines Heinz.

Writing – review & editing: Ines Heinz, Sabrina Baldofski, Katja Beesdo-Baum, Susanne

Knappe, Elisabeth Kohls, Christine Rummel-Kluge.

References
1. Vos T, Abajobir AA, Abate KH, Abbafati C, Abbas KM, Abd-Allah F, et al. Global, regional, and national

incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries,

1990–2016: a systematic analysis for the Global Burden of Disease Study 2016. The Lancet. 2017;

390:1211–59. https://doi.org/10.1016/S0140-6736(17)32154-2 PMID: 28919117

2. Cuijpers P, Sijbrandij M, Koole SL, Andersson G, Beekman AT, Reynolds CF. The efficacy of psycho-

therapy and pharmacotherapy in treating depressive and anxiety disorders: a meta-analysis of direct

comparisons. World Psychiatry. 2013; 12:137–48. https://doi.org/10.1002/wps.20038 PMID: 23737423

3. Boenisch S, Kocalevent R-D, Matschinger H, Mergl R, Wimmer-Brunauer C, Tauscher M, et al. Who

receives depression-specific treatment? A secondary data-based analysis of outpatient care received

by over 780,000 statutory health-insured individuals diagnosed with depression. Soc Psychiatry Psy-

chiatr Epidemiol. 2012; 47:475–86. https://doi.org/10.1007/s00127-011-0355-y PMID: 21350809

4. Trautmann S, Beesdo-Baum K. The Treatment of Depression in Primary Care. Dtsch Arztebl Int. 2017;

114:721–8. https://doi.org/10.3238/arztebl.2017.0721 PMID: 29143731

5. Wang PS, Angermeyer MC, Borges G, Bruffaerts R, Tat Chiu W, Girolamo G de, et al. Delay and failure

in treatment seeking after first onset of mental disorders in the World Health Organization’s World Men-

tal Health Survey Initiative. World Psychiatry. 2007; 6:177–85. PMID: 18188443

6. Corrigan P. How stigma interferes with mental health care. Am Psychol. 2004; 59:614–25. https://doi.

org/10.1037/0003-066X.59.7.614 PMID: 15491256

7. Schnyder N, Panczak R, Groth N, Schultze-Lutter F. Association between mental health-related stigma

and active help-seeking: systematic review and meta-analysis. Br J Psychiatry. 2017; 210:261–8.

https://doi.org/10.1192/bjp.bp.116.189464 PMID: 28153928

8. Schneider F, Kratz S, Bermejo I, Menke R, Mulert C, Hegerl U, et al. Insufficient depression treatment in

outpatient settings. Ger Med Sci. 2004; 2:Doc01. PMID: 19675684

9. Becker N, Abholz H-H. [Prevalence and Detection of Depressive Disorders in German General Practice

—A Systematic Review]. Z Allg Med. 2005; 81:474–81. https://doi.org/10.1055/s-2005-872584

10. Beesdo-Baum K, Knappe S, Einsle F, Knothe L, Wieder G, Venz J, et al. [How frequently are depressive

disorders recognized in primary care patients?: A cross-sectional epidemiological study in Germany].

Bundesgesundheitsblatt Gesundheitsforschung Gesundheitsschutz. 2018; 61:52–64. https://doi.org/

10.1007/s00103-017-2662-2 PMID: 29189872

11. Vega WA, Rodriguez MA, Ang A. Addressing stigma of depression in Latino primary care patients. Gen

Hosp Psychiatry. 2010; 32:182–91. https://doi.org/10.1016/j.genhosppsych.2009.10.008 PMID:

20302993

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 13 / 16

https://doi.org/10.1016/S0140-6736%2817%2932154-2

http://www.ncbi.nlm.nih.gov/pubmed/28919117

https://doi.org/10.1002/wps.20038

http://www.ncbi.nlm.nih.gov/pubmed/23737423

https://doi.org/10.1007/s00127-011-0355-y

http://www.ncbi.nlm.nih.gov/pubmed/21350809

https://doi.org/10.3238/arztebl.2017.0721

http://www.ncbi.nlm.nih.gov/pubmed/29143731

http://www.ncbi.nlm.nih.gov/pubmed/18188443

https://doi.org/10.1037/0003-066X.59.7.614

https://doi.org/10.1037/0003-066X.59.7.614

http://www.ncbi.nlm.nih.gov/pubmed/15491256

https://doi.org/10.1192/bjp.bp.116.189464

http://www.ncbi.nlm.nih.gov/pubmed/28153928

http://www.ncbi.nlm.nih.gov/pubmed/19675684

https://doi.org/10.1055/s-2005-872584

https://doi.org/10.1007/s00103-017-2662-2

https://doi.org/10.1007/s00103-017-2662-2

http://www.ncbi.nlm.nih.gov/pubmed/29189872

https://doi.org/10.1016/j.genhosppsych.2009.10.008

http://www.ncbi.nlm.nih.gov/pubmed/20302993

https://doi.org/10.1371/journal.pone.0248069

12. Cadoret RJ, Widmer RB, Troughton EP. Somatic complaints. J Affect Disord. 1980; 2:61–70. https://

doi.org/10.1016/0165-0327(80)90022-1 PMID: 6448881

13. Kratz S, Härter M, Bermejo I, Berger M, Schneider F, Gaebel W. [Reason for encounter and diagnosis

of depression in patients in general practice]. Z Arztl Fortbild Qualitatssich. 2003; 97 Suppl 4:50–6.

PMID: 14710668

14. Simon GE, VonKorff M, Piccinelli M, Fullerton C, Ormel J. An international study of the relation between

somatic symptoms and depression. N Engl J Med. 1999; 341:1329–35. https://doi.org/10.1056/

NEJM199910283411801 PMID: 10536124

15. Gates K, Petterson S, Wingrove P, Miller B, Klink K. You can’t treat what you don’t diagnose: An analy-

sis of the recognition of somatic presentations of depression and anxiety in primary care. Fam Syst

Health. 2016; 34:317–29. https://doi.org/10.1037/fsh0000229 PMID: 27598458

16. Bell RA, Franks P, Duberstein PR, Epstein RM, Feldman MD, Fernandez y Garcia E, et al. Suffering in

silence: reasons for not disclosing depression in primary care. Ann Fam Med. 2011; 9:439–46. https://

doi.org/10.1370/afm.1277 PMID: 21911763

17. Clement S, Schauman O, Graham T, Maggioni F, Evans-Lacko S, Bezborodovs N, et al. What is the

impact of mental health-related stigma on help-seeking? A systematic review of quantitative and quali-

tative studies. Psychol Med. 2015; 45:11–27. https://doi.org/10.1017/S0033291714000129 PMID:

24569086

18. Goffman E. STIGMA: NOTES ON THE MANAGEMENT OF SPOILED IDENTITY. By Erving Goffman.

Englewood Cliffs, New Jersey: Prentice-Hall, 1963. 147 pp. Social Forces. 1964; 43:127–8. https://doi.

org/10.1093/sf/43.1.127

19. Link BG, Yang LH, Phelan JC, Collins PY. Measuring mental illness stigma. Schizophr Bull. 2004;

30:511–41. https://doi.org/10.1093/oxfordjournals.schbul.a007098 PMID: 15631243

20. Rüsch N, Angermeyer MC, Corrigan PW. Mental illness stigma: concepts, consequences, and initia-

tives to reduce stigma. Eur Psychiatry. 2005; 20:529–39. https://doi.org/10.1016/j.eurpsy.2005.04.004

PMID: 16171984

21. Campbell DG, Bonner LM, Bolkan CR, Lanto AB, Zivin K, Waltz TJ, et al. Stigma Predicts Treatment

Preferences and Care Engagement Among Veterans Affairs Primary Care Patients with Depression.

Ann Behav Med. 2016; 50:533–44. https://doi.org/10.1007/s12160-016-9780-1 PMID: 26935310

22. Schomerus G, Matschinger H, Angermeyer MC. The stigma of psychiatric treatment and help-seeking

intentions for depression. Eur Arch Psychiatry Clin Neurosci. 2009; 259:298–306. https://doi.org/10.

1007/s00406-009-0870-y PMID: 19224105

23. Latalova K, Kamaradova D, Prasko J. Perspectives on perceived stigma and self-stigma in adult male

patients with depression. Neuropsychiatr Dis Treat. 2014; 10:1399–405. https://doi.org/10.2147/NDT.

S54081 PMID: 25114531

24. Pinel EC, Bosson JK. Turning Our Attention to Stigma: An Objective Self-Awareness Analysis of Stigma

and Its Consequences. Basic and Applied Social Psychology. 2013; 35:55–63. https://doi.org/10.1080/

01973533.2012.746593

25. Griffiths KM, Christensen H, Jorm AF. Predictors of depression stigma. BMC Psychiatry. 2008; 8:25.

https://doi.org/10.1186/1471-244X-8-25 PMID: 18423003

26. Griffiths KM, Crisp DA, Jorm AF, Christensen H. Does stigma predict a belief in dealing with depression

alone? Journal of Affective Disorders. 2011; 132:413–7. https://doi.org/10.1016/j.jad.2011.03.012

PMID: 21440305

27. Griffiths KM, Christensen H, Jorm AF, Evans K, Groves C. Effect of web-based depression literacy and

cognitive-behavioural therapy interventions on stigmatising attitudes to depression: randomised con-

trolled trial. Br J Psychiatry. 2004; 185:342–9. https://doi.org/10.1192/bjp.185.4.342 PMID: 15458995

28. Griffiths KM, Nakane Y, Christensen H, Yoshioka K, Jorm AF, Nakane H. Stigma in response to mental

disorders: a comparison of Australia and Japan. BMC Psychiatry. 2006; 6:21. https://doi.org/10.1186/

1471-244X-6-21 PMID: 16716231

29. Conner KO, Copeland VC, Grote NK, Koeske G, Rosen D, Reynolds CF, et al. Mental health treatment

seeking among older adults with depression: the impact of stigma and race. Am J Geriatr Psychiatry.

2010; 18:531–43. https://doi.org/10.1097/JGP.0b013e3181cc0366 PMID: 20220602

30. Schomerus G, Stolzenburg S, Freitag S, Speerforck S, Janowitz D, Evans-Lacko S, et al. Stigma as a

barrier to recognizing personal mental illness and seeking help: a prospective study among untreated

persons with mental illness. Eur Arch Psychiatry Clin Neurosci. 2019; 269:469–79. https://doi.org/10.

1007/s00406-018-0896-0 PMID: 29679153

31. Tomczyk S, Schmidt S, Muehlan H, Stolzenburg S, Schomerus G. A Prospective Study on Structural

and Attitudinal Barriers to Professional Help-Seeking for Currently Untreated Mental Health Problems in

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 14 / 16

https://doi.org/10.1016/0165-0327%2880%2990022-1

https://doi.org/10.1016/0165-0327%2880%2990022-1

http://www.ncbi.nlm.nih.gov/pubmed/6448881

http://www.ncbi.nlm.nih.gov/pubmed/14710668

https://doi.org/10.1056/NEJM199910283411801

https://doi.org/10.1056/NEJM199910283411801

http://www.ncbi.nlm.nih.gov/pubmed/10536124

https://doi.org/10.1037/fsh0000229

http://www.ncbi.nlm.nih.gov/pubmed/27598458

https://doi.org/10.1370/afm.1277

https://doi.org/10.1370/afm.1277

http://www.ncbi.nlm.nih.gov/pubmed/21911763

https://doi.org/10.1017/S0033291714000129

http://www.ncbi.nlm.nih.gov/pubmed/24569086

https://doi.org/10.1093/sf/43.1.127

https://doi.org/10.1093/sf/43.1.127

https://doi.org/10.1093/oxfordjournals.schbul.a007098

http://www.ncbi.nlm.nih.gov/pubmed/15631243

https://doi.org/10.1016/j.eurpsy.2005.04.004

http://www.ncbi.nlm.nih.gov/pubmed/16171984

https://doi.org/10.1007/s12160-016-9780-1

http://www.ncbi.nlm.nih.gov/pubmed/26935310

https://doi.org/10.1007/s00406-009-0870-y

https://doi.org/10.1007/s00406-009-0870-y

http://www.ncbi.nlm.nih.gov/pubmed/19224105

https://doi.org/10.2147/NDT.S54081

https://doi.org/10.2147/NDT.S54081

http://www.ncbi.nlm.nih.gov/pubmed/25114531

https://doi.org/10.1080/01973533.2012.746593

https://doi.org/10.1080/01973533.2012.746593

https://doi.org/10.1186/1471-244X-8-25

http://www.ncbi.nlm.nih.gov/pubmed/18423003

https://doi.org/10.1016/j.jad.2011.03.012

http://www.ncbi.nlm.nih.gov/pubmed/21440305

https://doi.org/10.1192/bjp.185.4.342

http://www.ncbi.nlm.nih.gov/pubmed/15458995

https://doi.org/10.1186/1471-244X-6-21

https://doi.org/10.1186/1471-244X-6-21

http://www.ncbi.nlm.nih.gov/pubmed/16716231

https://doi.org/10.1097/JGP.0b013e3181cc0366

http://www.ncbi.nlm.nih.gov/pubmed/20220602

https://doi.org/10.1007/s00406-018-0896-0

https://doi.org/10.1007/s00406-018-0896-0

http://www.ncbi.nlm.nih.gov/pubmed/29679153

https://doi.org/10.1371/journal.pone.0248069

the Community. J Behav Health Serv Res. 2020; 47:54–69. https://doi.org/10.1007/s11414-019-09662-

8 PMID: 31165415

32. Vogel DL, Wade NG, Haake S. Measuring the self-stigma associated with seeking psychological help.

J. Couns. Psychol. 2006; 53:325–37. https://doi.org/10.1037/0022-0167.53.3.325

33. Arnaez JM, Krendl AC, McCormick BP, Chen Z, Chomistek AK. The association of depression stigma

with barriers to seeking mental health care: a cross-sectional analysis. J Ment Health. 2020; 29:182–90.

https://doi.org/10.1080/09638237.2019.1644494 PMID: 31373519

34. Corrigan PW, Watson AC, Barr L. The Self–Stigma of Mental Illness: Implications for Self–Esteem and

Self–Efficacy. J Soc Clin Psychol. 2006; 25:875–84. https://doi.org/10.1521/jscp.2006.25.8.875

35. Link BG, Cullen FT, Frank J, Wozniak JF. The Social Rejection of Former Mental Patients: Understand-

ing Why Labels Matter. Am J Sociol. 1987; 92:1461–500. https://doi.org/10.1086/228672

36. Link BG. Understanding Labeling Effects in the Area of Mental Disorders: An Assessment of the Effects

of Expectations of Rejection. Am Sociol Rev. 1987; 52:96. https://doi.org/10.2307/2095395

37. Ritsher JB, Otilingam PG, Grajales M. Internalized stigma of mental illness: psychometric properties of

a new measure. Psychiatry Res. 2003; 121:31–49. https://doi.org/10.1016/j.psychres.2003.08.008

PMID: 14572622

38. Sirey JA, Bruce ML, Alexopoulos GS, Perlick DA, Friedman SJ, Meyers BS. Stigma as a barrier to

recovery: Perceived stigma and patient-rated severity of illness as predictors of antidepressant drug

adherence. Psychiatr Serv. 2001; 52:1615–20. https://doi.org/10.1176/appi.ps.52.12.1615 PMID:

11726752

39. Gearing RE, MacKenzie MJ, Ibrahim RW, Brewer KB, Batayneh JS, Schwalbe CSJ. Stigma and mental

health treatment of adolescents with depression in jordan. Community Ment Health J. 2015; 51:111–7.

https://doi.org/10.1007/s10597-014-9756-1 PMID: 25027014

40. Sirey JA, Bruce ML, Alexopoulos GS, Perlick DA, Raue P, Friedman SJ, et al. Perceived stigma as a

predictor of treatment discontinuation in young and older outpatients with depression. Am J Psychiatry.

2001; 158:479–81. https://doi.org/10.1176/appi.ajp.158.3.479 PMID: 11229992

41. Pyne JM, Kuc EJ, Schroeder PJ, Fortney JC, Edlund M, Sullivan G. Relationship between perceived

stigma and depression severity. J Nerv Ment Dis. 2004; 192:278–83. https://doi.org/10.1097/01.nmd.

0000120886.39886.a3 PMID: 15060401

42. Busby Grant J, Bruce CP, Batterham PJ. Predictors of personal, perceived and self-stigma towards

anxiety and depression. Epidemiol Psychiatr Sci. 2016; 25:247–54. https://doi.org/10.1017/

S2045796015000220 PMID: 25791089

43. Wittchen H-U, Perkonigg Axel. DIA-X-Screeningverfahren: Fragebogen DIA-X- ASQ, Screening für

Angststörungen; DIA-X-DSQ, Screening für Depressionen. Frankfurt: Swets & Zeitlinger; 1997.

44. Wittchen H-U. Depression 2000. Eine bundesweite Depressions-Screening-Studie in Allgemeinarzt-

praxen. MMW Fortschr Med. 2000:1–41.

45. Höfler M, Wittchen H-U. Why do primary care doctors diagnose depression when diagnostic criteria are

not met? Int. J. Method. Psychiat. Res. 2000; 9:110–20. https://doi.org/10.1002/mpr.85

46. DGPPN, BÄK, KBV, AWMF, AkdÄ, BPtK, et al. S3-Guideline/National Disease Management Guideline.

Unipolar Depression. Unipolar Depression. Berlin, Düsseldorf; 2015.

47. Calear AL, Griffiths KM, Christensen H. Personal and perceived depression stigma in Australian adoles-

cents: magnitude and predictors. J Affect Disord. 2011; 129:104–8. https://doi.org/10.1016/j.jad.2010.

08.019 PMID: 20863571

48. Boerema AM, van Zoonen K, Cuijpers P, Holtmaat CJM, Mokkink LB, Griffiths KM, et al. Psychometric

Properties of the Dutch Depression Stigma Scale (DSS) and Associations with Personal and Perceived

Stigma in a Depressed and Community Sample. PLoS ONE. 2016; 11:e0160740. https://doi.org/10.

1371/journal.pone.0160740 PMID: 27500969

49. Dietrich S, Mergl R, Rummel-Kluge C. Personal and perceived stigmatization of depression: a compari-

son of data from the general population, participants of a depression congress and job placement offi-

cers in Germany. Psychiatry Res. 2014; 220:598–603. https://doi.org/10.1016/j.psychres.2014.06.044

PMID: 25086761

50. World Health Organization. Process of translation and adaptation of instruments. Management of Sub-

stance Abuse. 2014. https://www.who.int/substance_abuse/research_tools/translation/en/. Accessed

13 Nov 2020.

51. Amarasuriya SD, Jorm AF, Reavley NJ, Mackinnon AJ. Stigmatising attitudes of undergraduates

towards their peers with depression: a cross-sectional study in Sri Lanka. BMC Psychiatry. 2015;

15:129. https://doi.org/10.1186/s12888-015-0523-9 PMID: 26087847

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 15 / 16

https://doi.org/10.1007/s11414-019-09662-8

https://doi.org/10.1007/s11414-019-09662-8

http://www.ncbi.nlm.nih.gov/pubmed/31165415

https://doi.org/10.1037/0022-0167.53.3.325

https://doi.org/10.1080/09638237.2019.1644494

http://www.ncbi.nlm.nih.gov/pubmed/31373519

https://doi.org/10.1521/jscp.2006.25.8.875

https://doi.org/10.1086/228672

https://doi.org/10.2307/2095395

https://doi.org/10.1016/j.psychres.2003.08.008

http://www.ncbi.nlm.nih.gov/pubmed/14572622

https://doi.org/10.1176/appi.ps.52.12.1615

http://www.ncbi.nlm.nih.gov/pubmed/11726752

https://doi.org/10.1007/s10597-014-9756-1

http://www.ncbi.nlm.nih.gov/pubmed/25027014

https://doi.org/10.1176/appi.ajp.158.3.479

http://www.ncbi.nlm.nih.gov/pubmed/11229992

https://doi.org/10.1097/01.nmd.0000120886.39886.a3

https://doi.org/10.1097/01.nmd.0000120886.39886.a3

http://www.ncbi.nlm.nih.gov/pubmed/15060401

https://doi.org/10.1017/S2045796015000220

https://doi.org/10.1017/S2045796015000220

http://www.ncbi.nlm.nih.gov/pubmed/25791089

https://doi.org/10.1002/mpr.85

https://doi.org/10.1016/j.jad.2010.08.019

https://doi.org/10.1016/j.jad.2010.08.019

http://www.ncbi.nlm.nih.gov/pubmed/20863571

https://doi.org/10.1371/journal.pone.0160740

https://doi.org/10.1371/journal.pone.0160740

http://www.ncbi.nlm.nih.gov/pubmed/27500969

https://doi.org/10.1016/j.psychres.2014.06.044

http://www.ncbi.nlm.nih.gov/pubmed/25086761

https://www.who.int/substance_abuse/research_tools/translation/en/

https://doi.org/10.1186/s12888-015-0523-9

http://www.ncbi.nlm.nih.gov/pubmed/26087847

https://doi.org/10.1371/journal.pone.0248069

52. Jorm AF, Wright A. Influences on young people’s stigmatising attitudes towards peers with mental disor-

ders: national survey of young Australians and their parents. Br J Psychiatry. 2008; 192:144–9. https://

doi.org/10.1192/bjp.bp.107.039404 PMID: 18245033

53. Reavley NJ, Mackinnon AJ, Morgan AJ, Jorm AF. Stigmatising attitudes towards people with mental

disorders: a comparison of Australian health professionals with the general community. Aust N Z J Psy-

chiatry. 2014; 48:433–41. https://doi.org/10.1177/0004867413500351 PMID: 23943633

54. Yap MBH, Mackinnon A, Reavley N, Jorm AF. The measurement properties of stigmatizing attitudes

towards mental disorders: results from two community surveys. Int. J. Method. Psychiat. Res. 2014;

23:49–61. https://doi.org/10.1002/mpr.1433 PMID: 24550065

55. Cohen J. Statistical power analysis for the behavioral sciences. 2
nd

ed. Hillsdale, N.J.: L. Erlbaum

Associates; 1988.

56. Dardas LA, Silva SG, Smoski MJ, Noonan D, Simmons LA. Personal and Perceived Depression Stigma

among Arab Adolescents: Associations with Depression Severity and Personal Characteristics. Arch

Psychiatr Nurs. 2017; 31:499–506. https://doi.org/10.1016/j.apnu.2017.06.005 PMID: 28927515

57. Wittchen H-U, Kessler RC, Beesdo K, Krause P, Höfler M, Hoyer J. Generalized anxiety and depression

in primary care: prevalence, recognition, and management. J Clin Psychiatry. 2002; 63 Suppl 8:24–34.

58. Keller AO, Valdez CR, Schwei RJ, Jacobs EA. Disclosure of Depression in Primary Care: A Qualitative

Study of Women’s Perceptions. Womens Health Issues. 2016; 26:529–36. https://doi.org/10.1016/j.

whi.2016.07.002 PMID: 27531601

59. Kluge M. Dittmann R., Lehmann M., Lindén M., Wehmeier P. M., Lilly Deutschland. Muscular Com-

plaints and Headache are Common Painful Physical Symptoms in Patients With Depression. German

Journal of Psychiatry.2006; 9(1):101–106.

60. Horsfield P, Stolzenburg S, Hahm S, Tomczyk S, Muehlan H, Schmidt S, et al. Self-labeling as having a

mental or physical illness: the effects of stigma and implications for help-seeking. Soc Psychiatry Psy-

chiatr Epidemiol. 2020; 55:907–16. https://doi.org/10.1007/s00127-019-01787-7 PMID: 31641830

61. Barney LJ, Griffiths KM, Jorm AF, Christensen H. Stigma about depression and its impact on help-seek-

ing intentions. Aust N Z J Psychiatry. 2006; 40:51–4. https://doi.org/10.1080/j.1440-1614.2006.01741.x

PMID: 16403038

62. Wrigley S, Jackson H, Judd F, Komiti A. Role of stigma and attitudes toward help-seeking from a gen-

eral practitioner for mental health problems in a rural town. Aust N Z J Psychiatry. 2005; 39:514–21.

https://doi.org/10.1080/j.1440-1614.2005.01612.x PMID: 15943655

63. Nair P, Bhanu C, Frost R, Buszewicz M, Walters KR. A Systematic Review of Older Adults’ Attitudes

Towards Depression and Its Treatment. Gerontologist. 2020; 60: e93–e104. https://doi.org/10.1093/

geront/gnz048 PMID: 31115449

64. Peluso EdTP Blay SL. Public stigma in relation to individuals with depression. Journal of Affective Disor-

ders. 2009; 115:201–6. https://doi.org/10.1016/j.jad.2008.08.013 PMID: 18842306

65. Angermeyer MC, Millier A, Rémuzat C, Refaï T, Schomerus G, Toumi M. Continuum beliefs and atti-
tudes towards people with mental illness: Results from a national survey in France. Int J Soc Psychiatry.

2015; 61:297–303. https://doi.org/10.1177/0020764014543312 PMID: 25061023

PLOS ONE Reasons for consultation and perceived depression stigma in primary care patients

PLOS ONE | https://doi.org/10.1371/journal.pone.0248069 March 5, 2021 16 / 16

https://doi.org/10.1192/bjp.bp.107.039404

https://doi.org/10.1192/bjp.bp.107.039404

http://www.ncbi.nlm.nih.gov/pubmed/18245033

https://doi.org/10.1177/0004867413500351

http://www.ncbi.nlm.nih.gov/pubmed/23943633

https://doi.org/10.1002/mpr.1433

http://www.ncbi.nlm.nih.gov/pubmed/24550065

https://doi.org/10.1016/j.apnu.2017.06.005

http://www.ncbi.nlm.nih.gov/pubmed/28927515

https://doi.org/10.1016/j.whi.2016.07.002

https://doi.org/10.1016/j.whi.2016.07.002

http://www.ncbi.nlm.nih.gov/pubmed/27531601

https://doi.org/10.1007/s00127-019-01787-7

http://www.ncbi.nlm.nih.gov/pubmed/31641830

https://doi.org/10.1080/j.1440-1614.2006.01741.x

http://www.ncbi.nlm.nih.gov/pubmed/16403038

https://doi.org/10.1080/j.1440-1614.2005.01612.x

http://www.ncbi.nlm.nih.gov/pubmed/15943655

https://doi.org/10.1093/geront/gnz048

https://doi.org/10.1093/geront/gnz048

http://www.ncbi.nlm.nih.gov/pubmed/31115449

https://doi.org/10.1016/j.jad.2008.08.013

http://www.ncbi.nlm.nih.gov/pubmed/18842306

https://doi.org/10.1177/0020764014543312

http://www.ncbi.nlm.nih.gov/pubmed/25061023

https://doi.org/10.1371/journal.pone.0248069

Copyright of PLoS ONE is the property of Public Library of Science and its content may not
be copied or emailed to multiple sites or posted to a listserv without the copyright holder’s
express written permission. However, users may print, download, or email articles for
individual use.

REVIEW ARTICLE

The relationship between potentially traumatic or stressful events, HIV

infection and neurocognitive impairment (NCI): a systematic review of
observational epidemiological studies
G. Spies a, S. Mallb, H. Wielera, L. Masilelab, E. Castelon Konkiewitzc and S. Seedat a

aDST/NRF South African Research Chair in PTSD, Department of Psychiatry, Faculty of Medicine and Health Sciences, Stellenbosch
University, Stellenbosch, South Africa; bDivision of Epidemiology and Biostatistics, School of Public Health, Faculty of Health Sciences,
University of the Witwatersrand, Johannesburg, South Africa; cFaculdade de Ciências Médicas e da Saúde, Universidade Federal da
Grande Dourados, Dourados, Brasil

ABSTRACT
Background: HIV/AIDS and potentially traumatic events (PTEs) or stressful life events (SLEs)
and/or PTSD are independently associated with neurocognitive impairment (NCI). Literatur

e

suggests that HIV and PTE/SLE exposure independently and consistently affect various
domains of cognition including language ability, working memory and psychomotor
speed. There are limited data on the interaction between HIV infection and PTEs and their
combined effect on NCI

.

Objective: In this systematic review, we synthesise evidence for the combined effect of HIV
infection and PTEs and SLEs and/or post-traumatic stress disorder (PTSD) on NCI of people
living with HIV/AIDS (PLWHA) from high-, middle- and low- income countries.
Method: Our inclusion criteria were observational epidemiological studies (case-control,
cohort and cross-sectional designs) that investigated the interaction of HIV infection, PTEs
and SLEs and/or PTSD and specifically their combined effect on NCI in adults. We searched
a number of electronic databases including Pubmed/Medline, PsycINFO, Scopus and Global
Health using the search terms: cognition, HIV/AIDS, observational studies, trauma and
permutations thereof.
Results: Fifteen studies were included in the review, of which the majority were conducted in
high-income countries. Ten of the fifteen studies were conducted in the United States of
America (USA) and five in South Africa. Seven of these focused on early life stress/childhood
trauma. The remaining studies assessed adult-onset PTEs and SLEs only. Eight studies included
women only. Overall, the studies suggest that PTE and SLE exposure and/or PTSD are
a significant risk factor for NCI in adults living with HIV, with impairments in memory and
executive functions being the most likely consequence of PTE and SLE exposure.
Conclusion: These findings highlight the need for trauma screening and for the integration
of trauma-focused interventions in HIV care to improve outcomes.

La relación entre eventos potencialmente traumáticos o estresantes,
infección por VIH y deterioro neurocognitivo (NCI): una revisión
sistemática de estudios epidemiológicos observacionales
Antecedentes: El VIH/SIDA y los eventos potencialmente traumáticos (PTEs) o los eventos
estresantes de la vida (SLEs) y/o TEPT se asocian independientemente con el deterioro
neurocognitivo (NCI). La literatura sugiere que la exposición al VIH, PTE y SLE afecta de
manera independiente y consistente varios dominios de la cognición, incluida la capacidad
del lenguaje, la memoria de trabajo y la velocidad psicomotora. Hay datos limitados sobre la
interacción entre la infección por VIH y los PTE, y su efecto combinado sobre el NCI.
Objetivo: En esta revisión sistemática sintetizamos evidencia del efecto combinado de la
infección por VIH, PTEs y SLEs, y/o TEPT en el NCI de personas que viven con VIH/SIDA
(PLWHA) en países de ingresos altos, medios y bajos.
Método: Nuestros criterios de inclusión fueron estudios epidemiológicos observacionales
(diseño de caso-control, cohortes y diseños transversales) que investigaron la interacción de
la infección por VIH, PTEs y SLEs y/o TEPT, y específicamente su efecto combinado sobre el
NCI en adultos. Se realizaron búsquedas en varias bases de datos electrónicas, que
incluyeron a Pubmed/Medline, PsycINFO, Scopus y Global Health, utilizando los términos
de búsqueda: cognición, VIH/SIDA, estudios de observación, trauma y permutaciones de los
mismos.
Resultados: Quince estudios se incluyeron en la revisión, de los cuales la mayoría se
realizaron en países de altos ingresos. Diez de los quince estudios fueron realizados en los
Estados Unidos de América (EE.UU.) y cinco en Sudáfrica. Siete de éstos se centraron en el
estrés de la vida temprana/trauma infantil. Los estudios restantes evaluaron PTEs y SLEs cuya

ARTICLE HISTORY
Received 28 February 2020
Revised 1 June 2020
Accepted 4 June 202

0

KEYWORDS
HIV; PLWHA; neurocognitive
impairment (NCI); cognitive
impairment; traumatic
events; stress; PTSD

PALABRAS CLAVE
VIH; PLWHA; Deterioro
neurocognitivo (NCI);
Deterioro cognitivo; Eventos
traumáticos; Estrés; TEPT

关键词
HIV; PLWHA; 神经认知障碍
(NCI); 认知障碍; 创伤事件;
应激; PTSD

HIGHLIGHTS
• HIV and the experience of
a traumatic event can have a
negative impact on brain
functioning

.

• This synthesis of evidence
shows that people living
with HIV who have endured
a trauma at any point in
their lives are more likely to
have trouble remembering,
paying attention and/or
multitasking.

CONTACT G. Spies ggiocos@sun.ac.za DST/NRF South African Research Chair in PTSD, Department of Psychiatry, Faculty of Medicine and
Health Sciences, Stellenbosch University, 7505, Stellenbosch, South Africa

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY
2020, VOL. 11, 1781432
https://doi.org/10.1080/20008198.2020.1781432

© 2020 The Author(s). Published by Informa UK Limited, trading as Taylor & Francis Group.
This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0/), which permits
unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.

http://orcid.org/0000-0003-0853-2813

http://orcid.org/0000-0002-5118-786X

http://www.tandfonline.com

https://crossmark.crossref.org/dialog/?doi=10.1080/20008198.2020.1781432&domain=pdf&date_stamp=2020-08-13

aparición fue en la vida adulta solamente. Ocho estudios incluyeron sólo mujeres. En
general, los estudios sugieren que la exposición a PTE y SLE y/o TEPT es un factor de riesgo
significativo para NCI en adultos que viven con VIH, con el deterioro en la memoria y las
funciones ejecutivas como la consecuencia más probable de la exposición a PTE y SLE.
Conclusión: Estos hallazgos resaltan la necesidad de la detección de traumas y la
integración de intervenciones centradas en el trauma en la atención del VIH para mejorar
sus resultados.

潜在创伤或应激事件, HIV感染和神经认知障碍 (NCI) 之间的关系:一项观
察性流行病学研究的系统综述
背景: HIV/AIDS和潜在创伤事件 (PTE) 或应激性生活事件 (SLE) 和/或PTSD与神经认知障碍
(NCI) 独立相关。文献表明, HIV, PTE和SLE暴露独立且持续地影响认知的各个领域, 包括语
言能力, 工作记忆和精神运动速度。关于HIV感染与PTE之间的相互作用及其对NCI的综合影
响的数据有限。
目的: 在本系统综述中, 我们综合了HIV感染, PTE和SLE和/或创伤后应激障碍 (PTSD) 对高,
中, 低收入国家 HIV/AIDS携带者(PLWHA)NCI的综合效应的证据。
方法: 我们的纳入标准是观察流行病学研究 (病例对照, 队列研究和横断面设计), 且考查了
HIV感染, PTE和SLE和/或PTSD的交互作用, 特别是它们对成人NCI的综合效应。我们在包括
Pubmed/Medline, PsycINFO, Scopus和Global Health的众多电子数据库中, 使用以下搜索词
进行搜索:‘认知’, ‘HIV/AIDS’, ‘观察性研究’, ‘创伤’及其排列。
结果: 15项研究被纳入综述, 其中大多数在高收入国家进行。 15项研究中, 10项在美国
(USA) 进行, 5项在南非进行。其中有7项关注生早期生活应激/童年创伤。其余研究仅评估
成年后发作型PTE和SLE。8项研究仅包括女性。总体而言, 研究表明PTE和SLE暴露和/或
PTSD是HIV成年携带者NCI的显著风险因素, 记忆和执行功能损伤最有可能由PTE和SLE暴露
导致。
结论: 这些发现强调了创伤筛查以及在HIV护理中整合聚焦创伤干预措施的必要性, 以改善
干预结果。

1. Background

Advances in the treatment of HIV have dramatically
improved survival rates. HIV has transformed from an
acute terminal disease to a chronic pharmacologically-
managed condition through increased access to combi-
nation antiretroviral therapy (cART) around the globe.
Nevertheless, HIV infection is associated with a range of
sequelae including neurocognitive impairment (NCI).
The neuropathogenesis of NCI is complex and is char-
acterized by events such as oxidative stress, neuroin-
flammation, synaptic pruning and neuronal death
(Kumar et al., 2009; Valcour et al., 2012; Williams,
Zulu, Stein, Joska, & Naude, 2020). HIV crosses the
blood brain barrier (BBB) early in the course of infec-
tion. There is considerable evidence that NCI related to
HIV has a variable clinical trajectory (Alford & Vera,
2018; Habib et al., 2013; Heaton et al., 2015; Rubin &
Maki, 2019a). Symptoms of NCI in people living with
HIV/AIDS (PLWHA) are generally on a spectrum ran-
ging from asymptomatic neurocognitive impairment to
mild cognitive impairment to HIV associated dementia
(Antinori et al., 2007). The range of dysfunctions
include mental slowing, memory loss, and difficulties
in complex tasks, motor disorders, and behavioural
abnormalities (Simioni et al., 2010). Studies suggest
that between 30 and 50% of PLWHA experience mild
forms of NCI. Data from the United States of America
(USA) suggest that approximately 2% of PLWHA
experience a more serious form of NCI, HIV associated
dementia (Heaton et al., 2010). Today, while the inci-
dence of the most severe phenotype, HIV-associated

dementia has declined, subtle forms of the disease
such as asymptomatic neurocognitive impairment per-
sist despite patients being on cART (Ambrosius, Gold,
Chan, & Faissner, 2019). This may, in part, be attributed
to the limited penetration of the BBB by certain anti-
retrovirals (e.g. HIV protease inhibitors, nucleoside
analogues). Evidence of the effectiveness of BBB pene-
tration of antiretrovirals on cognition is equivocal, with
higher and lower CNS (central nervous system) pene-
tration effectiveness (CPE) of antiretrovirals associated
with cognitive improvements, as well a few studies
finding no association between CPE and cognitive ben-
efits (Yuan & Kaul, 2019

).

NCI in PLWHA has a complex aetiology and it is
likely that a number of risk factors may interact with
the HIV infection to predispose its onset (Heaton
et al., 2015). A number of studies from both high
and low to middle-income countries using neuroi-
maging and epidemiological techniques have exam-
ined additional factors potentially associated with
NCI in PLWHA. These include potentially traumatic
events (PTEs), and post-traumatic stress disorder
[PTSD] (a potential consequence of PTEs) (Kessler
et al., 2014; McLaughlin et al., 2017; Scott et al.,
2018), which have also been associated with NCI in
HIV-negative samples (Jelinek et al., 2006; Lagarde,
Doyon, & Brunet, 2010). PTEs during childhood,
particularly neglect, are associated with decline in
memory, executive function, and processing speed
(Malan-Muller et al., 2013; Spies, Ahmed-Leitao,
Fennema-Notestine, Cherner, & Seedat, 2016).
PTEs encompass a broad spectrum of exposures

2 G. SPIES ET AL.

including sexual, physical, and emotional abuse dur-
ing early life, adolescence, or adulthood. PTEs have
been found to be highly prevalent in PLWHA (Brief
et al., 2004; Decker et al., 2016; Machtinger, Wilson,
Haberer, & Weiss, 2012). These PTEs have also been
associated with a range of additional psychopatholo-
gies, most frequently PTSD and depression (Spies
et al., 2012). A recent systematic review by Rubin
and Maki examined the relationship between depres-
sion and NCI in PLWHA and found that depression
contributed to impairments particularly in the
domains of executive function, processing speed,
learning, and motor function (Rubin & Maki,
2019b). A broad body of evidence links PTSD to
impairments in multiple cognitive systems, including
processing speed, learning, memory, and executive
function (Aupperle, Melrose, Stein, & Paulus, 2012;
Qureshi et al., 2011; Schuitevoerder et al., 2013; Scott
et al., 2015; Sumner et al., 2017). In a meta-analysis
based on data from 60 studies totalling 4,108 parti-
cipants, including 1,779 with PTSD, 1,446 trauma-
exposed comparison participants, and 895 healthy
comparison participants without trauma exposure,
significant neurocognitive effects were associated
with PTSD (Scott et al., 2015).

While literature suggests that HIV infection, PTEs,
and/or PTSD independently affect various domains of
cognition including language ability, working memory,
and psychomotor speed (Grant, 2008; Heaton et al.,
2010; Tomoda et al., 2011), few studies have examined
the combined effect of PTEs, PTSD, and/HIV infection
on NCI (i.e. HIV+PTEs and/or PTSD on NCI). Spies
and colleagues who work in South Africa, against the
backdrop of a substantial HIV epidemic, examined the
relationship between HIV, trauma and NCI in women,
both as independent and combined exposures. They
found that PTEs were significantly associated with
memory impairment in women living with HIV
(Spies, Fennema-Notestine, Archibald, Cherner, &
Seedat, 2012). Findings from the Women’s Inter-
Agency HIV Study (WIHS) indicated an interaction
between psychological risk factors, including perceived
stress, anxiety, post-traumatic stress, and depressive
symptoms, and NCI, such that perceived stress and
anxiety were more strongly associated with deficits in
learning and memory among women living with HIV
compared to their uninfected counterparts (Maki et al.,
2015). Specifically in this study, depressive symptoms
were associated with a lower level of cognitive perfor-
mance (Maki et al., 2015). The findings suggest that the
neurobiological effects of psychological risk factors,
such as stress on cognition, may be different in HIV-
positive and HIV-negative women, with putative
mechanistic links to stress responsivity and immune
function (Rubin et al., 2017). Stress, PTSD, and depres-
sion are immunomodulatory and influence the immune
response in women infected with HIV. However, stress

may have a stronger influence than depression given
that stress is more strongly associated with regulatory
mechanisms necessary to maintain immune cell homo-
eostasis (Rehm & Konkle-Parker, 2017), in turn impact-
ing brain homoeostasis (de Groot & Burgas, 2015).
Despite the gradual emergence of a small body of
research examining the combined effect of PTEs and
SLEs and NCI, to our knowledge no review has synthe-
sised observational epidemiological studies of PTEs or
SLEs and/or PTSD and NCI in PLWHA. For this
review, we sought to synthesise findings from derived
from high-, middle-, and low- income countries to
contribute to the growing body of research exploring
the relationship between PTEs and SLEs, and/or PTSD,
HIV infection and NCI.

2. Methods

The process as outlined in PRISMA was adhered to
(Moher, Liberati, Tetzlaff, & Altman, 2009).

2.1. Inclusion and exclusion criteria

Eligible studies had to be in English and include adults
(18 years and older) only. There was no limit on the
date of publication. We included observational epide-
miological studies (i.e. cross-sectional, cohort, or case-
control studies) that examine the relationship between
PTEs and/or PTSD and NCI in PLWHA. Intervention
studies were also excluded. In this instance, trauma was
broadly defined as traumatic or stressful events (PTEs
and SLEs) during childhood and/or adulthood. PTEs
included both early life and adult-onset traumatic
events that met Criterion A for a traumatic event for
PTSD as well as other stressful life events (SLEs, e.g.
economic hardship, food insecurity) and traumatic
brain injury. For a study to be included, NCI had to
be measured by a full neuropsychological battery.
Studies using self-reported measures of NCI were
excluded unless they included z-scores based on pub-
lished normative data. Moreover, studies using screen-
ing instruments only for cognitive impairment (e.g. the
(International) HIV Dementia Scale and the Montreal
Cognitive Assessment) were excluded. Inclusion and
exclusion criteria are presented in Table 1.

2.2. Search strategies

Prior to conducting the searches, we collaboratively
decided on the search terms. A number of electronic
databases were searched, namely PUBMED, PsycINFO,
Scopus, and Global Health using the search terms:
cognition, HIV/AIDS, observational studies, trauma,
and permutations thereof. The searches began in
March 2018. A total of 677 abstracts were extracted to
a Rayyan database, a web and mobile application for
systematic reviews (Ouzzani, Hammady, Fedorowicz, &

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 3

Elmagarmid, 2016). The searches were updated in
May 2020 using the following electronic databases:
PUBMED, Psych Info, and Google Scholar. Reference
lists of eligible studies were also scanned. The abstracts
were then reviewed independently by three reviewers to
see if they met inclusion criteria. A face-to-face meeting
was subsequently held whereby reviewers discussed dis-
crepancies and reached consensus about which studies
to include. Each reviewer flagged duplicate entries on
Rayyan, with removal of the duplicates from the data-
base via consensus. See Prisma flow diagram (Figure 1).

2.3. Data analysis

Data were extracted by two independent reviewers
using a standard data extraction form. The form
included the following fields: author and date,
country, study design, sample, trauma, and PTSD
measurement and summary of NCI measures.
Quality appraisal of included studies was conducted
by GS and SM. However, given that the seminal
studies were conducted by two independent
reviewers, a third objective reviewer was brought
in to assist with the quality appraisal of these stu-
dies. The quality appraisal was guided by the

Systematic Appraisal of Quality in Observational
Research (SAQOR) tool that comprises six domains
(each containing two to five questions): sample,
control/comparison group, exposure/outcome mea-
surements, follow-up, confounders, and reporting
of data (Ross et al., 2011). Table 2 presents the
quality appraisal.

3. Results

3.1. Quality assessment of included studies

The majority of studies (n = 12; 80%) included were
deemed to be of high quality. Two studies were of
moderate quality and only one study was deemed to
be of low quality due to inadequate description of
distorting influences/confounders and the sample and
comparison groups. See Table 2 for more detail on
quality appraisal of included studies.

3.2. Study characteristics

Tables 3 and 4 provide characteristics of the fifteen
selected studies meeting inclusion criteria. No studies
including a mixture of eligible and non-eligible

Table 1. Study inclusion and exclusion criteri

a.

Inclusion criteria Exclusion criteria

● Studies investigating neurocognitive impairment in the context of HIV
and potentially traumatic or stressful events, specifically the combined
impact of trauma on cognitive impairment in the context of HIV

● Studies not investigating the impact of potentially traumatic or
stressful events on cognitive impairment in the context of HIV

● Adult samples (≥18) ● Youth samples (<18

)

● Observational studies ● Randomised controlled trials, quasi-experimental study designs

and qualitative studies.

● English language studies only ● Non-English language studies
● NCI measured by a full neuropsychological battery or self-reported

measures of NCI that included z-scores based on published normative
data

● Self-reported measures of NCI that did not include z-scores
based on published normative data

● Screening instruments of cognitive impairment (e.g. IHDS; HDS;
MoCA, etc.)

Figure 1. PRISMA flow diagram of search procedure.

4 G. SPIES ET AL.

Ta
b

l

e

2

.
Q

ua
lit

y

as

se
ss

m
en

t

of

p
ap

e

r

s

i

n
cl

u

d
ed

i
n

t

h

e
s

y

st
em

at
ic

r
ev

ie
w

(
SA

Q
O

R

to

ol
).

St
ud

y
Sa

m
p

le
C

ou
n

tr
y

C
on

tr
ol

/c
om

p
ar

is
on

g
ro

up
Ex

p
os

ur
e/

ou
tc

o

m
e

m
ea

su
re

s
D

is
to

rt
in

g
i

n
flu

e

n
ce

s
Re

p
or

ti
n

g
o

f

d

at
a

O
ve

ra
ll

st
ud

y

q

ua
lit
y

C
la

rk
e

t

al

.,

20

12
Sa

m
p

le
r

ec
ru

it
ed

4
9

H
IV

+
a

n
d

47

–

in

d
iv

id
ua

ls
w

it
h

E
LS

t
o

ex
p

lo
re

e
ff

ec
ts

o
f

H
IV

a
n

d
E

LS

on
A

m
yg

d
al

a

m

or
p

h
om

et
ry

an

d
N

C
I.

In
s

u

m
m

ar
y,

s
am

p
le

is

c
le

ar
.

U
SA

C
om

p
ar
is
on

g
ro

up
(

47
H

IV
–

in
d

iv
id

ua
ls

)
in

cl
ud

e

d
,

ea
si

ly

i

d
en

ti
fia

b
le

,
so

ur
ce

o
f

co
m

p
ar
is
on
g
ro

up
c

le
ar

.

In

su

m
m
ar
y,

c
om

p
ar
is
on
g
ro

up
is

cl

ea

r.

M
ai

n
e

xp
os

ur
e:

E
ar

ly
L

ife
S

tr
es

s
Q

ue
st

io
n

n
ai

re

M
ai

n
o

ut
co

m
e:

s
M

RI
a

n
d

as
se
ss
m
en

ts
o

f

p

sy
ch

o

m
ot

or

s

p
ee

d
,

ex
ec

ut
iv

e

fu

n
ct

io
n

s,

ve
rb

al
m

em
or

y
an

d
f

in
e

m
ot
or

d
ex

te
ri

ty
.

In
s

um
m

ar
y,

q
ua

l

it
y

of

ex
p

os
ur

e/
ou

tc
om

e
m

ea
su

re
m

e

n
t

is
a

d
eq

ua
te

C
on

fo
un

d
er

s
(d

em
og

ra
p

h
ic

va

ri
ab

le
s

th
at

d
iff

er
ed

si

g
n

ifi
ca

n
tl

y

b

et
w

ee
n

g
ro

up
s)

.
In

s
um

m
ar

y,
d

es
cr

ip
ti

on
o

f
in

f

lu
en

ce
s

ar
e

un
cl

ea
r.

M
is

si
n

g
d

at
a

M
is
si
n
g
d
at
a

n
ot

re

p
or

te
d

.
In
s
um
m
ar

y,

re
p

or
ti

n
g

o
f

d
at

a
is

u
n

cl
ea

r.

H

ow
ev

er
,

d
at

a
ar

e
cl

ea
rl

y
an

d

ac
cu

ra
te

ly
p

re
se

n
te

d
w

it
h

p

v
al

ue
s

re
p

or
te

d
.

N
o

co
n

f

id
en

ce
i

n
te

rv
al

s
re

p
or
te
d
.
In
s
um
m
ar

y,
r

ep
or

ti
n
g
o

f
d

at
a
is
a
d
eq
ua
te
.

M
od

er
at

e
C
la
rk
e

t
al

.,
20

18
Sa

m
p
le
r
ec
ru
it
ed

4
4

H
IV

+

p
ar

ti
ci

p
an

ts
t

o

ex

p
lo

re

ch
an

g
es

i
n

b
ra

in
v

ol
um

e
af

fe
ct

ed
b

y
EL

S.
S

ou
rc

e,

m
et

h
od

a
n

d
e

n
tr

y
cr

it
er

ia

cl
ea

rl
y

st
at

ed
.

In
s
um
m
ar
y,

sa

m
p

le
i

s

ad

eq
ua

te
.

U
SA
C
om
p
ar
is
on
g
ro

up
i

n
cl

ud
ed

,
ea

si
ly

i
d

e

n
ti

fia
b

le
,

so
ur

ce
o

f

co

m
p

ar
is

o

n
s

ex
p

la
in

ed
,

co
m
p
ar
is
on

s
m

at
ch

ed
,
d
iff

er
en

ce
s

b
et

w
ee

n
c

as
es

a
n
d

co
m
p
ar
is
on

s
co

n
tr

ol
le

d
f

or
.

I

n

su
m

m
ar

y,
c

om
p

ar
is

on
g

ro
up

is

ad
eq

ua
te
.
M
ai
n
e
xp
os
ur
e:
E
ar
ly
L
ife
S
tr
es
s
Q
ue
st
io
n
n
ai

re
(

EL
SQ

)

M

ai
n

o
ut

co
m

e:
R

ea
ct

io
n

T
im

e
Ta

sk

St
ru

ct
ur

al
M

RI

Se
lf-

re
p
or
te

d
c

og
n

it
iv

e
fu
n
ct
io
n

t
h

ro
ug

h
T

h
e

H
IV

M

ed
ic

al
O

ut
co

m
es

S
ur

ve
y

(M
O

S-
H

IV
).

In
s
um
m
ar
y,

q
ua
lit

y
of

e
xp

os
ur
e/
ou
tc
om
e
m
ea
su
re
m

en
t

is
a
d
eq
ua
te
.
C
on
fo
un
d
er

s

(n

eu
ro

p
sy

ch
ia

tr
ic

sy

m
p

to
m

s
an

d
s

ca
n

n
er

t
yp

e)

ad
eq
ua
te

ly
c

on
tr

ol
le
d
f
or
.

In

su
m
m
ar
y,
d
es
cr
ip
ti
on
o
f
in

flu
en

ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
n
ot

r
ep

or
te
d
.

H
ow

ev
er

,
d

at
a
ar
e
cl
ea
rl
y

an
d

ac

cu
ra

te
ly

p
re

se
n

te
d

w
it

h

p
v

al
ue

s
re
p
or
te
d

.

N

o
co

n
fid

en
ce

i
n

te
rv

al
s

re
p
or
te
d
.
In
s
um
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

a
d

eq
ua
te
.

H
ig

h

D
ei

ss
e

t
al
.,
20

19
C

o

h
or

t
re

cr
ui

te
d

2

00

m
ili

ta
ry

b

en
ef

ic
ia

ri
es

l
iv

in
g

w
it

h
H

IV
t

o
ex
p
lo

re
t

h
e

re
la

ti
on

sh
ip

b
et
w
ee
n

N
C

I
an

d
m

en
ta

l

h

ea
lt

h
d

is
or

d
er

s.
S

ou
rc

e
of

t
h

e
co

h
or

t
is

u
n
cl
ea

r

.
M

et
h

o

d
a

n
d

en

tr
y

cr
it

er
ia

c
le

ar
ly

s
ta

te
d
.
In

su
m
m
ar
y,

c
oh

or
t

is
u

n
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
i
n
cl
ud
ed

(in

d
iv
id
ua
ls
w
it
h

ou
t

lif
et

im
e

h
is

to
ry

o
f

PT
SD

),
ea

si
ly

id

en
ti

fia
b
le
,
so
ur
ce
o

f
co

m
p
ar
is
on
g
ro
up
u
n
cl
ea

r.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

un

cl
ea
r.
M
ai
n
e
xp
os
ur
e:

C

om
p

os
it

e
In

te
rn

at
io

n
al

D

ia
g

n
os

ti
c

In
te

rv
ie

w
.

(

C
ID

I

)
M

ai
n
o
ut
co
m

e:
N

C
I

b
at

te
ry

m
ea
su

ri
n

g
v

er
b

al
f

lu
en

cy
,

at
te

n
ti

on
/w

or
ki

n
g

m
em

or
y,

vi

su
os

p
at

ia
l

fu
n

ct
io

n
in

g
,

i

n
fo

r

m
at

io
n

p
ro

ce
ss

in
g

,

le

a

r
n

in
g

/r
ec

al
l,

ex
ec
ut
iv
e
fu
n
ct
io
n

s,
m

ot
or

s
p

ee
d

a
n
d

d
ex
te
ri

ty
a

n
d
e
ff

or
t.

).
In

su
m
m
ar
y,
q
ua

lit
y

of
e

xp
os
ur
e/

ou

tc
om
e
m
ea
su
re
m
en
t
is

ad
eq
ua
te
.
C
on
fo
un
d
er

s
(a

g
e,

e
th

n
ic

it
y,

ed

uc
at

io
n

)
ad

eq
ua
te
ly

co
n
tr
ol
le
d
f
or
.

Po
te

n
ti

al

co
n
fo
un
d
er

s
in

cl
ud
in
g

cu

m
ul

at
iv

e

tr

au
m

as
,

w
ar

d

ep
lo

ye
d

t
o
an
d

r
eg

io
n
o
f

re
si

d
en

ce
n

ot
d

es
cr

ib
ed

.
In

su
m
m
ar
y,

d
es

cr
ip

ti
on
o
f

in
flu

en
ce

s
ar

e
un

cl
ea
r.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d

ac
cu
ra
te
ly

p
re

se
n
te
d
w
it
h

p
v
al
ue
s
an
d
c

on
fid

en
ce

in
te
rv
al
s
re
p
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

ad
eq
ua
te
.

Lo
w

Ka
p

et
an

ov
ic

et

a
l.,

2
02

0
18

7
in

d
iv
id
ua

ls
(

10
2

w
it
h
H

IV

an
d

I
PT

;
35

w
it
h
H

IV
o

n
ly

;
24

H

IV
-n

eg
at

iv
e

w
it

h
I

PT
;

26
H

IV
–

n
eg

at
iv

e)
. S

ou
rc
e
of
t
h
e
co
h
or
t
is
c
le

ar
.

M
et

h
od
a
n
d
e
n
tr
y
cr
it
er

ia
c

le
ar

ly
s

ta
te

d
.
In

su
m
m
ar

y,
s

am
p

le
i
s
ad
eq
ua
te
.
U
SA
C
om
p
ar
is
on
g
ro

up
in

cl
ud

e

d
(

H
IV

–
p

os
it
iv
e
w
it

h
n

o
IP

T
an

d
H

IV
–
n
eg
at
iv

e

w

it
h
a
n
d
w
it
h
ou
t

IP
T)

,
ea
si
ly
i
d
en
ti
fia
b
le
,
so
ur

ce

of
c

om
p
ar
is
on
s
ex
p
la
in
ed
,
co
m
p
ar
is
on
s
m
at
ch
ed
,
d
iff
er
en
ce
s
b
et
w
ee
n
c
as
es
a
n
d

co
m
p
ar
is
on
s
co
n
tr
ol
le
d
f
or
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up

is

ad
eq
ua
te
.
M
ai
n
e
xp
os
ur
e:

C
lie

n
t

D
ia

g
n

os
ti

c
Q

ue
st
io
n
n
ai
re
(

C
D

Q
).

M
ai
n
o
ut
co
m
e:

S
tr

uc
tu

ra
l

M
RI

an
d
a

ss
es

sm
en

ts
o

f

at

te
n

ti
on

/
w

or
ki
n
g

m
em

or
y,

p
sy
ch

om
ot

or
s

p
ee
d
,
ex
ec
ut
iv
e
fu
n
ct
io
n

s,
i

n
fo

rm
at

io
n

p

ro
ce

ss
in

g
,
ve
rb
al
f
lu
en
cy
,
le
ar
n
in
g
,
m
em
or
y,
a
n
d
a

d
ul

t
re

a

d
in

g

In
s
um
m
ar
y,
q
ua
lit
y
of

ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua
te
.
C
on
fo
un
d
er

s
(g

ro
up
s
ta

tu
s,

s
ex

,
ag

e,
r

ac
e,

a
n
d
e

d
uc

at
io

n
)

w
er

e
ad

ju
st

ed
f

or
.

A
lr

ea
d

y
ad

ju
st

ed

T-
sc

or
es

w
er

e
ap

p
lie

d
.
M
od

el
s

w
er

e
al

so
a

d
ju

st
ed

f
or

W
TA

R
(a

d
ul
t
re
ad
in

g
)

sc
or

es
.

In
s
um
m
ar
y,
d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
cl
ea
r.

D
at

a
ar
e
cl
ea
rl
y
an
d
a

cc
ur

at
el

y
p

re
se
n
te
d
w
it
h

p
v

al
ue
s
an
d

co
n

fid
en

ce
i
n
te
rv
al
s
re
p
or
te
d
.
In
s
um
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is
a
d
eq
ua
te
.
H
ig
h

(C
on

ti
nu

ed
)

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 5

Ta
b

le
2

.

(C

on
ti

n
ue

d
).

St
ud
y
Sa
m
p
le
C
ou
n
tr
y
C
on
tr
ol
/c
om
p
ar
is
on
g
ro
up
Ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
s
D
is
to
rt
in
g
i
n
flu
en
ce
s
Re
p
or
ti
n
g
o
f
d
at
a
O
ve
ra
ll
st
ud

y
q

ua
lit
y

Li
n

e
t

al
.,

20
11

96
4

H
IV

+
i

n
d
iv
id
ua
ls
w
it
h
T

BI
.

So
ur

ce
,

m
et
h
od
a
n
d
e
n
tr
y
cr
it
er
ia
c
le
ar
ly
s
ta
te
d
.
In

su
m
m
ar
y,
s
am
p
le
i
s
ad
eq
ua
te
.
U
SA
C
om
p
ar
is
on
g
ro
up
i
n
cl
ud
ed

(in
d
iv
id
ua
ls
w
it
h
ou
t

TB
I),

e
as

ily

id
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro
up
c
le
ar
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

cl
ea
r.
M
ai
n
e
xp
os
ur
e:

se

l

f-
re

p
or
te
d

m
ed

ic
al

h
is

to
ry
.
M
ai
n
o
ut
co
m
e:
N
C
I
b
at
te
ry

m
ea
su
ri
n
g
o

ve
ra

ll

vo

ca
b

ul
ar

y,

ve
rb
al
f
lu
en
cy
,
at
te
n
ti

on
/

w
or

ki
n

g
m

em
or

y,
l

ea
rn

in
g

,
in

fo
rm

at
io

n
p

ro
ce
ss
in
g
,
le
ar
n
in

g
/r

ec
al

l

,
ex

ec
ut

iv
e
fu
n
ct
io

n
s,

m
ot

or
s
p
ee
d
a
n
d

d

ex
te

ri
ty

).
In
s
um
m
ar

y,
q

ua
lit
y
of
e
xp
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua
te
.
C
on
fo
un
d
er
s
(a
g
e,

se

x,
e

d
uc
at
io

n
,

an
d
e
th
n
ic

it
y)

ad
eq
ua
te
ly

c
on

tr
ol

le
d

f
or

.
H

ow
ev
er
,

n
um

b
er

o
f

TB
Is

a
n
d

ty
p

e
of

T
BI

w
er

e

n

ot

co
n
tr
ol
le
d
f
or
.
In
s
um
m
ar
y,

d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
ad
eq
ua
te
ly

re
p
or
te

d
.

M
od
er
at
e

M
al

an
-M

ul
le

r
et

a
l.,

2
01

3
Th

e
sa

m
p
le
r
ec
ru
it
ed

1
28

w

om
en

.
So

ut
h

A
fr

ic
a

Th
e

sa
m

p
le
i
n
cl
ud

ed
n

=
8

3
w

h
o

w
er

e
H

IV
p

os
it

iv
e.

M
ai
n
e
xp
os
ur
e:

c
h

ild
h

oo
d

tr
au
m

a
m

ea
su

re
d

b
y

th
e

c

h
ild

h
oo

d
t

ra
um

a
q

ue
st
io
n
n
ai
re
(

C
TQ

)
m

ea
su
ri
n
g

28
e

x

p
er

ie
n

ce
s
of
c
h
ild
h
oo
d

tr
au

m
a.

M
ai
n
o
ut
co
m

e:

n
eu

ro
co

g
n
it
iv

e
te

st
in

g
u

si
n
g

th
e
In
te

rn
at

io
n
al

N
eu

ro
p

sy
ch

ol
og

ic
al

T
es

t
Ba

tt
er

y
d

ev
el

op
ed

b
y
th
e
H
IV

N
eu

ro
b

eh
av

io
ra

l
Re

se
ar

ch

C
en

tr
e

(H
N

RC
).

A
r

es
ea

rc
h

p
sy
ch
ol
og

is
t

co
n
d
uc
te
d
t
h
e
te
st
in

g
e

xa
m

in
in

g
m
ot
or

,
ve

r

b
al

f
lu

en
cy

,
at

te
n
ti
on

,
w

or
ki
n
g
m
em
or
y,
s
p
ee
d

,
le

ar
n
in
g

r
ec

al
l

an
d

e
xe

cu
ti

ve

fu
n
ct
io

n
.

C
on
fo
un
d
er
s
(a
g
e,

e
d

uc
at
io
n

,
Bo

d
y

M
as

s
In

d
ex

,
tr

au
m

a
su

b
–

ty
p

e,
t

ra
um
at
ic

l
ife

ex
p
er
ie
n

ce
s,

p
os

t-
tr

au
m
at
ic

st

re
ss

d
is

or
d

er
(

PT
SD

),
sy

m
p
to
m

at
ol

o

g
y

an
d

a
lc

o

h
ol

ab

us
e)

, L
TL

, A
RV

t
re

at
m

en
t.

A
s

al
l
p
ar
ti
ci
p
an

ts
w

er
e

b
la

ck
s

o
et

h
n

ic
it

y
w

as
n

ot
c

on
tr
ol
le
d

fo
r

i

n
t

h
e

an
al

ys
is

.
M
is
si
n
g
d
at
a
n
ot

m
en

ti
on

ed
.

H
ig
h

Sp
ie

s
et

a
l.,

20
12
Sa
m
p
le
r
ec
ru
it
ed
c
on

si
st

ed
o

f
83

(6

4

%
)

H
IV
p
os
it
iv

e
an

d
4

7
(3

6%
)

H
IV

-n
eg

at
iv

e
w

om
en
.

So
ut

h
A

fr
ic

a
Fo

rt
y-

ei
g

h
t

H
IV

-p
os

it
iv
e
w
om
en
w
er

e
ex

p
os

ed
t

o

ch

ild
h
oo
d

t
ra

um
a.

A

m
on

g
c

on
tr

ol
s,

2
0

h
ad

b
ee

n

ex
p

os
ed

t
o

ch
ild

h
oo
d
t
ra
um
a.
M
ai
n
e
xp
os
ur
e:
c
h
ild
h
oo
d

tr
au
m
a
m
ea
su
re
d
b
y
th
e
ch
ild
h
oo
d
t
ra
um
a
q
ue
st
io
n
n
ai
re
(
C
TQ
)
m
ea
su
ri
n
g

28
e

xp
er

ie
n
ce
s
of
c
h
ild
h
oo
d

tr
au
m
a.

M
ai
n
o
ut
co
m
e:

n
eu
ro
co
g
n
it
iv
e
te
st
in
g
u
si
n
g

th
e
In
te
rn
at
io
n
al

N
eu
ro
p
sy
ch
ol
og
ic
al
T
es
t
Ba
tt
er
y
d
ev
el
op
ed
b
y
th
e
H
IV
N
eu
ro
b
eh
av
io
ra
l
Re
se
ar
ch

C
en
tr
e
(H
N
RC
).
A
r
es
ea
rc
h

p
sy
ch
ol
og
is
t
co
n
d
uc
te
d
t
h
e
te
st
in
g
e
xa
m
in
in
g
m
ot
or
,
ve

rb
al

f
lu
en
cy
,
at
te
n
ti
on
,
w
or
ki
n
g
m
em
or
y,
s
p
ee
d
,
le
ar
n
in
g
r
ec
al
l
an
d
e
xe
cu
ti
ve

fu
n
ct
io
n
.

A
g

e
an
d
e
d
uc
at
io
n
c

or
re

ct
ed

z-

sc
or

es
w

er
e

ca
lc

ul
at

ed
f

ro
m

al

l
ra

w
N

P
d

at
a
an
d
g
ro

up
ed

in

to
d

om
ai

n
s

(m
ot

or
,

ve
rb
al

flu
en
cy
,
w
or
ki
n
g
m
em
or
y,

sp
ee

d
,
le
ar
n
in
g
,

re
ca

ll,
a

n
d

ex
ec
ut
iv
e
fu
n
ct
io

n
s)

.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.

Bo
th

st

at
is

ti
ca

lly
a

n
d

n
on

–
st

at
is
ti
ca

lly
s

ig
n

ifi
ca
n
t

va
ri

ab
le

s
ar

e
re

p
or
te
d
.
H
ig
h
(C
on
ti
nu
ed
)

6 G. SPIES ET AL.

Ta
b
le
2

.
(C

on
ti
n
ue
d
).
St
ud
y
Sa
m
p
le
C
ou
n
tr
y
C
on
tr
ol
/c
om
p
ar
is
on
g
ro
up
Ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
s
D
is
to
rt
in
g
i
n
flu
en
ce
s
Re
p
or
ti
n
g
o
f
d
at
a
O
ve
ra
ll
st
ud
y
q
ua
lit
y
Sp
ie
s
et
a
l.,

20

16
A

s
am
p
le
o
f

12
4

w
om

en
t

es
te

d

fo
r
H
IV
s
ta

tu
s

w
er
e
re
cr
ui
te
d
.
So
ut
h
A
fr
ic
a

62
w

er
e
H
IV
-p
os
it
iv

e,
3

2
w

it
h

a

h
is
to
ry
o
f
ch
ild
h
oo
d
t
ra
um

a
an

d

30
w

it
h

ou
t,

a
n

d
6

2
w
er
e
H
IV

–
n

eg
at

iv
e,

3
1

w
it

h
a

h
is
to
ry
o
f
ch
ild
h
oo
d
t
ra
um
a
an

d
3

1
w

it
h

ou
t.

M
ai
n
e
xp
os
ur
e:
c
h
ild
h
oo
d

tr
au
m
a
m
ea
su
re
d
b
y
th
e
ch
ild
h
oo
d
t
ra
um
a
q
ue
st
io
n
n
ai
re
(
C
TQ
)
m
ea
su
ri
n
g

28
e
xp
er
ie
n
ce
s
of
c
h
ild
h
oo
d

tr
au
m
a.

M
ai
n
o
ut
co
m
e:
N
eu
ro

co
g

n
it

iv
e
fu
n
ct
io

n
w

as
m

ea
su
re
d
w
it
h

10
t

es
ts

t
h

at
h

av
e

b
ee

n
u

se
d

co
m
m

on
ly

i
n

H
IV

r
es

ea
rc

h
a
n
d

co

ve
r

se
ve

n
a

b
ili

ty
d

om
ai
n
s

(le
ar

n
in
g
,

d
el

ay
ed

r
ec
al
l,
p
ro
ce
ss
in
g
s
p
ee
d
,
at
te
n
ti
on

/w
or

ki
n
g
m
em
or
y,

ex
ec
ut
iv
e
fu
n
ct
io
n
,
ve
rb
al

flu

en
cy

,
m

ot
or

a
b

ili
ty

).
A
g

e,
m

ar
it

al
s

ta
tu

s,
e

th
n

ic
it
y,

em
p

lo
ym

en
t
st
at

us
,

ye
ar

s
of

ed
uc
at
io
n
a
n
d
e

st
im

at
ed

in

tr
ac

ra
n

ia
l

va
ul

t,
b

ra
in

vo

lu
m

es
w
er
e

ca
p

tu
re

d
.

Es
ti

m
at

e

d
i

n
tr

ac
ra

n
ia

l
va

ul
t

w
as

a
d
d
ed

a
s

a
co

va
ri

at
e

to

co
n
tr
ol
f
or
i
n
d
iv
id
ua

l
d

iff
er

en
ce
s
in

h
ea

d
s

iz
e.

M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
Bo
th

st
at
is
ti
ca
lly
a
n
d
n
on
–
st
at
is
ti
ca
lly
s
ig
n
ifi
ca
n
t
va
ri
ab
le
s
ar
e
re
p
or
te
d
.
H
ig
h
Sp
ie
s
et
a
l.,

20

17
A

s
am
p
le
o
f

n
=

6
7

(5
1.

9%
)

H
IV
+

an
d

5
0(

38
.8

%
)
H
IV

−
w

om
en
.
So
ut
h
A
fr
ic
a

Fi
ft

y-
th

re
e

H
IV

+
w

om
en
a
n

d
1

8
co

n
tr

ol
s

w
er
e
ex
p
os
ed
t

o
ch

ild
h
oo
d
t
ra
um
a.
M
ai
n
e
xp
os
ur
e:
c
h
ild
h
oo
d

tr
au
m
a
m
ea
su
re
d
b
y
th
e
ch
ild
h
oo
d
t
ra
um
a
q
ue
st
io
n
n
ai
re
(
C
TQ
)
m
ea
su
ri
n
g

28
e
xp
er
ie
n
ce
s
of
c
h
ild
h
oo
d

tr
au
m
a.

M
ai
n
o
ut
co
m
e:
N
eu
ro
co
g
n
it
iv
e
fu
n
ct
io
n
w
as
m
ea
su
re
d
w
it
h

10
t
es
ts
t
h
at
h
av
e
b
ee
n
u
se
d

co
m
m
on
ly
i
n
H
IV
r
es
ea
rc
h
a
n
d

co
ve
r
se
ve
n
a
b
ili
ty
d
om
ai
n
s
(le
ar
n
in
g
,
d
el
ay
ed
r
ec
al
l,
p
ro
ce
ss
in
g
s
p
ee
d
,
at
te
n
ti
on
/w
or
ki
n
g
m
em
or
y,

ex
ec
ut
iv
e
fu
n
ct
io
n
,
ve
rb
al

flu
en
cy
,
m
ot
or
a
b
ili
ty
)
A
g
e,
m
ar
it
al
s
ta
tu
s,
e
th
n
ic
it
y,

ye
ar
s
of
e
d
uc
at
io
n
a
n
d

em
p
lo
ym
en
t
st
at

us
w

er
e
ca
p
tu
re
d
.

A
g

en
er

al
p

h
ys

ic
al

ex

a

m
in

at
io
n
w

as
c

on
d

uc
te

d

on
a

ll
p

ar
ti

ci
p

a

n
ts

.

V

ir
ol

og
ic

m

ar
ke

rs
o

f
d

is
ea

se
p

ro
g

re
ss
io
n

(C

D
4

ly
m

p
h

oc
yt

e
co

un
t

an
d

vi
ra
l

lo
ad

s)
w

er
e
ca
p
tu
re
d
.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
H
ig
h

Pu
ka

y-
M

ar
ti
n

et
a

l.,
2

00
3

25
1

H
IV
+
a
n
d

8
2

H
IV

–
g

ay
o

r
b

is
ex

ua
l

m
en

.
So

ur
ce
c
le
ar
ly

st
at
ed
.
M
et
h

od
a

n
d

e
n

tr
y
cr
it
er
ia
u
n
cl
ea
r.
In
s
um
m
ar
y,

sa
m
p
le

i
s

un
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
in
cl
ud
ed
(
H
IV

–
m

en
),

ea
si

ly
i

d
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro

up

cl
ea
r.
In
s
um
m
ar
y,
c
om
p
ar
is

on

g
ro
up
i

s
cl

ea
r.
M
ai
n
e
xp
os
ur
e:

P
sy

ch
ia
tr
ic

Ep

id
em

io
lo

g
y

Re
se

ar
ch

I
n

te
rv
ie
w

(
PE

RI
)

Li
fe

Ev

en
ts

S
ca

le
,
m
ea

su
ri

n
g

4
7

tr
au
m
at

ic
e

ve
n

ts
.

M
ai
n
o
ut
co
m
e:

W

ec
h

sl
er

A
d

ul
t
In
te

lli
g

en
ce
S
ca

le
-R

ev
is

ed
,

Se
le

ct
iv

e
Re

m
in
d
in

g
T

es
t,

V

is
ua

l.
M

em
or

y
Sp

an
F

or
w

ar
d

a
n
d

Ba
ck

w
ar
d
f
ro
m
t
h

e
W

ec
h
sl
er

M
em
or

y
Sc

al
e-

Re
vi

se
d

,
V

er
b
al

C
on

ce
p

t
A

tt
ai

n
m

en
t

Te
st

,
W

is
co

n
si

n
C

ar
d

S
or

ti
n
g

Te
st
,
V
er
b

al
F

lu
en
cy
,

Fi
g

ur
al

F
lu

en
cy

,
Tr

ai
l

M
ak

in
g

A
a
n

d
B

,
G

ro
ov

ed

Pe
g

b
oa

rd
,

an
d
t
h

e
Pa

ce
d

A

u

d
it

or
y

Se
ri

al
A

d
d

it
io

n
T

es
t.

In
s
um
m
ar
y,
q
ua
lit
y
of

ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua
te
.
C
on
fo
un
d
er

s
(im

p
ac

t
of
a
n

xi
et

y,

d
ep

re
ss
io
n
,
ag
e
an
d
e
d
uc
at
io
n
)
co
n
tr
ol
le
d
f
or
.
In
s
um
m
ar
y,
d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
an
d

d
at

a
an

al
ys

is

p
ro
ce
d

ur
es

n
ot

r
ep
or
te
d
.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d

ac
cu
ra
te
ly
p
re
se
n
te
d
w
it
h

p
v
al
ue
s
re
p
or
te
d
.
N
o
co
n
fid
en
ce
i
n
te
rv
al
s
re
p
or
te
d
.
In
s
um
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is
a
d
eq
ua
te
.
H
ig
h
(C
on
ti
nu
ed
)

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 7

Ta
b
le
2
.
(C
on
ti
n
ue
d
).
St
ud
y
Sa
m
p
le
C
ou
n
tr
y
C
on
tr
ol
/c
om
p
ar
is
on
g
ro
up
Ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
s
D
is
to
rt
in
g
i
n
flu
en
ce
s
Re
p
or
ti
n
g
o
f
d
at
a
O
ve
ra
ll
st
ud
y
q
ua
lit
y

Ru
b

in
e

t
al
.,
20

15
Sa

m
p
le
r
ec
ru
it
ed

1
00

9
H

IV
–

in
fe

ct
ed
a
n
d
4

96
a

t-
ri

sk
H

IV
–

un
in

fe
ct

ed
w

om
en
t
o
ex
p
lo
re

th

e
as

so
ci

at
io

n
b

et
w
ee
n

p

er
ce

iv
ed

s
tr

es
s

an
d

v
er

b
al

m
em
or

y

fr

om
t

h
e

W
om

en
’s

In

te
r-

A
g
en
cy
H
IV

S
tu

d
y

(W
IH

S)
.

In
s
um
m
ar
y,
s
am
p
le
i
s
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
i
n
cl
ud
ed

4
96

at

-r
is

k
H

IV
-u

n
in
fe
ct
ed
w
om
en
,
ea
si
ly
i
d
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro
up
c
le
ar
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

cl
ea
r.
M
ai
n
e
xp
os
ur
e:

P
er

ce
iv

ed
S

tr
es

s
Sc

al
e

(P
SS

-1

0)

M
ai
n
o
ut
co
m

e:
v

er
b
al

m
em
or
y
p
er
fo
rm

an
ce

u
si

n
g

th
e
H

op
ki

n
s

V
er

b
al

L
ea

rn
in

g

Te
st

(H

V
LT

).
O

t

h
er

d
om

ai
n

s
as

se
ss

ed
i

n
cl
ud
ed
v
er
b
al

le
ar
n
in
g
,
at
te
n
ti
on
a
n
d

co
n

ce
n

tr
at

io
n
,
ex
ec
ut
iv
e
fu
n
ct
io
n
in
g

(b
eh

av
io

ur
al

in
h

ib
it

io
n

, m
en

ta
l

fle
xi

b
ili

ty
,

w
or
ki
n
g
m
em
or

y)
,

p
sy

ch
om

ot
or
s
p
ee
d
,
ve
rb
al

flu
en
cy

,
an

d
f
in
e
m
ot
or
s

ki
lls

.
In
s
um
m
ar
y,
q
ua
lit
y
of

ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua

te
.

In

su
m
m
ar
y,
q
ua
lit
y
of
e
xp
os
ur

e/

ou
tc
om
e
m
ea
su
re
m
en
t
is

ad
eq
ua
te
.
C
on
fo
un
d
er
s
(a

n
n

ua
l

h
ou

se
h

ol
d

in
co

m
e;

d
ep

re
ss
io
n

;
H

ep
at

it
is

C
s

ta
tu

s;
s

el
f-

re
p
or
te
d
r
ec

en
t,

f
or

m
er

,
or

n
ev
er

u
se

o
f
m
ar

iju
an

a,
c

ra
ck

,
co

ca
in

e,
a

n
d

/o
r

h
er

oi
n

a
n
d

sm
ok

in
g

;
se

lf-
re

p
or
te
d
r
ec
en
t
h
ea

vy
a

lc
oh

ol
u

se
;

re
ce

n
t

an
ti

d
ep
re
ss

an
t

us
e;

a
n
d
s

tu
d

y
g

eo
g

ra
p
h
ic

s
it

e.
A

d
d
it
io
n
al

H

IV
-r

el
at

ed
c

lin
ic

al
v

ar
ia

b
le
s
of

in

te
re

st
i

n
cl
ud
ed

cA

RT
u

se

an
d

s
el

f-
re
p
or
te
d
a
d
h
er
en
ce
,
d

ur
at

io
n
o
f
an
ti

re
tr

ov
ir

al
t

h
er

ap
y

us
e,

cu

rr
en

t
C

D
4

co
un

t
<

20
0

ce
lls

/
m

m
3,

cu
rr
en

t
vi

ra
l
lo
ad

(u

n
d

et
ec

ta
b

le
,

< 10

,0
00

c
p

/m
l,

≥
10

,0
00
c
p
/
m

l),
a

n
d

n
ad

ir
C

D
4
co
un
t
< 20 0 ce lls

/m
m

3
ad

eq
ua
te
ly

re
p
or
te
d
.
In
s
um
m
ar
y,

d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
ad
eq
ua
te
ly

re
p
or
te
d
.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d

a
cc

ur
at

el
y

p
re
se
n
te
d
w
it

h
p

v
al
ue
s
re
p
or
te
d
.
N
o
co
n
fid
en
ce

in
te

rv
al
s
re
p
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

ad
eq
ua
te
.
H
ig
h
Ru
b
in
e
t
al
.,
20

16
Sa

m
p
le
r
ec
ru
it
ed

in
cl

ud
ed

3
8

H
IV

+

w
om

en
f

ro
m
t
h
e
W
om
en

’s

In
te

r-
A

g
en

cy
H

IV
S

tu
d

y
(W

IH
S)

to
e
xp
lo
re
t
h

e
n

eu
ro

b
io

lo
g

ic
al

fa

ct
or

s
co
n
tr

ib
ut

in
g
t
o

st
re

ss
–

re
la
te
d
m
em
or
y

im
p

ai
rm

en
t.

In
s
um
m
ar
y,
s
am
p
le
i
s
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
(
H
IV

-i
n

fe
ct
ed

lo
w

s
tr

es
s)

i
n
cl
ud
ed
,
ea
si
ly

id
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro
up
c
le
ar
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

cl
ea
r.
M
ai
n
e
xp
os
ur
e:
P
er
ce
iv
ed
S
tr
es
s
Sc
al
e
(P
SS
-1
0)
a
n

d
P

TS
D

C

h
ec

kl
is

t

–
C

iv
ili

an
V

er
si

o

n
(

PC
L-

C
)
M
ai
n
o
ut
co
m

e:
T

h
e

H
op

ki
n

s
V

er
b

al
l

ea
rn
in
g
T
es

t
(H

V
LT

)
to

m
ea

su
re
v
er
b
al

le
ar
n
in
g
a
n
d
m
em
or

y.

A
d
d
it
io
n

al
c

og
n
it
iv

e
d

om
ai
n

m
ea
su
re
d
i
n
cl
ud
ed

:
at

te
n
ti
on

/
co

n
ce
n
tr
at
io
n
,
ex
ec
ut
iv
e
fu
n
ct
io
n

s,
p

sy
ch
om
ot
or
s
p
ee
d

an
d
v
er
b
al
f
lu
en
cy

sM

RI
t

o
m

ea
su

re
b

ra
in

vo
lu
m

et
ri

c
ab

n
or

m
al

it
ie

s
In
s
um
m
ar
y,
q
ua
lit
y
of

ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua
te
.
C
on
fo
un
d
er
s
(a

g
e)

a
d
eq
ua
te
ly

re
p
or
te
d
.
In
s
um
m
ar
y,

d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

un
cl

ea
r.

H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d
a
cc
ur
at
el
y
p
re
se
n
te
d
w
it
h
p
v
al
ue
s
re
p
or
te
d
.
N
o
co
n
fid
en
ce

in
te
rv
al
s
re
p
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

ad
eq
ua
te
.
H
ig
h
(C
on
ti
nu
ed
)

8 G. SPIES ET AL.

Ta
b
le
2
.
(C
on
ti
n
ue
d
).
St
ud
y
Sa
m
p
le
C
ou
n
tr
y
C
on
tr
ol
/c
om
p
ar
is
on
g
ro
up
Ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
s
D
is
to
rt
in
g
i
n
flu
en
ce
s
Re
p
or
ti
n
g
o
f
d
at
a
O
ve
ra
ll
st
ud
y
q
ua
lit
y
Ru
b
in
e
t
al
.,
20

17
Sa

m
p
le
r
ec
ru
it
ed
i
n
cl
ud

ed
6

46

H
IV
+
a
n
d

3
00

H
IV

–
w

om
en

fr
om
t
h
e
W
om
en
’s
I
n
te
r-

A
g
en
cy
H
IV
S
tu
d

y
to

e
xp
lo
re

th
e
as
so
ci
at
io
n
b
et
w
ee
n

p
er
ce
iv
ed
s
tr
es
s
an
d

P
TS

D
i

n

le
ar
n
in
g
,
m
em
or
y,
a
n
d
f
lu
en

cy

im
p
ai
rm
en
ts
.
In
s
um
m
ar
y,

sa
m
p
le
i
s
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
(

30
0

H
IV
–
w
om
en
)
in
cl
ud
ed
,
ea
si
ly

id
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro
up
c
le
ar
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

cl
ea
r.
M
ai

n
ex

po
su

re
: P

er
ce
iv
ed
S
tr
es
s

Sc
al

e

(P

SS
-1

0)
a

nd
P

TS
D

Ch

ec
kl

is
t-

Ci
vi

lia
n

ve
rs

io
n

(P

CL
-C

)
M

ai
n

ou
tc

om
e:

le
ar

ni
ng

a
nd

m
em
or

y.
O

th
er

d
om

ai
ns

m
ea
su

re
in

cl
ud

ed
a

tt
en

tio
n,

ex
ec
ut
iv
e

fu
nc

tio
n,
f
lu
en
cy

,
ps

yc
ho

m
ot
or
s

pe
ed

, a
nd

m
ot
or

sk
ill

s
In
s
um
m
ar
y,
q
ua
lit
y
of
e
xp
os
ur
e/

ou
tc
om
e
m
ea
su
re
m
en
t
is

ad
eq
ua
te
.
C
on
fo
un
d
er

s
(w

av
e
of
e

n
ro

lm
en

t,

se
lf-

re
p
or
te
d
a
n
n
ua
l
h
ou
se
h
ol
d
i
n
co
m

e,
h

ar
m

fu
l

al
co

h
ol
u
se

,
sm

ok
in

g
s

ta
tu
s,

m
ar
iju
an

a

us

e,
a
n
d

c
ra

ck
,

co
ca

in
e,

a
n

d
/o

r
h

er
oi

n
u

se
,

p
sy
ch
ia
tr
ic
a
n

ti
d

ep
re

ss
an

t/

an
ti
p
sy

ch
ot

ic
m

ed
ic
at
io
n
,
an
d

p
os
it
iv
e

H
ep

at
it

is
C

a
n

ti
b

od
y)

ad
eq
ua
te
ly
r
ep
or
te
d
.
In

su
m
m
ar
y,
d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

un
cl
ea
r.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d
a
cc
ur
at
el
y
p
re
se
n
te
d
w
it
h
p
v
al
ue
s
re
p
or
te
d
.
N
o
co
n
fid
en
ce

in
te
rv
al
s
re
p
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

ad
eq
ua
te
.
H
ig
h

W
at

so
n

e
t
al
.,

20
18

Sa
m

p
le
r
ec

ru
it

ed
i
n
cl
ud
ed

1
22

H

IV
+

a
n

d
9

5
H

IV
–
in
d
iv
id
ua
ls

fr
om
t
h
e

M
ul

ti
-D

im
en

si
on

al

Su
cc

es
sf

ul
A

g
ei

n
g

a
m

on
g

A
d
ul

ts
l

iv
in

g

w
it
h
H

IV
s

tu
d
y
to
e
xp
lo
re
t
h

e
ef

fe
ct
s
of
t
ra

um
a,

e
co

n
om

ic

h
ar

d
sh

ip
,

an
d
s
tr
es
s

on
N

C
I
an
d

e
ve

ry
d

ay
f

un
ct

io
n
.
In

su
m
m
ar
y,
s
am
p
le
i
s
cl
ea
r.
U
SA
C
om
p
ar
is
on
g
ro
up
(

95
H

IV
–
in
d
iv
id
ua
ls
)
in
cl
ud
ed
,
ea
si
ly

id
en
ti
fia
b
le
,
so
ur
ce
o
f
co
m
p
ar
is
on
g
ro
up
c
le
ar
.
In

su
m
m
ar
y,
c
om
p
ar
is
on
g
ro
up
is

cl
ea
r.
M
ai
n
e
xp
os
ur
e:
s
el
f-
re
p
or

t
W

om
en

’s
H

ea
lt
h
I
n
it

ia
ti

ve

(

W
H

I)
Li

fe
E

ve
n

ts
S

ca
le

M
ai
n
o
ut
co
m

e:
e

xe
cu

ti
ve

fu
n
ct
io
n
in
g
,
le
ar
n
in
g
,
m
em
or

y
(d

el
ay

ed
r

ec
al

l),
w

or
ki
n
g

m
em
or

y,
v

er
b
al
f
lu
en
cy
,
sp
ee
d

of
i

n
fo
rm
at
io
n

p
ro

ce
ss
in
g

, a
n

d

co
m
p
le

x
m

ot
or

s
ki

lls
.

In
s
um
m
ar
y,
q
ua
lit
y
of

ex
p
os
ur
e/
ou
tc
om
e
m
ea
su
re
m
en
t
is
a
d
eq
ua
te
.
C
on
fo
un
d
er
s
(g

en
d

er
,
ye
ar
s
of

ed
uc
at
io
n

,
lif

et
im

e
M

D
D

,
lif
et
im

e
su

b
st

an
ce
u
se

d
is
or
d
er
,
lif
et
im
e
al

co
h

ol
u
se
,

ra
ce

/e
th

n
ic
it
y
an
d
l
ife

ti
m

e
ca

n
n

ab
is

u
se
)
ad
eq
ua
te
ly

re
p
or
te
d
.
In
s
um
m
ar
y,

d
es
cr
ip
ti
on
o
f
in
flu
en
ce
s
ar
e
ad
eq
ua
te
.
M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

un
cl
ea
r.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d
a
cc
ur
at
el
y
p
re
se
n
te
d
w
it
h
p
v
al
ue
s
re
p
or
te
d
.
N
o
co
n
fid
en
ce

in
te
rv
al
s
re
p
or
te
d
.
In

su
m
m
ar
y,
r
ep
or
ti
n
g
o
f
d
at
a
is

ad
eq
ua
te
.
H
ig
h
W
om

er
sl

ey

et
a
l.,
2

01
8

Sa
m
p
le
r
ec
ru
it

ed
1

28
H

IV

p
os
it
iv
e
w
om
en

b
et

w
ee

n
1

8
an

d
6

5
ye

ar
s

of
a

g
e,

w
h

o
w

er
e
flu
en

t
in

is
iX

h
os

a,
E

n
g

lis
h

a
n

d
/

or
A

fr
ik

aa
n

s
an
d
w
it
h

a
m
in

im
um

li
te

ra
cy

le
ve

l o
f

5t
h

g
ra

d
e

ed
uc

at
io
n
.
So
ut
h
A
fr
ic

a
Th

e
st

ud
y

ex
am

i

n
es

h
ow

ch
ild
h
oo
d
t
ra

um
a

in
te

ra
ct

s
w

it
h

A
p

oE
in

r
el

at
io
n
t

o
N

C
I s

o
co
m
p
ar
is

on
i

s
d

et
er

m
in
ed
b

y
ch

ild
h
oo
d
t
ra
um
a
ve
rs

us
n

o
ch
ild
h
oo
d
t
ra
um
a.
M
ai
n
e
xp
os
ur
e:
c
h
ild
h
oo
d

tr
au
m
a
m
ea
su
re
d
b
y
th
e
ch
ild
h
oo
d
t
ra
um
a
q
ue
st
io
n
n
ai
re
(
C
TQ
)
m
ea
su
ri
n
g

28
e
xp
er
ie
n
ce
s
of
c
h
ild
h
oo
d

tr
au
m
a.

M
ai
n
o
ut
co
m
e:
N
eu
ro
co
g
n
it
iv
e
fu
n
ct
io
n
a
s
m
ea
su
re
d
b
y
th
e
N
eu
ro
b
eh
av
io
ra
l
Re
se
ar

ch
C

en
tr

e
(H

N
RC

)
b

at
te

ry
.

Th
is

b
at

te
ry
c
on
si
st
s
of

1
7

te
st

s
an

d
u

se
s

cu
lt

ur
al
ly

ad
ap

te
d
f
or
t
h

e
So

ut
h
A
fr

ic
an

co
n
te

xt
.

Se
ve

n
d

om
ai
n
s
of

co
g
n
it
iv
e
fu
n
ct
io

n
:

m
ot
or

ab
ili

ty
,
p
ro
ce
ss
in
g
s
p
ee
d
,
ve
rb
al
f
lu
en
cy
,
le
ar
n
in
g
,
d
el
ay
ed
r
ec
al

l,
at

te
n
ti
on
,
w
or
ki
n
g
m
em
or
y
an
d

ex
ec
ut
iv
e
fu
n
ct
io
n
w
er
e
ex
am

in
ed

.
C
on
fo
un
d
er
i
n
cl
ud

ed
g

lo
b

al

co
g
n
it
iv
e
sc
or

es
.

M
is
si
n
g
d
at
a
n
ot
r
ep
or
te
d
.
H
ow
ev
er
,
d
at
a
ar
e
cl
ea
rl
y
an
d

ac
cu
ra
te
ly
p
re
se
n
te
d
w
it
h

p
v
al
ue
s
re
p
or
te
d
.
H
ig
h

SA
Q

O
R

–
Sy

st
em
at
ic
A
p

p
ra

is
al

o
f

Q
ua

lit
y

in
O

b
se

rv
at

io
n

al
R

es
ea
rc
h
.

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 9

participants (e.g. both adults and children) were
found; therefore, data disaggregation was not neces-
sary. Collectively the fifteen studies included 5140
participants (3866 PLWHA), with the average age of
participants in most studies in the 30s and 40s. The
average age for fourteen of fifteen studies, (one study
reported the age range only) was 40 years. Eight were
all-female studies (Malan-Muller et al., 2013; Rubin
et al., 2017, 2015, 2016; Spies et al., 2016, 2012; Spies,
Fennema-Notestine, Cherner, & Seedat, 2017;
Womersley, Spies, Seedat, & Hemmings, 2019) and
two were all-male studies (Deiss et al., 2019; Pukay-
Martin, Cristiani, Saveanu, & Bornstein, 2003), with
the remainder (five studies) including both men and
women (Clark, Arce Renteria, Hegde, & Morgello,
2018; Clark et al., 2012; Kapetanovic et al., 2020; Lin
et al., 2011; Watson et al., 2019).

3.2.1. Study design
There were ten case-control studies (Clark et al., 2012;
Kapetanovic et al., 2020; Malan-Muller et al., 2013;
Pukay-Martin et al., 2003; Rubin et al., 2017, 2015;
Spies et al., 2016, 2012, 2017; Watson et al., 2019) and
five cross-sectional studies (Clark et al., 2018; Deiss
et al., 2019; Lin et al., 2011; Rubin et al., 2016;
Womersley et al., 2019). Eight studies were longitudinal
in design (Malan-Muller et al., 2013; Rubin et al., 2017,
2015, 2016; Spies et al., 2016, 2012, 2017; Womersley
et al., 2019) but only one of these presented longitudinal
data examining the relationship between childhood
trauma and HIV and change in cognition over time
(Spies et al., 2017) and one examined the relationship
between perceived stress and PTSD and NCI over time
(Rubin et al., 2017).

3.2.2. Study location
All fifteen studies were limited to two countries; ten
studies (67%) were from the USA (high-income
country) and five (33%) from South Africa (middle-
income country). No studies from low-income coun-
tries/regions met entry criteria.

3.2.3. Trauma measurement
The majority of studies (n = 12) measured trauma using
self-report questionnaires, with the exception of one
study which included self-reported medical histories
of Traumatic Brain Injury (TBI) to examine if
PLWHA with a clear history of TBI had greater risk of
NCI compared to those without TBI (Lin et al., 2011),
one study which included the Composite International
Diagnostic Interview (CIDI) to assess for a DSM-IV
diagnosis of PTSD and other mental health conditions
(Deiss et al., 2019), and another study which included
the Client Diagnostic Questionnaire (CDQ) to assess
for interpersonal trauma (IPT history) and PTSD.
Overall, current or past PTSD was assessed for in four

studies (Deiss et al., 2019; Kapetanovic et al., 2020;
Rubin et al., 2017, 2016).

3.2.4. Trauma exposure
Of the fifteen studies, seven studies (Clark et al., 2018,
2012; Malan-Muller et al., 2013; Spies et al., 2016,
2012, 2017; Womersley et al., 2019) examined early
life trauma only (experienced before the age of 18)
and seven studies (Deiss et al., 2019; Lin et al., 2011;
Pukay-Martin et al., 2003; Rubin et al., 2017, 2015,
2016; Watson et al., 2019) examined adult-onset
trauma only (experienced after age 18). Only one
study considered both early life and adult-onset
trauma (Kapetanovic et al., 2020) but no studies
sought to separate out the effects of early-onset versus
adult-onset trauma. One study from the USA
assessed the combined effects of PTEs (through
a perceived stress measure) and PTSD (PTEs +
PTSD) on NCI (Rubin et al., 2017). Across the stu-
dies, sample sizes ranged from 38 to 1003 PLWHA.
In ten of the fifteen studies (67%) a HIV-negative
control group was included, with HIV-negative con-
trol groups ranging in size from 45 to 496. The
remaining five studies did not include a HIV-negative
control group but included a comparison group com-
prised of PLWHA.

3.3. Findings on trauma and NCI

All fifteen included studies that assessed NCI in PLWHA
demonstrated a significant association with trauma expo-
sure (Table 5). Five studies controlled for additional
confounding variables such as depression, substance
use, and antidepressant use (Clark et al., 2018, 2012;
Pukay-Martin et al., 2003; Rubin et al., 2015; Watson
et al., 2019). Among the fifteen studies measuring both
trauma and NCI in PLWHA, three studies (conducted in
the USA) also estimated the prevalence of NCI (Deiss
et al., 2019; Kapetanovic et al., 2020; Watson et al., 2019).
The prevalence of NCI in these studies was 19% (Deiss
et al., 2019), 39% (Watson et al., 2019), and 27%
(Kapetanovic et al., 2020) respectively. One study that
recruited participants from the CNS HIV Antiretroviral
Therapy Effects Research (CHARTER) estimated NCI
using a global deficit score (Lin et al., 2011). The remain-
ing studies reported associations between exposure to
PTEs and specific cognitive domains (Table 5).

3.3.1. Cognitive domains
In these studies, a cognitive domain was assessed
using more than one test. Five studies demonstrated
an association between PTEs and SLEs and executive
function deficits (Lin et al., 2011; Pukay-Martin et al.,
2003; Spies et al., 2016, 2017; Watson et al., 2019).
Regarding verbal fluency, five studies demonstrated
an association with PTEs and SLEs (Malan-Muller

10 G. SPIES ET AL.

Ta
b

le
3

.
St

ud
ie

s
in
cl
ud
ed
i
n
t
h
e
co
m
p
re

h
en

si
ve

r
ev
ie
w
o
f
th
e

lit
er

at
ur

e
on

t
h
e
as
so
ci
at
io
n
b
et
w
ee
n

p
ot

en
ti

al
ly

t
ra

um
at

ic
o

r
st

re
ss
fu
l

ev
en

ts
a

n
d

c
og

n
it
iv
e
fu
n
ct
io
n
a
m
on

g
P

LW
H

A
.

St
ud

y
To

ta
l

PL
W

H
A

n
H
IV

n
H
IV

+
c

om
p
ar
is
on

g
ro

up
C

ou
n
tr
y
M
al

e
(%

)
A

g
e

ra
n
g
e
or

m
ea

n
Tr

au
m

a
or

s
tr
es
s
m
ea
su
re

M
en

ta
l
h
ea

lt
h

a
ss

es
sm

en
t

Re
la

ti
on
sh
ip
b
et
w
ee
n
t
ra
um
a
an
d
N
C
I

in

PL
W
H
A
C
la
rk
e
t
al
.,
20

12
96

49
47

–
U

SA
58

45
EL

SQ
–

*

*

C

la
rk

e
t
al
.,
20
18

44
44

–
Ye

s
U

SA
59

45
EL
SQ
–

*
D

ei
ss

e
t
al
.,

20
19

18
9

18
9
–
Ye
s
U

SA
10

0
28

–4
4

–
C

ID
I

(c
ur

re
n

t
an

d
li

fe
ti

m
e
h
is
to
ry
o
f
al
co
h
ol

us
e
d
is
or
d
er
,
d
ep
re

ss
io

n
,
an
d
P
TS

D
)

**

Ka
p
et
an
ov
ic
e
t
al
.,

20
20

18
7

13
7

50
Ye

s
U

SA
63

50
C

D
Q

t
ra
um
a

in
ve

n
to

ry
C

D
Q
*
Li
n
e
t
al
.,
20
11

96
4

96
4
–
Ye
s
U

SA
77

44
Se

lf-
re
p
or
te
d

m
ed

ic
al

h
is
to

ry
–

*
M
al
an
-M
ul
le
r
et
a
l.,

20

13
83

45
–

So
ut
h

A
fr
ic
a

0
30

C
TQ

-S
F

–
**

Sp
ie
s
et
a
l.,
2
01

2
12

8
83

47
–

So
ut
h

A
fr
ic
a
0
30
C
TQ
-S
F

–
*

Sp
ie
s
et
a
l.,
2
01

6
12

4
62

62
–

So
ut
h

A
fr
ic
a
0
30

;
18

–5
0

C
TQ
-S
F
–
*
Sp
ie
s
et
a
l.,
2
01

7
11

7
67

50
–

So
ut
h

A
fr
ic
a

0
33

;
18
–5
0
C
TQ
-S
F
–
**
Pu
ka
y-
M
ar
ti
n
e
t
al
.,
20

03
33

3
25

1
82

–
U
SA
10
0
33

PE
RI

L
ife

E
ve

n
ts
S
ca

le
–

**
Ru
b
in
e
t
al
.,
20

15
14

99
10

03
49

6
–

U
SA

0
46

;
25

–8
7

–
PS

S-
10

**
*

Ru
b
in
e
t
al
.,
20

16
38

3

8
–

Ye
s

U
SA

0
44

;
27

–5
9

PC
L-

C
PS

S-
10

*
Ru

b
in

e
t
al
.,

20
17

96
4

64
6

30
0

–
U

SA
0

45
PC

L-
C

PS
S-

10
*

W
at
so
n
e
t
al
.,
20
19

21
7

12
2

95
–

U
SA

78
51

;
35

–6
5

W
H

I
Li

fe
E
ve
n
ts
S
ca
le
–
*
W
om
er
sl

ey
e

t
al
.,
20

19
12

8
12

8
–
Ye
s
So
ut
h

A
fr
ic
a
0
33

;
21

–5
0
C
TQ
-S
F
*

*
=

p
<

0
.0

5;
*

*
=
p
< 0 .0

1;
*

**
=

p
< 0 .0

01

C
D

Q
=

C
lie
n
t
D
ia
g
n
os
ti
c
Q
ue
st
io
n
n
ai

re
;

C
ID

I
=

C
om

p
os

it
e

In
te
rn
at
io
n

al
D

ia
g
n
os
ti
c
In
te
rv
ie

w
;

C
TQ

-S
F

=
C

h
ild
h
oo

d
T

ra
um

a
Q

ue
st
io
n
n
ai

re
-S

h
or

t
Fo

rm
;

EL
SQ

=
E

ar
ly
L
ife
S
tr
es
s

Q
ue

st
io

n
n

ai
re

;
PC

L-
C

=
P

TS
D

C
h

ec
kl
is
t-

C
iv

ili
an

V
er

si
on

;
PE

RI
=

P
sy
ch
ia
tr
ic

E
p

id
em
io
lo
g
y
Re
se
ar
ch
I
n
te
rv
ie
w

;
PS

S-
10
=
P
er
ce
iv
ed
S
tr
es
s
Sc
al

e;
W

H
I

=
W

om
en
’s
H
ea
lt
h
I
n
it
ia
ti

ve
L

ife
E

ve
n
ts
S
ca
le
.

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 11

Ta
b

le
4

.
H

IV
d

is
ea
se
p

ar
am

et
er
s
of
i
n
cl
ud

ed
s

tu
d

ie
s.

St
ud

y
PL

W
H

A

n
%

o
n

A

RV
s

M
ea

n
(

SD
)

d
ur

at
io
n
o

f
ill

n
es
s
in

y
ea

rs
M

ea
n

(
SD

)
C

D
4

T-
ce

ll
co

un
t

(c
el

ls
/m

m
3

)
M

ea
n
(
SD

)
H

IV
v

ir
al

l
oa

d
(

co
p

ie
s/

m

L)

C
la
rk
e
t
al
.,
20

12
49

84
H

ig
h

E
LS

(n

=
2

5)

13
.0

4
(7

.9
8)

Lo
w
E
LS

(n
=
2

4)

14
.4

6
(6

.6
0)

H
ig

h
E

LS

(n
=

2
5)

46

0.
92

(
25

7.
35

)
Lo
w
E
LS

(n
=
2
4)

56
0.

96
(

31
3.

51
)

H
ig
h
E
LS

(n
=
2
5)

20

U
n

d
et

ec
ta

b
le

(
<

75
)

4
H

ig
h

(
>

40
0)

Lo
w
E
LS

(n
=
2
4)

16
U

n
d
et
ec
ta
b

le
(

< 75

)
6

H
ig

h
(

>
40

0

)
C

la
rk
e
t
al
.,
20
18

44
10

0
H

ig
h
E
LS

(n
=
2

6)

16
.0

0
(7

.2
2)

Lo
w
E
LS

(n

=
1

8)

14
.5

6
(6

.6
4)

H
ig
h
E
LS

60
8.

85

(2
72

.1
9)

Lo
w
E
LS

59

4.
28

(
29

8.
40

)
H

ig
h
E
LS

65

%
<

50
Lo

w
E

LS

53
%

< 50

D
ei
ss
e
t
al
.,
20

19
18

9
66

5
(2

.1
1

m
ed

ia
n

I
Q

R)
54

2
(4

13
,6

97
m

ed
ia

n
I

Q
R)

48
c

op
ie

s/
m

L
(4

8,
32

82
m

ed
ia
n
I
Q
R)
Ka
p
et
an
ov
ic
e
t
al
.,
20
20
13
7

10
0

IP
T+

(n

=
1

02
)

17
.9

(
9.

0)

IP
T-

(n
=

3
5)

14

.1
(

9.
7)

IP
T+

(n
=
1
02
)

N
ad

ir
C
D
4

20
1.

5
(1

77
.5

)
IP
T-
(n
=
3
5)

N

ad
ir

C
D

4
21

7.
9

(1
69

.1
)

IP
T+

(n
=
1
02
)

A
ll

w
er

e
av

ir
em

ic
(

< 40

c
op

ie
s/
m
L)
IP
T-
(n
=
3
5)

A

ll
w

er
e

av
ir

em
ic

(
<

40
c

op
ie
s/
m
L)
Li
n
e
t
al
.,
20
11
96
4

66
n

r
TB

I+

(n
=

1
10

)
N

ad
ir
C
D
4
21

3.
3

(1
88

.5
)

TB
I-

(n
=

5
90

)
N
ad
ir
C
D
4
21

9.
9

(2
10

.7
)

TB
I+

(n
=
1

10
)

65
.1

%
d

et
ec
ta
b
le
(

b
as

ed
o

n
l

im
it

o
f
d
et

ec
ti

on

of
5

0)
TB
I-
(n
=
5
90

)

70

.1
%

d
et

ec
ta
b
le

(
b

as
ed

o
n

l
im

it
o

f
d
et
ec
ti
on

of

5
0)

M
al
an
-M
ul
le
r
et
a
l.,

20
13
83

16
n

r
n
r
n
r
Sp
ie
s
et
a
l.,
2
01

2
83

16
n

r
40

5
(2

59
.8

0)
10

5,
16

9.
5

(4
07

,4
59

.5
)
Sp
ie
s
et
a
l.,
2
01

6
62

49
n

r
43

1.
20

(
26

7.
75

)
16

,1
72

.0
8

(4
39

,0
37

.7
6)

Sp
ie
s
et
a
l.,
2
01
7
67

81
n

r
C

T+
(

n
=

5
3)

44

2.
04

(
23

4.
88

)
C

T-
(

n
=

1
4)

41

3.
57

(
28

5.
02

)
C
T+
(
n
=
5
3)

61

,5
05

.0
4

(1
74

,0
46

.0
9)

C
T-

(
n

=
1
4)

81
,1

69
.9

3
(1

85
,2

06
.4

4)
Pu

ka
y-

M
ar

ti
n
e
t
al
.,

20
03

25
1

n
r

n
r

49
.1

%
˃

40
0m

m

3
;

27
.5

%
≥

2
00
m
m

3
a

n
d

≤
4

00

m
m
3
;

23
.4

%
˂

20
0m

m
3
n
r
Ru
b
in
e
t
al
.,
20

15
10

03
76

n
r
H
ig

h
s

tr
es

s
(n

=
3

84
)

N
ad

ir
2

08
(

15
7)

44
%
˃

50
0

40
%

≥
20

0
an

d
<

50
0

16
%

< 20

0
Lo
w
s
tr
es
s
(n
=

6
19

)
N
ad
ir

2
18

(
15

8)

55
%

˃
50

0
34

%
≥
20
0
an
d
< 50

0
11

%
<

20
0

H
ig
h
s
tr
es
s
(n
=
3
84
)

47
%

u
n
d
et
ec
ta
b
le

37

%
<

10
,0

00

16
%

≥
10

,0
00
Lo
w
s
tr
es
s
(n
=
6
19

)
55

%
u

n
d
et
ec
ta
b

le

32
%

< 10

,0
00

13

%
≥
10
,0
00
Ru
b
in
e
t
al
.,
20
16
38

79
n

r
H

ig
h
s
tr
es
s
(n
=

1
8)

N
ad
ir

2
87

(
17

7)

28
%

˃
50

0
55

%
≥
20
0
an
d
< 50

0
17

%
< 20 0 Lo w s tr es s (n =

2
0)

N
ad
ir

3
25

(
22

6)

74
%

˃
50

0
16

%
≥
20
0
an
d
< 50

0

10

%
< 20 0 H ig h s tr es s (n = 1 8)

44
%

u
n
d
et
ec
ta
b
le

28

%
< 10 ,0 00 28 % ≥ 10 ,0 00 Lo w s tr es s (n = 2 0) 70 % u n d et ec ta b le

20
%

< 10 ,0 00 10 % ≥ 10 ,0 00 Ru b in e t al ., 20

17
64

6
77

n
r

52
1

(4
01

m
ed
ia
n
I
Q

R)
29

%
<

50
0

W
at
so
n
e
t
al
.,
20
18
12
2

96
17

.1
(

8.
7)

63
3(

42
5,

8
51

m
ed
ia
n
I
Q

R)
6.

6%
d

et
ec
ta
b

le
W

om
er

sl
ey

e
t
al
.,
20
18

12
8

46
n

r
n
r
n
r

n
r

=
n

ot
r

ep
or
te
d

;
EL

S
=

e
ar

ly
l

ife
s

tr
es

s,
T

BI
=

t
ra
um
at

ic
b

ra
in
i
n

ju
ry

;
C

T
=

c
h
ild
h
oo
d
t
ra

um
a;

I
Q

R
=

i
n

te
rq

ua
rt

ile
r

an
g

e

12 G. SPIES ET AL.

et al., 2013; Pukay-Martin et al., 2003; Rubin et al.,
2017; Spies et al., 2016, 2017; Watson et al., 2019).
For attention/working memory, seven studies
demonstrated an association with trauma (Clark
et al., 2018; Kapetanovic et al., 2020; Lin et al., 2011;
Pukay-Martin et al., 2003; Spies et al., 2016; Watson
et al., 2019; Womersley et al., 2019); four studies
demonstrated an association with processing speed,
(Clark et al., 2012; Pukay-Martin et al., 2003; Spies
et al., 2016, 2012); six studies demonstrated an asso-
ciation with learning (Rubin et al., 2017, 2015, 2016;
Spies et al., 2016, 2012; Watson et al., 2019) while
three studies demonstrated an association with motor
function (Pukay-Martin et al., 2003; Spies et al., 2016;
Womersley et al., 2019). Two studies (Deiss et al.,
2019; Pukay-Martin et al., 2003) reported an associa-
tion between global NCI and trauma (i.e. not distin-
guished by cognitive domain).

3.3.2. Associations between PTEs and SLEs and NCI
Across all fifteen studies, there was consistent evidence
for an association between PTEs and SLEs and NCI. In
the nine studies that included a HIV-negative control
group, results consistently showed that trauma was
a significant risk factor for neurocognitive dysfunction
among PLWHA. For example, a recent study showed
that interpersonal trauma (IPT) exposure and HIV
infection have a synergistic effect on daily functioning
and cortical thickness in PLWHA (Kapetanovic et al.,
2020). Participants were classified as IPT+ if they
experienced any of the following: childhood physical
or sexual abuse, intimate partner violence as an adult,
physical, or sexual assault as an adult, direct combat,
seeing people harming one another in the family as
a child, or losing a child to death (Kapetanovic et al.,
2020). Of all fifteen studies included, this was the only
study that considered both early life and adult-onset
trauma, although this study did not separate out the
effects of early-onset versus adult-onset trauma on NCI

in PLWHA. This study found that attention/working
memory test performances were significantly worse in
PLWHA with IPT compared with PLWHA without
IPT and HIV-negative counterparts without IPT. The
HIV+ IPT+ group had significantly greater compro-
mise in activities of daily living and also had ‘impaired’
scores on a measure of subjective cognitive function
compared with the remaining three groups
(Kapetanovic et al., 2020). In another study, a case con-
trol design, stressful life events (47 events in seven
different categories: relationships and love, family, resi-
dence, crime and legal matters, finances, health, and
other), measured by the Psychiatric Epidemiology
Research Interview (PERI) Life Events Scale, were asso-
ciated with cognitive impairment only among male
PLWHA (Pukay-Martin et al., 2003). In a cohort
study, where participants were recruited across six
USA based sites, a significant HIV by stress interaction
was found for verbal memory among HIV-infected
women only, with high stress associated with lower
performance on verbal memory tests compared to
women with low stress (Rubin et al., 2015). In another
cohort study, assessing the combined effects of multiple
traumatic and stressful experiences among PLWHA,
a higher composite stress score (consisting of multiple
adverse experiences including trauma, economic hard-
ship, and stress) was associated with poorer executive
function, learning, working memory, and greater
decline in activities of daily living, even after controlling
for relevant demographic, psychiatric, substance use,
and HIV disease covariates. On their own, individual
stress composite scores did not predict these outcomes,
suggesting that the accumulation of adverse experiences
may additively or synergistically harm cognitive health
(Watson et al., 2019). In the remaining five studies, that
included a comparator group of PLWHA, similar
results were demonstrated. In two studies, one a USA
based cross-sectional study design and the other a USA
based MRI study, HIV-positive participants were

Table 5. Relationship between potentially traumatic or stressful life events and cognitive function among PLWHA.
Cognitive domains

Study PLWHA n On ARVs (%) EF FLU ATT/WM PS LRN/MEM Motor Global NCI

Clark et al., 2012 49 84 **
Clark et al., 2018 44 100 *
Deiss et al., 2019 189 66 **
Kapetanovic et al., 2020 137 100 *
Lin et al., 2011 964 66 * *
Malan-Muller et al., 2013 83 16 **
Spies et al., 2012 83 16 * *
Spies et al., 2016 62 49 * * * * * *
Spies et al., 2017 67 81 ** **
Pukay-Martin et al., 2003 251 nr * ** ** * * **
Rubin et al., 2015 1003 76 ***
Rubin et al., 2016 38 79 *
Rubin et al., 2017 646 77 * **
Watson et al., 2019 122 96 * * *
Womersley et al., 2019 128 46 * *

nr = not reported; * = p < 0.05; ** = p < 0.01; *** = p < 0.001 Cognitive domains: EF = executive function; FLU = Fluency; ATT/WM = attention and working memory; PS = processing speed; LRN/MEM = learning and

memory; VS = visuospatial and visuoconstructive functions; NCI = neurocognitive impairment.

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 13

categorised into low stress and high stress groups (Clark
et al., 2018; Rubin et al., 2016). In one of these studies by
Rubin et al., which compared HIV-infected women
with low stress PSS-10 scores in the lower two tertiles)
and HIV-infected women with high stress (PSS-10
scores in the top tertile), high stress and HIV seemed
to be associated with worse performance on measures
of verbal learning and memory and smaller brain
volumes (Rubin et al., 2016). Similarly in the other
study by Clark et al, participants with high early life
stress (ELS) (defined as an endorsement of three or
more adverse childhood events [ACEs]) exhibited
poorer working memory compared to those with low-
ELS (defined by endorsement of fewer than three
ACEs) (Clark et al., 2018). In studies that included
comparison groups with and without trauma (viz.
childhood trauma) (Womersley et al., 2019), PTSD
(Deiss et al., 2019), and TBI (Lin et al., 2011),
PLWHA who were trauma exposed consistently per-
formed more poorly than their counterparts without
trauma.

3.3.3. PTSD findings
Overall, current or past PTSD was assessed for in four
studies (Deiss et al., 2019; Kapetanovic et al., 2020;
Rubin et al., 2017, 2016). PTSD was not dealt with as
a mediator of trauma exposure and NCI but rather as
an independent predictor of NCI in one study (Deiss
et al., 2019). Although PLWHA were assessed for
PTSD, the independent effect of PTSD on NCI was
not reported in two of these studies (Kapetanovic
et al., 2020; Rubin et al., 2016). Instead, the effects
of high vs. low stress (viz. perceived stress) (Rubin
et al., 2016) and interpersonal trauma (IPT)
(Kapetanovic et al., 2020) on NCI were reported. In
the study including PTSD as an independent predic-
tor of NCI, lifetime history of PTSD was indepen-
dently associated with NCI in a cohort of
189 U.S. military men living with HIV (Deiss et al.,
2019). Individuals with a lifetime history of PTSD
were six times more likely to be diagnosed with
NCI than individuals without PTSD (Deiss et al.,
2019).

3.3.4. PTE exposure and PTSD findings
The fourth study reporting on current or past PTSD
sought to assess both the separate and interactive effects
of HIV and stress or PTSD on NCI (Rubin et al., 2017).
This longitudinal study from the USA assessed the com-
bined effects of PTEs (through a perceived stress mea-
sure) and PTSD on NCI (Grant, 2008), finding that
higher levels of stress (viz. perceived stress) and PTSD
symptoms, more so than depressive symptoms, may
contribute to different patterns of detrimental neurocog-
nitive performance among PLWHA, compared to lower
levels of stress and PTSD (Rubin et al., 2017). Across time
points, PTSD was associated with lower performance on

both learning and memory among women living with
HIV but not HIV-negative counterparts. This longitudi-
nal study provides evidence that elevated perceived stress
and PTSD symptoms in the context of HIV are linked to
alterations in verbal abilities (learning, memory, fluency)
over time (Rubin et al., 2017).

4. Discussion

4.1. Associations between PTEs and SLEs and NCI

In this paper we present, to our knowledge, the most
comprehensive review of evidence of the relationship
between broad concepts of PTEs and NCI in PLWHA.
Despite the small evidence base (especially from low
and middle-income countries [LMIC]) and the metho-
dological limitations of the available studies, it is notable
that in all the studies included there was an association
between trauma and NCI in PLWHA. In order to
provide a robust body of evidence, the studies in the
present review include measures of different kinds of
adversity including exposure to PTEs (both early life
and adult-onset trauma), economic hardship (food
insecurity and low SES), perceived stress, and traumatic
brain injury. These findings of the relationship between
NCI and PTE have emerged in several previous studies
(Pavlovic, Pekic, Stojanovic, & Popovic, 2019).
Perceived stress may include stressful life events con-
sidered as potentially traumatic events. Stressful life
events are thought to increase risk for disease when
one perceives that the demands these events impose
tax or exceed a person’s adaptive capacity (Cohen,
Janicki-Deverts, & Miller, 2007). Exposures to chronic
stress are considered the most toxic because they are
most likely to result in long-term or permanent changes
in the emotional, physiological, and behavioural
responses that influence susceptibility to and course of
disease (McEwen, 2006). A high level of perceived stress
has been shown to be accompanied by symptoms of
depression and/or anxiety (Wiegner, Hange,
Björkelund, & Ahlborg, 2015). Irrespective of the type
of exposure, the results show compelling evidence for
the contributory role of trauma in NCI in PLWHA. All
15 included studies that assessed NCI in PLWHA
demonstrated a significant association with PTEs and
SLE and/or PTSD as broadly defined in our study.
Three studies reported prevalences of 19%, 39%, and
27% of NCI (Deiss et al., 2019; Kapetanovic et al., 2020;
Watson et al., 2019). All 15 included studies demon-
strated an association of PTEs and SLEs and/or PTSD
and various cognitive domains, including executive
function (Lin et al., 2011; Pukay-Martin et al., 2003;
Spies et al., 2016, 2017; Watson et al., 2019), attention/
working memory (Clark et al., 2018; Kapetanovic et al.,
2020; Lin et al., 2011; Pukay-Martin et al., 2003; Spies
et al., 2016; Watson et al., 2019; Womersley et al., 2019),
processing speed (Clark et al., 2012; Pukay-Martin et al.,

14 G. SPIES ET AL.

2003; Spies et al., 2016, 2012), verbal fluency (Malan-
Muller et al., 2013; Pukay-Martin et al., 2003; Rubin
et al., 2017; Spies et al., 2016, 2017), learning/memory
(Rubin et al., 2017, 2015, 2016; Spies et al., 2016, 2012;
Watson et al., 2019), and motor function (Pukay-
Martin et al., 2003; Spies et al., 2016; Womersley et al.,
2019), with some studies showing an association with
global cognitive function among PLWHA (Deiss et al.,
2019; Pukay-Martin et al., 2003).

4.2. Methodological considerations of included
studies

Potential confounders were adequately controlled for in
the majority of studies including (Clark et al., 2018;
Deiss et al., 2019; Lin et al., 2011; Malan-Muller et al.,
2013; Pukay-Martin et al., 2003; Rubin et al., 2017, 2015,
2016; Spies et al., 2016, 2012; Watson et al., 2019).
A recent systematic review by Rubin and Maki found
that depression contributed to impairments particularly
in the domains of executive function, processing speed,
learning, and motor function (Rubin & Maki, 2019b),
highlighting the importance for studies to adequately
consider the effects of depression on NCI or control for
these effects. In some studies included in this review,
depression was controlled for or included in analyses to
determine the independent effects thereof on NCI
(Clark et al., 2018, 2012; Deiss et al., 2019; Lin et al.,
2011; Pukay-Martin et al., 2003; Rubin et al., 2017,
2015). The majority of studies were cross-sectional
and as such, a causal relationship between trauma and
NCI cannot be inferred. However, in the longitudinal
study by Spies et al., the effects of HIV and childhood
trauma on NCI were sustained at 12-month follow-up
despite better ART uptake and improved HIV disease
status (Spies et al., 2017). In the era of effective ART and
increased virologic suppression, the experience of
trauma over the life course, the occurrence of stressors,
or PTSD among individuals living with HIV infection
may hold more relevance in the development of NCI
than was previously recognised. Future longitudinal
studies are needed. Across studies there were discrepan-
cies in the way trauma, stressors and adversity were
defined and measured. Few HIV-trauma-NCI studies
included mental health measures (e.g. PTSD, depres-
sion etc.) and those that did mostly included self-report
screeners of PTSD and depression. Only one study
included a structured diagnostic interview to assess
PTSD and depression (Deiss et al., 2019). The majority
of studies either assessed early-onset or adult-onset
trauma. Only one study considered both (Kapetanovic
et al., 2020) but no studies sought to separate out the
effects of early-onset versus adult-onset trauma. Across
studies there were also discrepancies in the measure-
ment of the severity of exposures and in the timing of
trauma in relation to NCI. There was a lack of consis-
tency in instrumentation used, both for the assessment

of trauma exposure and NCI. Moreover, only three
studies estimated the prevalence of NCI based on avail-
able norms (Deiss et al., 2019; Kapetanovic et al., 2020;
Watson et al., 2019), overwhelmingly studies reported
associations between trauma and NCI in the absence of
neuropsychological norms (Malan-Muller et al., 2013;
Spies et al., 2016, 2012, 2017; Womersley et al., 2019).
Across studies, there were inconsistencies in how HIV-
related clinical characteristics such as HIV RNA, use of
ART, illness stage etc. were dealt with. Moreover, the
characterisation of participants across studies in terms
of parameters such as gender, age, and HIV co-morbid-
ities was varied. To our knowledge, there are no studies
from Europe, South America, or Asia, highlighting the
lack of geographical diversity. Finally, more studies are
needed to parse out the effects of HIV and PTEs and
SLEs on NCI from the effects of HIV and mental health
disorders (e.g. PTSD) on NCI.

4.3. Clinical implications

The experience of trauma over the life course, the
occurrence of highly stressful events, and PTSD
deserve more scrutiny in the clinical assessment
and management of PLWHA as they may signal
the presence of NCI, more than was previously
recognised. The findings of the present review,
therefore, highlight the need for PTE and SLE
screening and detection and implementation of sec-
ondary prevention strategies in PLWHA who have
PTEs and SLEs. PTEs and SLEs and PTSD symptoms
are treatable targets and early intervention may serve
to improve cognitive abilities in PLWHA. Trauma-
focused interventions in youth and adults, that
include resilience building and coping strengthening
elements, may be beneficial in mitigating the nega-
tive psychological sequelae of PTEs and SLEs. The
integration of trauma-focused interventions in pri-
mary HIV care aimed at improving cognitive and
HIV disease outcomes, as well as emotional well-
being and quality of life, among PLWHA, warrants
investigation.

4.4. Implications for research

The quality analysis of included studies in this review
highlights the need for future studies to report their
methodology and findings in a standardised way to
ultimately improve methodological rigour in observa-
tional epidemiological studies. Future research should
aim to unpack the mechanisms underlying the associa-
tion between trauma and NCI among PLWHA.
Prospective studies in a range of LMIC and controlling
for different types of trauma are needed. Measurement
of the severity of exposure and the timing of trauma in
relation to NCI should be considered. Due to the cross-
sectional nature of these studies, it is not possible to

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 15

determine the temporal pattern of these changes.
Future longitudinal studies that track the course of
neurocognitive changes are needed to provide more
clarity. Such studies have the potential to provide
greater insights into potential targets for therapeutic
interventions.

5. Limitations

5.1. Limitations of reviewed studies

First, the use of self-report screeners of PTSD and
other psychiatric diagnoses rather than a structured
diagnostic interview in the majority of studies may
overestimate the number of individuals meeting cri-
teria for a clinical diagnosis. Second, the heterogene-
ity of trauma measures and variations in sample size,
sampling strategy, and study design, all of which may
have impacted on the results reported herein. Third,
publication bias, namely a tendency for journals to
publish positive findings, may overestimate the
strength or consistency of the association between
trauma and NCI.

5.2. Limitations of this study

In our study, trauma was broadly defined as poten-
tially traumatic events (PTEs) and stressful live events
(SLEs) during childhood and/or adulthood. Given the
paucity of studies in this area, both studies of poten-
tially TEs (PTEs) and stressful life events (SLEs) were
included. For example, studies assessing high levels of
stress (viz. perceived stress) in the context of everyday
stressors were also included. Measuring trauma expo-
sure without PTSD means that these events are not
DSM Criterion A events that cause PTSD and there-
fore should be considered potentially traumatic
events. In our review, studies that assessed for poten-
tially traumatic and stressful life events were
included. This should therefore be taken into account
when interpreting the findings of these studies. The
reliability and validity of the SAQOR tool need to be
established. As with other quality assessment tools,
the potential impact of differential weighting of the
component domains of the SAQOR tool should also
be determined. In our use of this instrument, each of
the five component domains (sample, control, out-
come/exposure, follow-up, distorting influences,
reporting of data) was given equal weighting in the
overall quality level assigned. Variations in weighting
may be appropriate depending upon the research
question.

6. Conclusion

The results of the included studies underscore the
importance of considering trauma, PTSD, and other

common mental health outcomes when conducting
research on cognition in PLWHA, as well as when
making a diagnosis of HIV associated Neurocognitive
Disorders (HAND). In particular, these studies high-
light the need for trauma screening over the life
course in PLWHA. Stress reduction and post-trau-
matic stress prevention and treatment programmes
could help improve and maintain cognitive function
in PLWHA and in turn may contribute to optimal
HIV treatment adherence, viral suppression and
improved quality of life.

Disclosure statement

No potential conflict of interest was reported by the
authors.

Funding

This work is supported by the South African Research
Chair in PTSD awarded to S Seedat and hosted by
Stellenbosch University, funded by the DSI and adminis-
tered by NRF and the Faculty of Medicine and Health
Sciences, Stellenbosch University (Deputy Dean’s strategic
fund for postdoctoral fellows and NRF postdoctoral fellow-
ship). Additional research support was provided by a
CFAR grant awarded to S Seedat [P30-AI036214]. GS has
received incentive funding for rated researchers from the
NRF (grant number 119880). SM has received funding
support from the Claude Leon Foundation, a self-initiated
research fellowship from the SA MRC and incentive fund-
ing for rated researchers from the NRF (grant number
119375).

ORCID

G. Spies http://orcid.org/0000-0003-0853-2813
S. Seedat http://orcid.org/0000-0002-5118-786X

References

Alford, K., & Vera, J. H. (2018, June 2). Cognitive
Impairment in people living with HIV in the ART era:
A review. British Medical Bulletin [Internet], 127(1),
55–68.

Ambrosius, B., Gold, R., Chan, A., & Faer, S. (2019, May 1).
Antineuroinflammatory drugs in HIV-associated neuro-
cognitive disorders as potential therapy. Neurology –
Neuroimmunology Neuroinflammation [Internet], 6(3),
e551. Retrieved from http://nn.neurology.org/content/6/
3/e551.abstract

Antinori, A., Arendt, G., Becker, J. T., Brew, B. J., Byrd, D. A.,
Cherner, M., & Wojna, V. E. (2007, October). Updated
research nosology for HIV-associated neurocognitive
disorders. Neurology, 69(18), 1789–1799.

Aupperle, R. L., Melrose, A. J., Stein, M. B., & Paulus, M. P.
(2012, February). Executive function and PTSD:
Disengaging from trauma. Neuropharmacology, 62(2),
686–694.

Brief, D. J., Bollinger, A. R., Vielhauer, M. J., Berger-
Greenstein, J. A., Morgan, E. E., Brady, S. M., … The
Hivaids treatment adherenc, F. (2004). Understanding

16 G. SPIES ET AL.

http://nn.neurology.org/content/6/3/e551.abstract

http://nn.neurology.org/content/6/3/e551.abstract

the interface of HIV, trauma, post-traumatic stress dis-
order, and substance use and its implications for health
outcomes. AIDS Care, 16(Suppl 1), S97–120.

Clark, U. S., Arce Renteria, M., Hegde, R. R., & Morgello, S.
(2018). Early life stress-related elevations in reaction
time variability are associated with brain volume reduc-
tions in HIV+ adults. Frontiers in Behavioral
Neuroscience, 12, 6.

Clark, U. S., Cohen, R. A., Sweet, L. H., Gongvatana, A.,
Devlin, K. N., Hana, G. N., & Tashima, K. T. (2012,
July). Effects of HIV and early life stress on amygdala
morphometry and neurocognitive function. Journal of
the International Neuropsychological Society, 18(4),
657–668.

Cohen, S., Janicki-Deverts, D., & Miller, G. E. (2007,
October). Psychological stress and disease. JAMA, 298
(14), 1685–1687.

de Groot, N. S., & Burgas, M. T. (2015, December). Is
membrane homeostasis the missing link between inflam-
mation and neurodegenerative diseases? Cellular and
Molecular Life Sciences, 72(24), 4795–4805.

Decker, M. R., Benning, L., Weber, K. M., Sherman, S. G.,
Adedimeji, A., Wilson, T. E., & Golub, E. T. (2016,
November). Physical and sexual violence predictors: 20
Years of the women’s interagency HIV study cohort.
American Journal of Preventive Medicine, 51(5),
731–742.

Deiss, R., Campbell, C. J., Watson, C. W.-M., Moore, R. C.,
Crum-Cianflone, N. F., Wang, X., … Moore, D. J.
(2019). Posttraumatic stress disorder and neurocognitive
impairment in a U.S. military cohort of persons living
with HIV. Psychiatry, 82(3), 228–239.

Grant, I. (2008). Neurocognitive disturbances in HIV.
International Review of Psychiatry, 20(1), 33–47.

Habib, A. G., Yakasai, A. M., Owolabi, L. F., Ibrahim, A.,
Habib, Z. G., Gudaji, M., … Nashabaru, I. (2013,
October). Neurocognitive impairment in HIV-1-infected
adults in Sub-Saharan Africa: A systematic review and
meta-analysis. International Journal of Infectious
Diseases, 17(10), e820–31.

Heaton, R. K., Clifford, D. B., Franklin, D. R. F.,
Woods, S. P., Ake, C., Vaida, F., … Grant, I. (2010,
December). HIV-associated neurocognitive disorders
persist in the era of potent antiretroviral therapy:
CHARTER study. Neurology, 75(23), 2087–2096.

Heaton, R. K., Jr, D. R. F., Deutsch, R., Letendre, S.,
Ellis, R. J., Casaletto, K., … Teshome, M. (2015,
February). Neurocognitive change in the era of HIV
combination antiretroviral therapy: The longitudinal
CHARTER study. Clinical Infectious Diseases, 60(3),
473–480.

Jelinek, L., Jacobsen, D., Kellner, M., Larbig, F.,
Biesold, K. H., Barre, K., & Moritz, S. (2006, August).
Verbal and nonverbal memory functioning in posttrau-
matic stress disorder (PTSD). Journal of Clinical and
Experimental Neuropsychology, 28(6), 940–948.

Kapetanovic, S., Norato, G., Nair, G., Julnes, P. S.,
Traino, K. A., Geannopoulos, K., & Nath, A. (2020,
March). Effect of HIV and interpersonal trauma on
cortical thickness, cognition, and daily functioning.
JAIDS Journal of Acquired Immune Deficiency
Syndromes. doi:10.1097/QAI.0000000000002358

Kessler, R. C., Rose, S., Koenen, K. C., Karam, E. G.,
Stang, P. E., Stein, D. J., … Carmen Viana, M. (2014,
October). How well can post-traumatic stress disorder
be predicted from pre-trauma risk factors? An explora-
tory study in the WHO World Mental Health Surveys.

World Psychiatry: Official Journal of the World
Psychiatric Association (WPA), 13(3), 265–274.

Kumar, A. M., Fernandez, J. B., Singer, E. J., Commins, D.,
Waldrop-Valverde, D., Ownby, R. L., & Kumar, M.
(2009, May). Human immunodeficiency virus type 1 in
the central nervous system leads to decreased dopamine
in different regions of postmortem human brains.
Journal of Neurovirology, 15(3), 257–274.

Lagarde, G., Doyon, J., & Brunet, A. (2010, May). Memory
and executive dysfunctions associated with acute post-
traumatic stress disorder. Psychiatry Research, 177(1–2),
144–149.

Lin, K., Taylor, M. J., Heaton, R., Franklin, D., Jernigan, T.,
Fennema-Notestine, C., … The CHARTER Group, F.
(2011, March). Effects of traumatic brain injury on cog-
nitive functioning and cerebral metabolites in
HIV-infected individuals. Journal of Clinical and
Experimental Neuropsychology, 33(3), 326–334.

Machtinger, E. L., Wilson, T. C., Haberer, J. E., &
Weiss, D. S. (2012, November). Psychological trauma
and PTSD in HIV-positive women: A meta-analysis.
AIDS and Behavior, 16(8), 2091–2100.

Maki, P. M., Rubin, L. H., Valcour, V., Martin, E.,
Crystal, H., Young, M., … Anastos, K. (2015, January).
Cognitive function in women with HIV: Findings from
the women’s interagency HIV study. Neurology, 84(3),
231–240.

Malan-Muller, S., Hemmings, S. M., Spies, G., Kidd, M.,
Fennema-Notestine, C., & Seedat, S. (2013). Shorter
telomere length – A potential susceptibility factor for
HIV-associated neurocognitive impairments in South
African women [corrected]. PloS One, 8(3), e58351.

McEwen, B. S. (2006). Protective and damaging effects of
stress mediators: Central role of the brain. Dialogues in
Clinical Neuroscience, 8(4), 367–381.

McLaughlin, K. A., Koenen, K. C., Bromet, E. J.,
Karam, E. G., Liu, H., Petukhova, M., & Kessler, R. C.
(2017, November). Childhood adversities and
post-traumatic stress disorder: Evidence for stress sensi-
tisation in the World Mental Health Surveys. British
Journal of Psychiatry, 211(5), 280–288.

Moher, D., Liberati, A., Tetzlaff, J., & Altman, D. G. (2009,
July). Preferred reporting items for systematic reviews
and meta-analyses: The PRISMA statement. PLoS
Medicine, 6(7), e1000097.

Ouzzani, M., Hammady, H., Fedorowicz, Z., &
Elmagarmid, A. (2016). Rayyan-a web and mobile app
for systematic reviews. Systematic Reviews, 5(1), 210.

Pavlovic, D., Pekic, S., Stojanovic, M., & Popovic, V. (2019,
June). Traumatic brain injury: Neuropathological, neu-
rocognitive and neurobehavioral sequelae. Pituitary, 22
(3), 270–282.

Pukay-Martin, N. D., Cristiani, S. A., Saveanu, R., &
Bornstein, R. A. (2003). The relationship between stress-
ful life events and cognitive function in HIV-infected
men. The Journal of Neuropsychiatry and Clinical
Neurosciences, 15(4), 436–441.

Qureshi, S. U., Long, M. E., Bradshaw, M. R., Pyne, J. M.,
Magruder, K. M., Kimbrell, T., & Kunik, M. E. (2011).
Does PTSD impair cognition beyond the effect of
trauma? Journal of Neuropsychiatry, 23(1), 16–28.

Rehm, K. E., & Konkle-Parker, D. (2017, September).
Association of CD4+ T cell subpopulations and psycho-
logical stress measures in women living with HIV. AIDS
Care, 29(9), 1107–1111.

Ross, L. E., Grigoriadis, S., Mamisashvili, L., Koren, G.,
Steiner, M., Dennis, C.-L., & Mousmanis, P. (2011,

EUROPEAN JOURNAL OF PSYCHOTRAUMATOLOGY 17

https://doi.org/10.1097/QAI.0000000000002358

December). Quality assessment of observational studies
in psychiatry: An example from perinatal psychiatric
research. International Journal of Methods in
Psychiatric Research [Internet], 20(4), 224–234.
Retrieved from http://www.scopus.com/inward/record.
url?eid=2-s2.0-79951999327&partnerID=tZOtx3y1

Rubin, L. H., Cook, J. A., Springer, G., Weber, K. M.,
Cohen, M. H., Martin, E. M., & Maki, P. M. (2017,
November). Perceived and post-traumatic stress are
associated with decreased learning, memory, and fluency
in HIV-infected women. AIDS, 31(17), 1401–2393.

Rubin, L. H., Cook, J. A., Weber, K. M., Cohen, M. H.,
Martin, E., Valcour, V., & Maki, P. M. (2015, August).
The association of perceived stress and verbal memory is
greater in HIV-infected versus HIV-uninfected women.
Journal of NeuroVirology, 21(4), 422–432.

Rubin, L. H., & Maki, P. M. (2019a, August). Neurocognitive
complications of HIV infection in women: Insights from the
WIHS cohort. Current Topics in Behavioral Neurosciences.

Rubin, L. H., & Maki, P. M. (2019b, February). HIV, depres-
sion, and cognitive impairment in the era of effective anti-
retroviral therapy. In Current HIV/AIDS Reports. Berlin,
Heidelberg: Springer. doi:10.1007/7854_2019_101

Rubin, L. H., Meyer, V. J. J., Conant, R.,
Sundermann, E. E., Wu, M., Weber, K. M., &
Maki, P. M. (2016, August). Prefrontal cortical volume
loss is associated with stress-related deficits in verbal
learning and memory in HIV-infected women.
Neurobiology of Disease, 92(Pt B), 166–174.

Rubin, L. H., Wu, M., Sundermann, E. E., Meyer, V. J.,
Smith, R., Weber, K. M., & Maki, P. M. (2016,
December). Elevated stress is associated with prefrontal
cortex dysfunction during a verbal memory task in women
with HIV. Journal of NeuroVirology, 22(6), 840–851.

Schuitevoerder, S., Rosen, J. W., Twamley, E. W., Ayers, C. R.,
Sones, H., Lohr, J. B., & Thorp, S. R. (2013, August). A
meta-analysis of cognitive functioning in older adults with
PTSD. Journal of Anxiety Disorders, 27(6), 550–558.

Scott, J. C., Matt, G. E., Wrocklage, K. M., Crnich, C.,
Jordan, J., Southwick, S. M., & Schweinsburg, B. C.
(2015, January). A quantitative meta-analysis of neuro-
cognitive functioning in posttraumatic stress disorder.
Psychological Bulletin, 141(1), 105–140.

Scott, K. M., Koenen, K. C., King, A., Petukhova, M. V.,
Alonso, J., Bromet, E. J., & Kessler, R. C. (2018, January).
Post-traumatic stress disorder associated with sexual
assault among women in the WHO World Mental
Health Surveys. Psychological Medicine, 48(1), 155–167.

Simioni, S., Cavassini, M., Annoni, J.-M., Rimbault
Abraham, A., Bourquin, I., Schiffer, V., … Du
Pasquier, R. A. (2010, June). Cognitive dysfunction in
HIV patients despite long-standing suppression of
viremia. AIDS, 24(9), 1243–1250.

Spies, G., Afifi, T. O., Archibald, S. L., Fennema-
Notestine, C., Sareen, J., & Seedat, S. (2012). Mental
health outcomes in HIV and childhood maltreatment:

A systematic review. Systematic Reviews, 1, 30.
doi:10.1186/2046-4053-1-30.

Spies, G., Ahmed-Leitao, F., Fennema-Notestine, C.,
Cherner, M., & Seedat, S. (2016, April). Effects of HIV
and childhood trauma on brain morphometry and neu-
rocognitive function. Journal of NeuroVirology, 22(2),
149–158.

Spies, G., Fennema-Notestine, C., Archibald, S. L.,
Cherner, M., & Seedat, S. (2012). Neurocognitive deficits
in HIV-infected women and victims of childhood
trauma. AIDS Care, 24(9), 1126–1135.

Spies, G., Fennema-Notestine, C., Cherner, M., & Seedat, S.
(2017, January). Changes in cognitive function in
women with HIV infection and early life stress. AIDS
Care, 29(1), 14–23.

Sumner, J. A., Hagan, K., Grodstein, F., Roberts, A. L.,
Harel, B., & Koenen, K. C. (2017, April). Posttraumatic
stress disorder symptoms and cognitive function in
a large cohort of middle-aged women. Depression and
Anxiety, 34(4), 356–366.

Tomoda, A., Sheu, Y.-S., Rabi, K., Suzuki, H.,
Navalta, C. P., Polcari, A., & Teicher, M. H. (2011,
January). Exposure to parental verbal abuse is associated
with increased gray matter volume in superior temporal
gyrus. Neuroimage, 54, S280–6.

Valcour, V., Chalermchai, T., Sailasuta, N., Marovich, M.,
Lerdlum, S., Suttichom, D., … Ananworanich, J. (2012,
July). Central nervous system viral invasion and inflam-
mation during acute HIV infection. The Journal of
Infectious Diseases, 206(2), 275–282.

Watson, C. W.-M., Sundermann, E. E., Hussain, M. A.,
Umlauf, A., Thames, A. D., Moore, R. C., … Moore, D. J.
(2019, January). Effects of trauma, economic hardship, and
stress on neurocognition and everyday function in HIV.
Health Psychology, 38(1), 33–42.

Wiegner, L., Hange, D., Björkelund, C., & Ahlborg, G. J.
(2015, March). Prevalence of perceived stress and asso-
ciations to symptoms of exhaustion, depression and
anxiety in a working age population seeking primary
care–an observational study. BMC Family Practice, 16
(1), 38.

Williams, M. E., Zulu, S. S., Stein, D. J., Joska, J. A., &
Naude, P. J. W. (2020, March). Signatures of HIV-1
subtype B and C Tat proteins and their effects in the
neuropathogenesis of HIV-associated neurocognitive
impairments. Neurobiology of Disease, 136, 104701.

Womersley, J. S., Spies, G., Seedat, S., & Hemmings, S. M. J.
(2019, April). Childhood trauma interacts with ApoE to
influence neurocognitive function in women living with
HIV. Journal of NeuroVirology, 25(2), 183–193.

Yuan, N. Y., & Kaul, M. (2019, August). Beneficial and
Adverse Effects of cART Affect Neurocognitive Function
in HIV-1 Infection: Balancing Viral Suppression against
Neuronal Stress and Injury [published online ahead of
print, 2019 Aug 6]. J Neuroimmune Pharmacol.
doi:10.1007/s11481-019-09868-9

18 G. SPIES ET AL.

http://www.scopus.com/inward/record.url?eid=2-s2.0-79951999327%26partnerID=tZOtx3y1

http://www.scopus.com/inward/record.url?eid=2-s2.0-79951999327%26partnerID=tZOtx3y1

https://doi.org/10.1007/7854_2019_101

https://doi.org/10.1186/2046-4053-1-30

https://doi.org/10.1007/s11481-019-09868-9

Copyright of European Journal of Psychotraumatology is the property of Taylor & Francis
Ltd and its content may not be copied or emailed to multiple sites or posted to a listserv
without the copyright holder’s express written permission. However, users may print,
download, or email articles for individual use.

• Abstract

Abstract
Abstract

• 1. Background
• 2. Methods

2.1. Inclusion and exclusion criteria
2.2. Search strategies
2.3. Data analysis

• 3. Results

3.1. Quality assessment of included studies
3.2. Study characteristics
3.2.1. Study design
3.2.2. Study location
3.2.3. Trauma measurement
3.2.4. Trauma exposure
3.3. Findings on trauma and NCI
3.3.1. Cognitive domains
3.3.2. Associations between PTEs and SLEs and NCI
3.3.3. PTSD findings
3.3.4. PTE exposure and PTSD findings

• 4. Discussion

4.1. Associations between PTEs and SLEs and NCI
4.2. Methodological considerations of included studies
4.3. Clinical implications
4.4. Implications for research

• 5. Limitations

5.1. Limitations of reviewed studies
5.2. Limitations of this study

• 6. Conclusion

Disclosure statement
Funding
References