Perinatally-acquired HIV remains a global health concern even in the era of antiretroviral therapy (ART), Africa is particularly severely affected, with 3.2 million perinatally infected children and adolescents . There is a large spectrum of neurologic conditions that may be caused by HIV in the brain, for example, epilepsy, cerebrovascular disease, peripheral neuropathy, and opportunistic infections . Current WHO HIV guidelines recommend universal access to ART . For infected children and adolescents, this means increased life expectancy, and the potential for social, economic, and cultural contributions throughout adulthood.
However, the extent of these contributions and the associated quality of life is tempered by risk for disease-associated cognitive impairment [4,5]. HIV-associated cognitive impairment persists despite earlier and better antiretrovirals access . Cognitive impairments are associated with mental health problems (e.g. anxiety and depression) and with diminished adaptive functioning (e.g. poor self-care and social communication) [5,7,8]. Hence, persistent and undiagnosed HIV-associated cognitive impairment in children and adolescents has long-term consequences for their academic, occupational, and socioeconomic achievement. Studies focusing on the typical cognitive phenotype of perinatally HIV-infected adolescents are that there is an overall intellectual disability and relative weaknesses in various other cognitive domains [9–11]. Based on the findings of a recently published systematic review and meta-analysis, the most commonly affected domains are working memory, processing speed, and executive function .
The gold standard for assessing cognitive function, and detecting impairment, is administering a comprehensive neuropsychological battery. Such administration is rarely feasible as a first-line option in low-income and middle-income countries (LMICs). Neuropsychological batteries are time-consuming to administer, require specialist training and knowledge to interpret, rely on expensive materials, and hence are not feasible in the already burdened healthcare systems in LMICs. Brief, valid screeners could address these issues, and might assist in ensuring that appropriate specialist assessment and intervention is provided for those at risk of being cognitively impaired . Hence, cognitive screening is an essential part of service delivery for this vulnerable group .
The current study aimed to validate a modified version of the International HIV Dementia Scale (IHDS), the y-IHDS (youth-IHDS), as a cognitive screening test for HIV-infected children and adolescents. The IHDS is an easy-to-use and widely employed screening tool that has good psychometric properties in HIV-infected adults [14,15].
Study design, setting, and participants
This was a cross sectional study of children enrolled in the Cape Town Adolescent Antiretroviral Cohort (CTAAC) . Perinatally HIV-infected children and adolescents (n = 203) were recruited from community health clinics. Inclusion criteria were being between 9 and 12 years of age, having acquired HIV vertically, having been stable on ART for more than 6 months, and having no history of central nervous system conditions or serious head injury. Major reasons for exclusion were previous major psychiatric conditions, central nervous system conditions, and loss of consciousness for more than 5 min. Age-matched uninfected controls (n = 44), from similar socioeconomic backgrounds, were recruited from a local youth clinic.
The CTAAC study was approved by the Human Research Ethics Committee of the University of Cape Town's Faculty of Health Sciences (HREC REF: 051/2013). Parents gave informed consent for their own participation and for their child to participate, and children and adolescents signed informed assent documents.
Measures and procedures
All study procedures were completed in private rooms within the University of Cape Town (UCT) Department of Psychiatry and Mental Health. Each participant was administered a comprehensive neuropsychological test battery by trained research assistants. All tests (including the y-IHDS) were translated into isiXhosa and back-translated into English to ensure compatibility with the original. The neuropsychological test battery assessed performance within various cognitive domains and included the following tests, among others: the Wechsler Abbreviated Scale of Intelligence; the Digit Span, Symbol Search, and Digit Symbol-Coding subtests of the Wechsler Intelligence Scale for Children; the Inhibition and Fingertip Tapping subtests of the NEPSY; and the Children's Colour Trails Test (refer to Hoare et al. for full description). See Supplementary file 1, http://links.lww.com/QAD/B444.
Each participant also completed the y-IHDS. Administration followed standard procedures for adults with one exception: adolescents were allowed to practice the words and motor tasks until they were able to complete them correctly at least once before they proceeded with the timed tasks. The y-IHDS is not without limitations, even when administered in adult populations , but of particular note is that sensitivity and specificity may be improved by the addition of screening questions which may tap into cognitive impairment . To this end, we asked the children and adolescents if they had ever repeated a grade at school, with responses coded as either yes or no. Repeating a grade at school has been shown to be an indicator of functional impairment in the school domain  and is also associated with cognitive impairment more generally . We refer to this modified version of the y-IHDS (i.e. a version that features increased practice opportunity on the verbal memory and motor tasks, and that includes the repeated grades question) as the y-IHDS.
The y-IHDS took approximately 3–5 min to complete and was administered by trained research assistants. Supplement 2, http://links.lww.com/QAD/B445, provides English and isiXhosa test administration guidelines for the instrument. Supplementary videos 1 and 2, http://links.lww.com/QAD/B442, http://links.lww.com/QAD/B443, demonstrate English and isiXhosa administrations of the instrument.
Parents of the adolescent participants completed measures related to their child's functioning. Here, we analysed data from two of those measures [the Vinelands Adaptive Behaviour Scale II (VABS2)  and the Child Behaviour Checklist (CBCL)]  to assess functional impairment in adolescents. Parents also completed basic questionnaires regarding their child's performance at school.
We captured all data from the neuropsychological test battery and from the y-IHDS into SPSS (version 25; IBM, Armonk, New York, USA). We used the same software to complete all inferential statistical analyses, with α set at 0.05 and confidence intervals at 95%.
Using data from the neuropsychological test battery, we created 10 composite cognitive domain scores (one each for general intellectual functioning, motor coordination, processing speed, attention, working memory, visual memory, verbal memory, visual–spatial ability, language, and executive function) using our previously described theoretical–statistical method . Briefly, we grouped test outcome variables based on theory regarding the cognitive function they assessed, and then tested the inter-relatedness (internal consistency) of the variables within each group using Cronbach's alpha correlation coefficient. Only those groups with alpha values more than 0.70 were retained. Eight participants were unable to complete some of the neuropsychological tests due to severe cognitive impairment. For these participants, individual composite scores were calculated based on the tests they were able to complete. For these eight participants, only two were diagnosed as having a major neurocognitive disorder (ND). We scored the y-IHDS as follows: Total score = [(Items 2 + 3 + 4) − (Item 1)] (refer to Supplement 1, http://links.lww.com/QAD/B444). This scoring of the y-IHDS takes into account school performance. The scoring of the y-IHDS does not include the time allowed for practice. Only the actual timed tasks, word recall, and school performance on the y-IHDS are scored. Regarding the behavioural data, adolescents were classified as having functional impairment if either their VABS2 Total score or CBCL Total Competence score was lower than the manual-based cut-off score for impairment.
Regarding descriptive statistics, we conducted between-group comparisons on sociodemographic, cognitive, and behavioural data to provide an overview of the sample.
As a first step in our inferential analyses, we conducted bivariate correlational analyses (using the Spearman rho's coefficient for non-normally distributed data) to determine the magnitude of association between y-IHDS score and scores within each composite cognitive domain.
Youth-International HIV Dementia Scale validation process
Next, we followed a four-step process to validate the y-IHDS:
- We applied the youth HIV-associated neurocognitive disorder diagnostic criteria  to each participant's neuropsychological profile to screen for HIV-associated neurocognitive disorders and to classify each as having either a major neurocognitive disorder (major ND), minor neurocognitive disorder (minor ND), or no impairment (no ND). The criteria draw on the adolescent's cognitive performance and functional ability, with the taxonomy structured as follows: major ND = performance of more than 2SD below the mean in two separate cognitive domains, and the presence of functional impairment; minor ND = performance of more than 1SD below the mean in two separate cognitive domains, with or without functional impairment; and no ND = no cognitive impairment.
- We performed a series of cross-tabulations (i.e. y-IHDS x youth HIV-associated neurocognitive disorders diagnosis) at various y-IHDS cut-off scores (i.e. 8, 9, 9.5, 10, and 10.5) to determine the ratio of true positives vs. false positives, and false negatives vs. true negatives.
- Based on the frequencies obtained at the previous step, we calculated sensitivity and specificity values for each cut-off score using a MedCalc application (https://www.medcalc.org/calc/diagnostic_test.php).
- We constructed receiver operating characteristic (ROC) curves using the youth neurocognitive disorders classification as the state variable, plotted against the y-IHDS raw scores for major ND, minor ND, and all HIV-associated neurocognitive disorders or impairments.
After calculating the optimum cut-off score on the y-IHDS, we used independent samples t tests to determine if there were significant differences in cognitive performance (estimated using composite cognitive domain scores) between those HIV-infected children and adolescents scoring at or above that cut-off point and those scoring below.
Demographic and clinical characteristics of the adolescents are described in Table 1. Analyses detected no significant between-group differences with regard to age, sex, language, or years of education. However, significantly more participants in the HIV-infected group had repeated at least one grade at school. Regarding the VABS2, there were no significant between-group differences. There was a significant difference seen on the CBCL total competence scores, with the controls displaying better total competence than the HIV-infected group.
Table 2 presents the cognitive data. Analyses detected significant between-group differences for y-IHDS total score, as well as for score in every cognitive domain except motor coordination. In each case, the healthy control group outperformed the HIV-infected group.
The series of bivariate correlational analyses detected significant positive associations between y-IHDS score and performance in each cognitive domain, except attention (i.e. higher y-IHDS scores were correlated with overall better cognitive performance; refer to Table 3). This association tells us that the y-IHDS is able to tap into changes in cognition and to detect either improvements or decline in the various cognitive domains, making it suitable as a screening tool to detect the risk of HIV-associated neurocognitive disorder. This is further supported by the fact that only the domain of motor-coordination is significantly correlated with y-IHDS scores in the control group.
Youth-International HIV Dementia Scale validation
Based on the youth neurocognitive disorder criteria, we classified eight participants as having a major ND (3%), 96 as minor ND (39%), and 144 as no ND (or no impairment, 58%). The mean y-IHDS score for the eight participants classified as major ND was 6.8 with a SD of 2.5. Using those classifications, we created the cross-tabulations shown in Table 4 and hence examined the ability of the y-IHDS (at cut-off scores of 8, 9, 9.5, 10, and 10.5, in which in each case a score at or below the cut-off indicated impairment) to discriminate among major ND and nonimpaired participants, minor ND and nonimpaired participants, and any form of ND (major or minor) and nonimpaired participants. A y-IHDS cut-off score of 9 or less yielded the best balance between sensitivity and specificity, Table 4. However, the cut-off score of 10 or less yielded the best sensitivity (but had poor specificity) and is therefore regarded as the best value for cut-off for the y-IHDS.
Follow-up ROC analyses indicated that, at each the preferred cut-off value of 10 or less, the y-IHDS was significantly better than chance at discriminating major ND, minor ND, and any form of ND from no impairment (Figs. 1–3). Although the area under the curve (AUC) value was acceptable in each case, the strongest of those values (0.859, P < 0.001) was associated with the ability of the instrument to discriminate major ND from no impairment. The ROC for the y-IHDS to screen for any form of ND yielded an AUC of 0.695 (P = 0.001) and for minor ND the AUC was 0.682 (P < 0.001).
Finally, an independent samples t test detected significant performance differences in each cognitive domain (except attention) between HIV-infected adolescents with y-IHDS scores 10 or less and those with scores more than 10 (Table 5).
The current study aimed to validate an adapted version of the IHDS (the Youth IHDS, or y-IHDS) as a cognitive screening test for HIV-infected children and adolescents. Our analyses found that, at a cut-off score of 10 or less, the instrument had good sensitivity but low specificity in discriminating children and adolescents with HIV-associated cognitive impairment from those with unimpaired function, but also displayed high negative predictive ratios, suggesting that the y-IHDS as a screening tool (which requires minimal resources) may be a useful addition for clinicians working in resource-poor contexts in with a high adolescent HIV prevalence.
Our validation process encompassed four steps. At the first step, we found that 46% of HIV-infected participants could be classified as displaying either a major (3%; n = 8) or a minor (43%; n = 88) ND. The proportions of participants with either a minor or major ND are slightly lower to those previously found in a smaller cohort  comparable with the CTAAC sample. Although it may seem that the rates of major ND are declining, subtle or minor forms of neurocognitive impairment may go undetected, and therefore a validated quick screening tool for all forms of, neurocognitive impairment encompassing major and minor ND, is needed. It is encouraging to see that only 3% of HIV-infected youth had a major ND, this could possibly be a sign of the effectiveness and benefit of prolonged ARV treatment in protecting the brain .
Of note is that 18% of HIV-negative controls (n = 8) were classified as having a minor neurocognitive disorder, consistent with prior reports . The relatively high rate of minor neurocognitive impairment in controls may be a result of socioeconomic disadvantage resulting in, for instance, malnutrition and a lack of stimulating home environments . In our study, of the group of 144 participants who had no cognitive impairment, 103 had some form of functional impairment. In these cases, their functional impairment could be caused by factors other than their cognitive ability [e.g. behavioural problems or attention deficit hyperactivity disorder (ADHD)]. The controls in this study are from similar risk backgrounds as the HIV-infected adolescents, and are exposed to and experience the same social struggles, as noted above, and thus the rates of cognitive impairment and repeated grades in this control group is higher than one would normally expect to see in a control group from a well-resourced country.
At the second validation step, we found that at the lowest y-IHDS cut-off score (i.e. 8) the rate of true positives was lower than that of false positives. At higher cut-off scores (i.e. 9, 9.5, 10, and 10.5), however, the rate of true positives was higher than that of false positives. Screening tools should be optimized to minimize false negative and false positive results. Considering the rates of cognitive impairment we observed in this cohort, it becomes important to minimize the rates of false negatives when screening for neurocognitive disorders. High rates of false negatives would provide an incorrect representation of the extent of the problem in real-world terms and may miss detecting an adolescent with a neurocognitive disorder. Reducing false positives is also important in a low resource setting as it is not feasible to conduct full neuropsychological testing on individuals who may not require it.
We found that a cut-off of 9 produced the best balance between sensitivity and specificity, but that a cut-off of 10 gave higher sensitivity. We recommend using the cut-off of 10 or less because the associated negative predictive value (NPV) value is 85%, which means at this cut-off we can correctly exclude cases of no cognitive impairment. In the public health context of South Africa (and other low-resource settings), the ideal is to reduce false negatives and identify as many true positive cases as possible because in the long-term public health costs of false negatives is potentially higher than the cost of assessing false positive cases . In this case, we could therefore accept a screening tool with high sensitivity and relatively poor specificity. This tool is not intended to formally diagnose HIV-associated neurocognitive disorders, but rather to screen for the risk of having either a major or minor ND. Youth who screen as high risk should be referred for further clinical evaluation for a formal diagnosis.
In a setting like South Africa, where the rates of HIV-associated neurocognitive disorders are high, one would rather run the risk of over screening as opposed to under screening and possibly missing cases which require further assessment and intervention. A high sensitivity of at least 80% is recommended for screening , therefore, based on these results and taking into account the high rates of all forms of HIV-associated cognitive impairment and the high NPVs, a cut-off score of 10 or less is suggested as optimal so as not to falsely exclude cases.
For the fourth and last step in the validation process, our ROC analyses findings suggest that using a y-IHDS cut-off score of 10 or less yields a moderate screening utility for detecting all forms of cognitive impairment and minor ND. Furthermore, using a y-IHDS cut-off score of 10 or less was able to accurately screen for risk of major ND in all of the cases confirmed as having a major ND as determined by the youth HIV-associated neurocognitive disorder diagnostic criteria.
Overall, these results suggest that the use of the y-IHDS could be valuable in both clinical practice and research. Specifically, our results suggest that y-IHDS is a valid screening tool for neurocognitive disorders in HIV-infected children and adolescents. Early detection of HIV-associated neurocognitive disorders, using an instrument such as the y-IHDS, should be combined with early initiation of ARVs (WHO Treat All policy has been implemented by 70% of LMICs ) and other elements of a multidisciplinary interventions approach to optimize long-term health, and minimize occupational health and safety risks, of infected children and adolescents [29,30].
The current findings are especially relevant to clinicians working in low-resource settings with high HIV prevalence rates. The benefits of having a screening tool that can be administered by lay professionals and that can detect the risk of all forms of HIV-associated cognitive impairment may enable rapid screening and referral of children and adolescents, and alleviate pressure on already overburdened healthcare systems . Of particular note here is that administration, scoring, and interpretation of the y-IHDS does not require specialist training. It takes a maximum of 5–7 min to administer (including the practice runs and depending on the child's ability) and score and requires only minimal resources (a paper and pen).
The fact that higher numbers of HIV-infected participants compared with the HIV-uninfected controls were included for logistical reasons related to the larger study, may be seen as a study limitation; however, these groups were well matched, and the types of analyses performed here are not sensitive to unequal variances between the groups. A limitation is that although all the measures on the neuropsychological test battery were translated and back-translated into isiXhosa and administered in the participant's preferred language, we did not validate this battery. However, these tests were selected based on their widespread usability in adolescents in low-resource settings, taking into consideration their reported psychometric properties . We also took steps to ensure that the battery was administered in accordance with the International Test Commission Guidelines .
More research into possibly adding cognitive screening questions to the y-IHDS will strengthen the findings here and may possibility produce a tool which is even more suited to screening for HIV-associated cognitive impairment in children and adolescents. Adding cognitive screening questions to the IHDS has been shown to improve the sensitivity and specificity in adult populations .
In conclusion, this study shows that the y-IHDS, at a cut-off score of 10 or less may be a useful and quick screening tool to assess the risk of all forms of HIV-associated neurocognitive disorders in perinatally infected children and adolescents. We recommend that the instrument be considered for addition to The Western Cape Consolidated Guidelines for HIV Treatment (HIV/AIDS/STI/TB Directorate, 2015) for children and adolescents . Although these guidelines recommend neurocognitive screening as part of the monitoring process for children and adolescents, the recommended tool only assesses neurodevelopment in children up to the age of 6 years. In contrast, the y-IHDS can be administered to children aged 9 years and older and is suited for use throughout adolescence.
Tembela Molwana and Yvonne Sani, for assistance in data collection and capturing. Teboho Linda and Bulelwa Mtukushe, for assistance with isiXhosa translations. A big thank you to the parents and adolescents who participated in the CTAAC neuro substudy.
The research was supported by Eunice Kennedy Shriver National Institute of Child Health and Human Development under grant R01HD074501. N.J.P. receives funding from the South African Medical Research Council (SAMRC) in terms of the National Health Scholarships Programme (NHSP) towards her PhD studies. All authors have no financial relationships relevant to this article to disclose.
Contributor's statements: N.J.P. developed the concept for the article, was responsible for overseeing data collection and quality control thereof, conducting the ground work and final statistical analysis. She is responsible for the write-up of the final article and for submission to the journal, and she approved the final article as submitted. N.J.P. will also be the corresponding author for this article and will address any reviewer comments received. N.J.P. receives funding from the South African Medical Research Council in terms of the National Health Scholarships Programme towards her PhD studies for 2015–2017.
K.G.F.T. provided input regarding the calculation of each specific cognitive domain and provided assistance with the statistical analyses. K.G.F.T. critically reviewed the article and made relevant changes prior to submission to the journal. K.G.F.T. has no disclosures.
L.M. provided consultation regarding statistical analyses, reviewed the article, and suggested/made relevant changes prior to submission to the journal. L.M. has no disclosures.
N.S. is the author of the original IHDS screening tool and assisted in conceptualizing this article. He provided guidance on the procedures for test validation and reporting. N.S. reviewed the article and suggested/made relevant changes prior to submission to the journal. N.S. has no disclosures.
H.J.Z. is the principle investigator for CTAAC project within which this study is nested, and provided insight into clinical relevance. H.J.Z. reviewed the article and suggested/made relevant changes prior to submission to the journal. H.J.Z. is supported by the South African Medical Research Council.
D.J.S. assisted with the concept for the article and provided supervision of this work throughout the entire process. Furthermore, D.J.S. critically reviewed the article and suggested/made relevant changes prior to submission to the journal. D.J.S. is supported by the South African Medical Research Council.
J.H. assisted with the concept for the article and provided clinical input during the entire process (e.g. which age categories to assess, division of HIV groups, and clinical relevance of the data presented). Furthermore, J.H. critically reviewed the article and suggested/made relevant changes prior to submission to the journal. J.H. has no disclosures.
All authors approved the final article as submitted and agree to be accountable for all aspects of the work.
Screening for HIV-Associated Neurocognitive Disorders in Perinatally Infected Adolescents: y-IHDS Validation.
Conflicts of interest
There are no conflicts of interest.
1. World Health Organization. Definitions of key terms
. Geneva, Switzerland: World Health Organization; 2013.
2. Donald KA, Hoare J, Eley B, Wilmshurst JM. Neurologic complications of pediatric human immunodeficiency virus: implications for clinical practice and management challenges in the African setting
. Semin Pediatr Neurol
3. World Health Organization. Consolidated guidelines on the use of antiretroviral drugs for treating and preventing HIV infection
. Geneva, Switzerland: World Health Organization; 2016.
4. World Health Organization. Fact sheet on HIV/AIDS. http://wwwwhoint/mediacentre/factsheets/fs360/en/2016
. [Accessed; July 2017]
5. Heaton RK, Marcotte TD, Mindt MR, Sadek J, Moore DJ, Bentley H, et al. The impact of HIV-associated neuropsychological impairment on everyday functioning
. J Int Neuropsychol Soc
6. Heaton R, Clifford D, Franklin D, Woods S, Ake C, Vaida F, et al. HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy CHARTER Study
7. Phillips N, Amos T, Kuo C, Hoare J, Ipser J, Thomas KG, et al. HIV-associated cognitive impairment in perinatally infected children: a meta-analysis
8. Wachsler-Felder JL, Golden CJ. Neuropsychological consequences of HIV in children: a review of current literature
. Clin Psychol Rev
9. Smith R, Chernoff M, Williams PL, Malee KM, Sirois PA, Kammerer B, et al. Impact of human immunodeficiency virus severity on cognitive and adaptive functioning during childhood and adolescence
. Pediatr Infect Dis J
10. Kandawasvika G, Kuona P, Chandiwana P, Masanganise M, Gumbo F, Mapingure M, et al. The burden and predictors of cognitive impairment among 6- to 8-year-old children infected and uninfected with HIV from Harare, Zimbabwe: a cross-sectional study
. Child Neuropsychol
11. Cohen S, Ter Stege JA, Geurtsen GJ, Scherpbier HJ, Kuijpers TW, Reiss P, et al. Poorer cognitive performance in perinatally HIV-infected children versus healthy socioeconomically matched controls
. Clin Infect Dis
12. Bloch M, Kamminga J, Jayewardene A, Bailey M, Carberry A, Vincent T, et al. A screening strategy for HIV-associated neurocognitive disorders that accurately identifies patients requiring neurological review
. Clin Infect Dis
13. Barber TJ, Bradshaw D, Hughes D, Leonidou L, Margetts A, Ratcliffe D, et al. Screening for HIV-related neurocognitive impairment in clinical practice: challenges and opportunities
. AIDS Care
14. Sacktor NC, Wong M, Nakasujja N, Skolasky RL, Selnes OA, Musisi S, et al. The International HIV Dementia Scale: a new rapid screening test for HIV dementia
15. Joska JA, Westgarth-Taylor J, Hoare J, Thomas KG, Paul R, Myer L, et al. Validity of the international HIV dementia scale in South Africa
. AIDS Patient Care STDS
16. Brittain K, Asafu-Agyei NA, Hoare J, Bekker L-G, Rabie H, Nuttall J, et al. Association of adolescent- and caregiver-reported antiretroviral therapy adherence with HIV viral load among perinatally-infected South African adolescents
. AIDS Behav
17. Hoare J, Fouche J-P, Spottiswoode B, Donald K, Philipps N, Bezuidenhout H, et al. A diffusion tensor imaging and neurocognitive study of HIV-positive children who are HAART-naïve “slow progressors”
. J Neurovirol
18. Haddow LJ, Floyd S, Copas A, Gilson RJC. A systematic review of the screening accuracy of the HIV Dementia Scale and International HIV Dementia Scale
. PLoS One
19. Biederman J, Monuteaux MC, Doyle AE, Seidman LJ, Wilens TE, Ferrero F, et al. Impact of executive function deficits and attention-deficit/hyperactivity disorder (ADHD) on academic outcomes in children
. J Consult Clin Psychol
20. Under review, Phillips N, Mtukushe B, Myer L, Zar HJ, Stein DJ, Hoare J. HIV-associated neurocognitive disorders in perinatally infected children and adolescents: how we measure functioning matters
21. Sparrow SS, Cicchetti DV, Balla D. Vineland adaptive behavior scales (Vineland II), survey interview form/caregiver rating form
. Livonia, MN: Pearson Assessments; 2005.
22. Achenbach TM, Ruffle TM. The Child Behavior Checklist and related forms for assessing behavioral/emotional problems and competencies
. Pediatr Rev
23. Phillips NJ, Hoare J, Stein DJ, Myer L, Zar HJ, Thomas KG. HIV-associated cognitive disorders in perinatally infected children and adolescents: a novel composite cognitive domains score
. AIDS Care
24. Hoare J, Phillips N, Joska JA, Paul R, Donald KA, Stein DJ, et al. Applying the HIV-associated neurocognitive disorder diagnostic criteria to HIV-infected youth
25. Heaton RK, Franklin DR, Ellis RJ, McCutchan JA, Letendre SL, LeBlanc S, et al. HIV-associated neurocognitive disorders before and during the era of combination antiretroviral therapy: differences in rates, nature, and predictors
. J Neurovirol
26. Grantham-McGregor S, Cheung YB, Cueto S, Glewwe P, Richter L, Strupp B, et al. Developmental potential in the first 5 years for children in developing countries
27. Franchini M, Pieroni S, Passino C, Emdin M, Molinaro S. The CARPEDIEM algorithm: a rule-based system for identifying heart failure phenotype with a precision public health approach
. Front Public Health
28. Ford N, Vitoria M, Doherty M. Providing antiretroviral therapy to all who are HIV positive: the clinical, public health and programmatic benefits of treat all
. J Int AIDS Soc
29. Dobroszycki J, Lee P, Romo DL, Rosenberg MG, Wiznia A, Abadi J. Antiretroviral therapy in HIV-infected adolescents: clinical and pharmacologic challenges
. Expert Rev Clin Pharmacol
30. Flynn PM. A broader look at adolescents with perinatal HIV
. Memphis, Tennessee: Nature Publishing Group; 2018.
31. Cysique LA, Murray JM, Dunbar M, Jeyakumar V, Brew BJ. A screening algorithm for HIV-associated neurocognitive disorders
. HIV Med
32. Bartram D. The development of international guidelines on test use: the International Test Commission Project
. Int J Test
33. Gouse H, Casson-Crook M, Decloedt EH, Joska JA, Thomas KG. Adding a brief self-report cognitive tool to the IHDS improves effectiveness of identifying patients with HIV-associated dementia in South Africa
. J Neurovirol
34. Provincial Government of the Western Cape Department of Health HASTHD. The Western Cape Consolidated Guidelines for HIV Treatment: prevention of mother-to-child transmission of HIV (PMTCT), children, adolescents and adults
. Western Cape Department of Heath Website.Cape Town, South Africa: Western Cape Government; 2015.