Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2012 Jun 8.
Published in final edited form as: J Child Adolesc Ment Health. 2011 Aug 22;23(1):17–28. doi: 10.2989/17280583.2011.594245

Validation of the Pediatric Symptom Checklist in HIV-infected Batswana

Elizabeth Lowenthal 1,2,3,*, Kathy Lawler 2, Nurit Harari 2,3, Lesedi Moamogwe 3, Japhter Masunge 4,5, Motshodi Masedi 6, Bolefela Matome 5, Esther Seloilwe 4, Michael Jellinek 7, Michael Murphy 7, Robert Gross 2,3
PMCID: PMC3370684  NIHMSID: NIHMS350057  PMID: 22685483

Abstract

Objective

To determine the validity of the Pediatric Symptom Checklist (PSC), a brief measure of psychosocial health, for screening HIV+ Batswana children.

Method

Setswana versions of the parent and child PSC were administered to 509 HIV+ Batswana children (age 8–16) and their parents/guardians. Test properties were evaluated and cut-off scores were derived using receiver operating characteristic curve analysis. Scores on the parent-completed PSC and the child-completed PSC-Y were compared to parental and clinic staff reports of concern about the child’s psychosocial health and to scores on the Children’s Depression Inventory and the Revised Children’s Manifest Anxiety Scale.

Results

The Setswana PSC has high internal consistency (Cronbach’s alpha 0.87 for the parent-completed version). Comparing PSC scores to parental reports of concern and child-reported depression symptoms, a cut-off score of 20 on the PSC and PSC-Y maximised the sensitivity and specificity.

Conclusions

The PSC performed well in Setswana-speaking children and is a promising screening tool for paediatric psychosocial problems in busy clinical settings. Screening with the PSC may allow for early detection and treatment of psychosocial problems. This is likely to be of particular value for HIV+ children for whom HIV treatment non-adherence may result from untreated psychosocial dysfunction.

Background

Psychosocial dysfunction contributes greatly to morbidity among children and adolescents (Kelleher and Wolraich 1996, Prince et al. 2007). Most psychosocial problems are unrecognized and untreated, even among young people attending clinics for other healthcare needs (Costello et al. 1988, Navon et al. 2001). Screening for psychosocial dysfunction is not routinely done in most health centres since such screening can be extremely time-consuming and most screening modalities require special training (Chambers et al. 1985, Jellinek and Murphy 1990). However, children and adolescents presenting for chronic medical care are particularly likely to have psychosocial co-morbidities (Prince et al. 2007, Saini et al. 2007). Children and adolescents infected with and affected by HIV are at particularly high risk of suffering from psychosocial problems (Pao et al. 2000, Scharko 2006, DeLaMora, Aledort and Stavola 2006, Fielden et al. 2006, Bauman et al. 2007, Cluver, Gardner and Operario 2007, Lee et al. 2007, Rao et al. 2007, Mellins et al. 2008, Hazra, Siberry and Mofenson 2010). Since most HIV-infected children, even in resource-limited settings, attend clinic regularly for their HIV-related medical care, clinic staff members have the opportunity to screen for and address psychosocial concerns in this vulnerable population.

Despite impressive improvements in preventing mother-to-child transmission of HIV in Botswana, the burden of HIV among children and adolescents in this country is large (Stover et al. 2008). UNAIDS estimates that about 16 000 children aged 0–14 years are infected with HIV (UNAIDS 2011). Successful treatment programmes have allowed children infected with HIV around the time of birth to survive through adolescence, even in resource-limited settings such as those found in Botswana (Kline et al. 2004, Bolton-Moore et al. 2007). HIV-infected children have variable needs for psychosocial support as they grow older. Generally, these needs are higher among HIV-infected children than among uninfected children in the same populations (UNICEF 2010).

Screening for psychosocial dysfunction among children and adolescents seeking routine medical care can be facilitated using brief screening questionnaires (Eisert, Sturner and Mabe 1991, Sturner 1991). The Pediatric Symptom Checklist (PSC) was developed to allow for rapid identification of individuals between the ages of 6 and 16 years who would benefit most from further services to evaluate and treat emotional and behavioural problems (Jellinek, Murphy and Burns 1986, Jellinek and Murphy 1990). The PSC consists of 35 items that are rated as never, sometimes or often present (scored 0, 1 or 2). A 17-item version of the PSC (PSC-17) has also been used successfully to detect youth with psychosocial impairment (Gall et al. 2000, Borowsky, Mozayeny and Ireland 2003, Duke, Ireland and Borowsky 2005, Gardner et al. 2007). Both versions are scored by simply adding the scores on each individual item. No special training is required for scoring or interpretation. The PSC is typically completed by a parent, however, a youth self-report version (PSC-Y) is also available (Pagano et al. 1996). Like the parent-completed version of the PSC, results of the PSC-Y correlate well with more comprehensive measures of psychosocial dysfunction. However, children identified as at risk through the PSC-Y are frequently not identified as at risk when measured concurrently with the PSC (Pagano et al. 1996).

Parent-completed questionnaires such as the PSC are more routinely recommended than child-completed questionnaires such as the PSC-Y for screening school-aged children for psychosocial problems (Conner 1979). However, there are a few reasons to consider utilising child-report questionnaires in addition to or instead of parent report questionnaires for screening HIV+ children. While parents can generally report accurately about behaviour problems, they are less aware of children’s dysphoria and anxiety than the children themselves (Weissman, Orvaschel and Padian 1980, Herjanic and Reich 1982, Wren, Bridge and Birmaher 2004). In addition, the parent or guardian who knows the child best may not be the same person who routinely brings the child to clinic for routine HIV-related follow-up. For example, a child’s primary caregiver might be the grandmother, but an uncle who lives in a different household might be the most available to take the child to clinic appointments. Adults who do not know the child well are likely to be limited in their ability to accurately report on the child’s psychosocial health.

For assessment of depressive symptoms in children in sub-Saharan Africa, the Children’s Depression Inventory has been used by several research groups (Cluver and Orkin 2009, Ndetei et al. 2009, Ruiz-Casares et al. 2009, Bach and Louw 2010). The Revised Children’s Manifest Anxiety Scale has also been used extensively in sub-Saharan Africa to evaluate anxiety symptoms in children (Pela and Reynolds 1982, Cluver et al. 2007). Because of the lack of validated standards for the measurement of depression and anxiety symptoms among Batswana children, we chose tools that have been trusted for use in similar populations. Both these scales are based on the children’s self-report of their symptoms.

The PSC is completed by parents and the PSC-Y is completed by the children themselves. A positive score on either version of the PSC suggests the need for further evaluation. Data from early studies with the PSC indicate that about two out of three children who have a positive score on the PSC will have been correctly identified as having a moderate or serious impairment in psychosocial functioning (Walker, LaGrone and Atkinson 1989). A score of 28 or higher is considered positive when the English language PSC is used in US populations of school-going age (Jellinek et al. 1999). For the PSC-17, a cut-off score of 15 is recommended (Gardner et al. 2007).

The original English language version of the PSC has been validated against various criterion standards, including longer diagnostic questionnaires and extensive diagnostic interviews (Jellinek et al. 1995, Anderson et al. 1999, Simonian and Tarnowski 2001, Gardner, Kelleher and Pajer 2002, Gardner et al. 2007). Prior studies have suggested that similar populations are identified when responses to single question screens are compared with extensive diagnostic interviews (Pagano et al. 1996, Cuijpers and Smit 2001, Miller et al. 2008). Thus, while single question screens are less sensitive than more comprehensive screening methodologies for the identification of at-risk children, they can be helpful for establishing score cut-offs in new populations for tools previously validated by rigorous criterion standards in other populations.

A single question asking the parent or child to indicate if the child has a problem for which (s) he needs help has been a part of PSC screening for nearly two decades (Murphy et al. 1996, Pagano et al. 1996). It is included for the express purpose of allowing a preliminary exploration of the validity of the screen when it is used in a new population. In an outpatient paediatric sample using the parent rating of whether the child had a problem as the ‘gold standard’, PSC case classifications were in categorical agreement 91% of the time (Hacker et al. 2006). In a school based health centre, the same question was associated with an 86% rate of agreement with the PSC-Y (Gall et al. 2000). In an outpatient child psychiatry clinic where presumably all children had significant problems, 80% of parents answered yes to this question on the PSC (Murphy unpublished data). Recent work by members of our group has also found a high degree of association (88%) between a single question screen and the United State Department of Agriculture’s 18 item Food Security Measure of hunger (Kleinman et al. 2007, Miller et al. 2008).

The PSC has been used in several different cultural contexts and has been translated into other languages including Dutch (Reijneveld et al. 2006), German (Thun-Hohenstein and Herzog 2008), Japanese (Ishizaki 2000) and Spanish (Jutte et al. 2003). Translated versions have been tested in the appropriate ethnic groups with different score cut-offs proving more appropriate in different settings. For the Japanese version of the PSC, a cut-off score of 17 is recommended (Ishizaki 2000); for the German form, the recommended cut-off is 24 (Thun-Hohenstein and Herzog 2008); and for the Dutch version, a cut-off of 25 is recommended (Reijneveld et al. 2006). To date, no reports have been published on use of the PSC on the African continent and its validity as a screening tool for African youth has not been previously demonstrated.

We assessed the test properties of a Setswana version of the PSC in a population of HIV-infected Batswana children and adolescents. The purpose of our study was to determine whether the PSC is likely to contribute to the early detection of psychosocial problems among African children and adolescents in treatment for HIV infection.

Methods

Translation, cultural adaptation and piloting

The PSC and PSC-Y were translated into Setswana by a University of Botswana linguist. Two Batswana paediatric healthcare professionals back-translated the tools into English. A group of eight professionals who are fluent in both English and Setswana and whose work involves mental health support or research related to children and adolescents was then assembled to finalise the wording of the tool. The goal of this group was to ensure that the tools would be easily understood by the local population and that the original meaning of the items was maintained. Finally, the tools were piloted in a group of Batswana children aged 8 to 16 years and further minor modifications were made based on their input. The final English version used is identical to the previously published English version (Jellinek et al. 1988, Jellinek et al. 1990, Jellinek et al. 1991, Pagano et al. 1996, Anderson et al. 1999, Reijneveld et al. 2006). The Setswana version is available for free download at http://www2.massgeneral.org/allpsych/psc/psc_home.htm.

Study population and administration of tools

The PSC and PSC-Y were administered separately to the parent/guardian (parent) of HIV-infected children aged 8 to 16 years and the children themselves in infectious disease clinics in Nyangabgwe Hospital in Francistown and Letsholathebe Hospital in Maun, Botswana. These two public clinics are the largest HIV treatment sites in north-east and north-west Botswana respectively, where both English and Setswana are official national languages. Combined, they treat approximately 9 000 HIV-infected patients of all ages with highly active antiretroviral therapy. The only exclusion criteria for the study were severe cognitive limitations, prohibiting the child from being able to answer simple questions or presentation to the clinic without a parent or guardian who could give consent for study participation. All eligible children presenting for care at the sites during the recruitment period were enrolled. Only Botswana nationals were eligible for treatment in public HIV treatment clinics in Botswana therefore children and adolescents from other countries were not included.

The tools were administered by a research assistant in order to overcome literacy issues and minimise missing data. Administration of each tool typically took 5 to 10 minutes. Subjects completed the tools in whichever language (English or Setswana) they were most comfortable. Demographic information and clinical data were collected for each subject and the children also completed the Children’s Depression Inventory (CDI) (Kovacs 1992) and the Revised Children’s Manifest Anxiety Scale (RCMAS) (Reynolds and Richmond 1978). The CDI and RCMAS went through the same process described above for translation, cultural adaptation and piloting. Parents and clinic staff were also asked to state whether or not they thought each child had emotional or behavioural problems for which they needed help. A single research assistant administered all tests to each subject and parent during a single session. The research assistant was therefore not specifically blinded to the reference standards. However, the research assistant was instructed to ask the questions in a uniform way and did not participate in test scoring.

Data analysis

Cronbach’s alpha was calculated as a measure of the internal consistency of each tool. Cut-off scores for the Setswana versions of the PSC, PSC-Y, PSC-17 and PSC-17-Y were derived using receiver operator characteristic (ROC) analyses (Hanley and McNeil 1982). Areas under the curves (AUC) with standard errors were derived and the sensitivity and specificity with binomial exact 95% confidence intervals were compared at each cut-point. For the PSC and PSC-17, ROC curves were generated by comparing scores to the dichotomised parent and clinic staff reports of concern about the child. For the PSC-Y and PSC-17-Y, ROC curves were generated using scores on the CDI and the RCMAS as comparators. Test scores were compared by age and sex using two-sample t-tests. The per cent of subjects who achieved each cut-off score was determined and Pearson’s chi-square test was used to evaluate for differences between demographic groups: boys vs. girls, older vs. younger children (≥10 years vs. < 10 years), higher vs. lower school level (≥ standard 5 vs. < standard 5), at appropriate school grade level vs. >2 years behind, orphans vs. non-orphans. All analyses were done using Stata 11 (StataCorp LP, College Station, TX).

Ethical approvals

The study was approved by the Botswana Health Research Development Committee Ethics Board, the Institutional Review Boards (IRBs) at the University of Pennsylvania and Children’s Hospital of Philadelphia, and by the appropriate personnel at all recruitment sites.

Results

Sample

Five hundred and nine child-parent dyads completed the study, 455 from Francistown and 54 from Maun. Demographic data for the enrolled children are presented in Table 1. About half (50.2%) of subjects with pre-HAART CD4 counts available (N = 478) had a nadir CD4 count <200 cells/mm3 or CD4 per cent <15%. Seventy-seven per cent of the children and 95% of those aged 12 or older had been informed of their HIV+ status. Of the subjects over the age of 12, 25% had taken part in a support group for HIV-infected teens. Only eight subjects had a previous psychiatric (five with depression) or substance abuse-related (three) diagnosis.

Table 1.

Demographics (N = 509)

Age Year
 Median (IQR) 12.0 (10.2–13.8))
 Percent >10 years 78.4
Sex Number (%)
 Female 267 (52.5)
Suspected mode of HIV-Infection
 Perinatal 458 (89.9)
 Breastfeeding 2 (0.4)
 Abuse 1 (0.2)
 Other/Unknown 48 (9.4)
Orphan status
 Double orphans 73 (14.4)
 Single orphans 200 (39.4)
 Both parents living 234 (46.2)
Living in orphanage
16 (3.1)
Enrolled in school
 Yes 503 (98.8%)
Last grade in school completed
 Standard 1 29 (5.7)
 Standard 2 77 (15.2)
 Standard 3 84 (16.5)
 Standard 4 95 (18.7)
 Standard 5 77 (15.2)
 Standard 6 59 (11.6)
 Standard 7 48 (9.5)
 Form 1 25 (4.9)
 Form 2 11 (2.2)
 Form 3 1 (0.2)
 Never attended 2 (0.4)
Within 2 years of expected grade
439 (86.6)
Language preferred
 Setswana 498 (97.8)
 English 11 (2.2)
Open in a new tab

PSC

Cronbach’s alpha for the 35 question PSC was 0.87. A total of 11.5% of parents reported that their child needs help for an emotional or behavioural problem. Figure 1a compares the parents’ report of concern to scores on the PSC. The AUC was 0.85 with a standard error of 0.024. Using a cut-off score of 20 on the full-length PSC resulted in a sensitivity of 0.62 (95% confidence interval 0.58–0.66) and a specificity of 0.86 (95% confidence interval 0.83–0.89) for detecting children whose parents expressed concern about their psychosocial health. Table 2 outlines the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of various PSC cut-off scores compared with parental concern. Based on clinic staff report, 11.5% of children were thought to need help for a psychosocial problem. However, only 3.5% of children were of concern to both parents and clinic staff. Clinic staff concern discriminated relatively poorly with PSC scores, having an AUC of 0.70 (95% confidence interval 0.65–0.74).

Figure 1.

Figure 1

Open in a new tab

ROC analysis of PSC vs. parent/caregiver report of psychosocial concerns

Table 2.

PSC cut-off scores compared with parental report of concern

PSC
cut-off score		 Per cent
positive		 Sensitivity Specificity PPV NPV
15 35 0.80 0.71 0.27 0.97
20 19.5 0.62 0.86 0.36 0.95
22 15.9 0.55 0.89 0.39 0.94
25 11 0.47 0.93 0.47 0.93
28   8.5 0.36 0.95 0.49 0.92
Open in a new tab

With scores of 20 or greater considered positive, 19.5% of children in our study were identified as at risk. Using a cut-off score of 28, only 8% of children would have tested positive. The latter cut-off score had a sensitivity of only 0.36 (95% confidence interval 0.32–0.40) for identifying children whose parents have concerns about their psychosocial health. The mean score on the PSC was 12.7 (SD = 9.1; range 0–58). Scores did not vary significantly by sex (p = 0.32). The mean score for children under the age of 10 years was less than one point different from the mean score for children >10 years of age: 12.0 vs. 12.9, although this difference was statistically significant (p < 0.01). There was no significant difference in the proportion of subjects classified as testing positive when scores were stratified by sex (p > 0.5), age (p = 0.14), grade in school (p > 0.5), being within 2 years of their expected grade level (p > 0.5), or orphan status (p > 0.5). The highest 5th, 10th, 15th, 20th and 25th percentile scores were 31, 25, 22, 19 and 17 respectively.

PSC-17

Cronbach’s alpha for the PSC-17 was 0.80. Figure 1b compares the parents’ report of concern to scores on the PSC-17. The AUC was 0.84 with a standard error of 0.025. Using a cut-off score of 15 to indicate positivity on the PSC-17, only 5.5% of our population would have been considered to have tested positive and only 25% of children whose parents thought they needed help for emotional or behavioural problems would have tested positive. Using a cut-off score of 10 on the PSC-17 resulted in a sensitivity of 0.60 (95% confidence interval 0.56–0.64) and a specificity of 0.88 (95% confidence interval 0.85–0.91) for detecting children whose parents expressed concern about their psychosocial health. A score of 10 or greater on the PSC-17 correctly identified most subjects who would have been identified as positive using a cut-off of 20 or greater on the full-length PSC (PPV = 0.92). A score of <10 on the PSC-17 correctly excluded most subjects who would have been excluded as negative using a cut-off of <20 on the full-length PSC (NPV = 0.95). The mean score on the PSC-17 was 4.5 (SD = 4.5; range 0–24). The difference in mean scores for males vs. females was not clinically significant (5 vs. 4.6, although it was statistically significant, p < 0.01). The mean score for children under the age of 10 years was also less than half a point different from the mean score for children >10 years of age (4.5 vs. 4.9, p < 0.01). There was no significant difference in the proportion of subjects classified as testing positive when scores were stratified by sex (p > 0.5), age (p > 0.5), grade in school (p > 0.5), being within 2 years of their expected grade level (p > 0.5), or orphan status (p = 0.47). The highest 5th, 10th, 15th, 20th and 25th percentile scores were 15, 12, 10, 8 and 7 respectively.

PSC-Y and PSC-Y-17

Like the PSC and the PSC-17, the youth self-report versions of these tests had high internal consistency. Cronbach’s alpha for the 35 question PSC-Y was 0.86. For the PSC-17-Y, Cronbach’s alpha was 0.78. For comparisons of the PSC-Y to the CDI, a cut-off score of 19 was considered positive on the CDI (Kovacs 1992). A total of 7.7% of our subjects met this threshold for severe depressive symptoms. Figure 2a compares the CDI scores (<19 vs. ≥19) to scores on the PSC-Y. The AUC was 0.81 with a standard error of 0.043. Using a cut-off score of 20 as positive on the PSC-Y had a sensitivity of 0.64 (95% confidence interval 0.60–0.68) and a specificity of 0.88 (95% confidence interval 0.85–0.91) for detecting children with a positive CDI. A cut-off score on the PSC of 28 had poor sensitivity (0.36; 95% confidence interval 0.32–0.40) for detecting children with a CDI score ≥19. Table 3 outlines the sensitivity, specificity, positive predictive value (PPV) and negative predictive value (NPV) of various PSC cut-off scores compared with CDI scores dichotomised at a score of 19.

Figure 2.

Figure 2

Open in a new tab

ROC analysis of PSC-Y vs. Children’s Depression Inventory Score (<19 vs. ≥19)

Table 3.

PSC-Y cut-off scores compared with high (≥19) CDI scores

PSC
cut-off score		 Per
cent positive		 Sensitivity Specificity PPV NPV
15 27.7 0.74 0.63 0.21 0.97
20 15.9 0.64 0.88 0.31 0.97
22 12.4 0.59 0.92 0.37 0.96
25 9.6 0.49 0.94 0.39 0.96
28 7.3 0.36 0.95 0.38 0.95
Open in a new tab

For the comparison of the PSC-17-Y to CDI scores (<19 vs. ≥19), the AUC was 0.82 with a standard error of 0.040. Figure 2b compares the CDI scores (<19 vs. ≥19) to scores on the PSC-17-Y. Using a cut-off score of 9 on the PSC-17-Y had a sensitivity of 0.67 (95% confidence interval 0.63–0.71) and a specificity of 0.90 (95% confidence interval 0.87–0.93) for detecting children with a positive CDI. This cut-off compared favourably with cut-offs of 10 or 15 which had sensitivities of 0.55 and 0.31 respectively for detecting children with a positive CDI. A score of ≥9 on the PSC-17-Y correctly identified most subjects who would have been identified as positive using a cut-off of ≥20 on the full-length PSC (PPV = 0.89). A score of <10 on the PSC-17-Y correctly excluded most subjects who would have been excluded as negative using a cut-off of <20 on the full-length PSC-Y (NPV = 0.96).

The highest 5th, 10th, 15th, 20th and 25th percentile scores on the PSC-Y were 31, 24, 20, 17 and 15 respectively. The highest 5th, 10th, 15th, 20th and 25th percentile scores on the PSC-17-Y were 14, 10, 8, 7 and 6 respectively. The difference in mean scores on the PSC-Y for males vs. females was not clinically significant (11.1 vs. 11.8, p<0.01). The difference in mean scores on the PSC-17-Y for males vs. females was also not clinically significant (4.31 vs. 4.36, although it too was statistically significantly different, p = 0.05). The mean score on the PSC-17-Y for children <10 years of age was 3.9 and that for children ≥10 years of age was 4.5 (p < 0.01). There was no significant difference in the proportion of subjects classified as testing positive when scores on the PSC-Y and PSC-17-Y were stratified by sex, age, grade in school or orphan status (p > 0.5 for all comparisons). Children who were at least two grade levels below their expected grade level were significantly more likely to have positive scores on the PSC-17-Y (p = 0.02), but not on the PSC-Y (p = 0.14).

Using the recommended cut-off score of 19 as positive on the RCMAS (Reynolds et al. 1978), 42% of subjects met the threshold for high anxiety symptom scores. The top 10% of scores on the RCMAS were >25. The AUC for the ROC curve comparing the top 10% of scores on the RCMAS to PSC-Y values was 0.81 with a standard error of 0.028. The AUC for the ROC curve comparing the top 10% of scores on the RCMAS to PSC-17-Y values was 0.78 with a standard error of 0.029. The PSC-Y and the PSC-17-Y have poor sensitivity for detecting children with the top 10% of scores on the RCMAS. Only 51% of subjects with an RCMAS score >25 also had a PSC score >20. Only 39% of subjects with an RCMAS score >25 also had a PSC-17-Y score >10.

Discussion

We have demonstrated that the Setswana versions of the PSC and PSC-Y are promising tools for screening HIV-infected children in Botswana for psychosocial dysfunction. Our subjects all completed the 35 question tools and our analyses were repeated limiting the data to only the questions included in the PSC-17 (Gall et al. 2000, Duke et al. 2005, Gardner et al. 2007) and PSC-17-Y. All four tools have strong internal consistency. The ROC analysis indicates that high scores on the PSC and PSC-17 help to discriminate between children with and without parent report of concern and identify some children not reported as being of concern based on a single-question screen. The PSC-Y and PSC-Y-17 show similarly good agreement with high scores on the CDI. Agreement with the RCMAS was less strong, not an unexpected finding given that only a single question on the PSC specifically addresses anxiety and prior studies have shown only moderate agreement between PSC scores and anxiety in children (DeMaso et al. 2000, Wren et al. 2004).

Cut-off score of 20 on the PSC and the PSC-Y and cut-off scores of 10 and 9 on the PSC-17 and PSC-17-Y identify 15–20% of children in our population as being at risk. In prior studies utilizing the PSC, between about 5% and 25% of children tested positive. Higher percentages of children test positive in populations which include children living in poverty and those with chronic illnesses (Anderson et al. 1999, Murphy et al. 1992, Navon et al. 2001, Stoppelbein et al. 2005). The sensitivity and specificity of the above thresholds for detecting children with suspected psychosocial dysfunction are similar to those reported previously for other translations of the PSC (Jellinek et al. 1986, Jellinek et al. 1988, Reijneveld et al. 2006).

We recommend that the scoring thresholds discussed above be considered for use in screening HIV+ Batswana children for psychosocial dysfunction. Our data and prior studies using the PSC suggest that most children scoring above these thresholds will have problems for which further evaluation and treatment would be of benefit. In some resource-limited settings, different thresholds might need to be chosen, taking into account the number of children the site has the capacity to screen further and treat for psychosocial problems. For example, a large clinic with a single child psychologist available to evaluate and treat children with severe psychosocial problems might decide to begin by screening all children who had not previously been referred to the psychologist and referring those with scores in the expected highest 5%.

The amount of time needed to complete the 35 question PSC and PSC-Y was only 5 to 10 minutes. However, our data suggest that the shorter PSC-17 and PSC-17-Y perform similarly to the longer versions of the tools and could reasonably be substituted for initial screening of Setswana-speaking children. Many of our subjects required assistance to complete the PSC due to literacy issues. Since some clinic staff time may be required for administration in busy clinical settings, use of the shorter version may be desirable.

There were several limitations to our study. We did not evaluate test-retest validity in this population. Because our study was limited to children between the ages of 8 and 16 years, we cannot comment on how well the tool would work in older and younger children. Likewise, since our population was limited to Batswana at two clinical sites and mainly in a single city, we cannot comment on the generalisability of the findings outside of these populations. Furthermore, since we did not do in-depth diagnostic interviews and studies to determine the per cent of subjects with high scores on the PSC who have true psychosocial problems, we cannot comment on how many children with psychosocial problems are likely to be missed by PSC screening alone. It was infeasible for us to perform in-depth assessments of psychosocial dysfunction. While there is precedent for use of a single question response (Pagano et al. 1996, Cuijpers and Smit 2001, Miller et al. 2008), in-depth assessments would have been a preferred ‘gold standard’.

Further studies are needed to determine what percentage of children with high scores on the Setswana PSC, PSC-17, PSC-Y and PSC-17-Y would be diagnosed with psychological problems using in-depth evaluations. The types of problems and their prevalence among HIV+ children and adolescents in Africa need to be elucidated to allow for targeted allocation of social support resources within HIV treatment programmes. If anxiety is found to be an important contributor to psychosocial morbidity among HIV+ children and adolescents, then a simple screening tool for anxiety should be considered as a supplement to the PSC for routine screening of HIV+ children and adolescents for psychosocial dysfunction.

We believe that identification of HIV+ children and adolescents with psychosocial problems is critical for reducing morbidity and mortality. Numerous studies have demonstrated that failure to recognize and treat psychosocial problems can lead to more severe impairment with increased utilisation of health care services and increased cost to society (Koot and Verhulst 1992, Visser et al. 1999, Moffitt and Caspi 2001). For children and adolescents with HIV, psychosocial problems can increase their risk of HIV treatment failure (Mellins et al. 2004, DeLaMora et al. 2006, Williams et al. 2006, Rudy et al. 2009, Hazra et al. 2010). Successful treatment of mental disorders in low and middle income settings has been demonstrated (Patel et al. 2007). Recognising those children at highest risk during their routine clinic visits by using a simple tool such as the PSC can allow for targeting psychosocial support resources to improve both mental health and HIV treatment outcomes.

References

  1. Anderson DL, Spratt EG, Macias MM, Jellinek MS, Murphy JM, Pagano M, Griesemer DA, Holden KR, Barbosa E. Use of the pediatric symptom checklist in the pediatric neurology population. Pediatric Neurology. 1999;20:116–120. doi: 10.1016/s0887-8994(98)00121-0. [DOI] [PubMed] [Google Scholar]
  2. Bach JM, Louw D. Depression and exposure to violence among Venda and Northern Sotho adolescents in South Africa. African Journal of Psychiatry. 2010;13:25–35. doi: 10.4314/ajpsy.v13i1.53426. [DOI] [PubMed] [Google Scholar]
  3. Bauman LJ, Silver EJ, Draimin BH, Hudis J. Children of mothers with HIV/AIDS: unmet needs for mental 26 health services. Pediatrics. 2007;120:e1141–e1147. doi: 10.1542/peds.2005-2680. [DOI] [PubMed] [Google Scholar]
  4. Bolton-Moore C, Mubiana-Mbewe M, Cantrell RA, Chintu N, Stringer EM, Chi BH, Sinkala M, Kankasa C, Wilson CM, Wilfert CM, Mwango A, Levy J, Abrams EJ, Bulterys M, Stringer JS. Clinical outcomes and CD4 cell response in children receiving antiretroviral therapy at primary health care facilities in Zambia. Journal of the American Medical Association. 2007;298:1888–1899. doi: 10.1001/jama.298.16.1888. [DOI] [PubMed] [Google Scholar]
  5. Borowsky IW, Mozayeny S, Ireland M. Brief psychosocial screening at health supervision and acute care visits. Pediatrics. 2003;112:129–133. doi: 10.1542/peds.112.1.129. [DOI] [PubMed] [Google Scholar]
  6. Chambers WJ, Puig-Antich J, Hirsch M, Paez P, Ambrosini PJ, Tabrizi MA, Davies M. The assessment of affective disorders in children and adolescents by semistructured interview. Test-retest reliability of the schedule for affective disorders and schizophrenia for school-age children, present episode version. Archives of General Psychiatry. 1985;42:696–702. doi: 10.1001/archpsyc.1985.01790300064008. [DOI] [PubMed] [Google Scholar]
  7. Cluver L, Gardner F, Operario D. Psychological distress amongst AIDS-orphaned children in urban South Africa. Journal of Child Psychology and Psychiatry. 2007;48:755–763. doi: 10.1111/j.1469-7610.2007.01757.x. [DOI] [PubMed] [Google Scholar]
  8. Cluver L, Orkin M. Cumulative risk and AIDS-orphanhood: interactions of stigma, bullying and poverty on child mental health in South Africa. Social Science Medicine. 2009;69:1186–1193. doi: 10.1016/j.socscimed.2009.07.033. [DOI] [PubMed] [Google Scholar]
  9. Conner C. Rating scales. In: Noshpitz J, editor. Basic handbook of child psychiatry. New York: Basic Books; 1979. pp. 675–685. [Google Scholar]
  10. Costello EJ, Edelbrock C, Costello AJ, Dulcan MK, Burns BJ, Brent D. Psychopathology in pediatric primary care: the new hidden morbidity. Pediatrics. 1988;82:415–424. [PubMed] [Google Scholar]
  11. Cuijpers P, Smit F. Assessing parental alcoholism: a comparison of the family history research diagnostic criteria versus a single-question method. Addiction Behavior. 2001;26:741–748. doi: 10.1016/s0306-4603(00)00155-6. [DOI] [PubMed] [Google Scholar]
  12. DeLaMora P, Aledort N, Stavola J. Caring for adolescents with HIV. Current HIV/AIDS Reports. 2006;3:74–78. doi: 10.1007/s11904-006-0021-2. [DOI] [PubMed] [Google Scholar]
  13. DeMaso DR, Spratt EG, Vaughan BL, D’Angelo EJ, Van der Feen JR, Walsh E. Psychological functioning in children and adolescents undergoing radiofrequency catheter ablation. Psychosomatics. 2000;41:134–139. doi: 10.1176/appi.psy.41.2.134. [DOI] [PubMed] [Google Scholar]
  14. Duke N, Ireland M, Borowsky IW. Identifying psychosocial problems among youth: factors associated with youth agreement on a positive parent-completed PSC-17. Child: Care, Health and Development. 2005;31:563–573. doi: 10.1111/j.1365-2214.2005.00551.x. [DOI] [PubMed] [Google Scholar]
  15. Eisert DC, Sturner RA, Mabe PA. Questionnaires in behavioral pediatrics: guidelines for selection and use. Journal of Developmental and Behavioral Pediatrics. 1991;12:42–50. [PubMed] [Google Scholar]
  16. Fielden SJ, Sheckter L, Chapman GE, Alimenti A, Forbes JC, Sheps S, Cadell S, Frankish JC. Growing up: perspectives of children, families and service providers regarding the needs of older children with perinatally-acquired HIV. AIDS Care. 2006;18:1050–1053. doi: 10.1080/09540120600581460. [DOI] [PubMed] [Google Scholar]
  17. Gall G, Pagano ME, Desmond MS, Perrin JM, Murphy JM. Utility of psychosocial screening at a school-based health center. Journal of School Health. 2000;70:292–298. doi: 10.1111/j.1746-1561.2000.tb07254.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Gardner W, Kelleher KJ, Pajer KA. Multidimensional adaptive testing for mental health problems in primary care. Medical Care. 2002;40:812–823. doi: 10.1097/00005650-200209000-00010. [DOI] [PubMed] [Google Scholar]
  19. Gardner W, Lucas A, Kolko DJ, Campo JV. Comparison of the PSC-17 and alternative mental health screens in an at-risk primary care sample. Journal of the American Academy of Child and Adolescent Psychiatry. 2007;46:611–618. doi: 10.1097/chi.0b013e318032384b. [DOI] [PubMed] [Google Scholar]
  20. Hacker KA, Myagmarjav E, Harris V, Suglia SF, Weidner D, Link D. Mental health screening in pediatric practice: factors related to positive screens and the contribution of parental/personal concern. Pediatrics. 2006;118:1896–1906. doi: 10.1542/peds.2006-0026. [DOI] [PubMed] [Google Scholar]
  21. Hanley JA, McNeil BJ. The meaning and use of the area under a receiver operating characteristic (ROC) curve. Radiology. 1982;143:29–36. doi: 10.1148/radiology.143.1.7063747. [DOI] [PubMed] [Google Scholar]
  22. Hazra R, Siberry GK, Mofenson LM. Growing up with HIV: children, adolescents, and young adults with perinatally acquired HIV infection. Annual Review of Medicine. 2010;61:169–185. doi: 10.1146/annurev.med.050108.151127. [DOI] [PubMed] [Google Scholar]
  23. Herjanic B, Reich W. Development of a structured psychiatric interview for children: agreement between child and parent on individual symptoms. Journal of Abnormal Child Psychology. 1982;10:307–324. doi: 10.1007/BF00912324. [DOI] [PubMed] [Google Scholar]
  24. Ishizaki Y, Fukai Y, Kobayashi Y, Ozawa K. Validation and cutoff score of the Japanese version of the Pediatric Symptom Checklist: Screening of school-aged children with psychosocial and psychosomatic disorders (in Japanese) The Journal of the Japan Pediatric Society. 2000;104:831–840. [Google Scholar]
  25. Jellinek M, Little M, Murphy JM, Pagano M. The Pediatric Symptom Checklist. Support for a role in a managed care environment. Archives of Pediatric and Adolescent Medicine. 1995;149:740–746. doi: 10.1001/archpedi.1995.02170200030004. [DOI] [PubMed] [Google Scholar]
  26. Jellinek MS, Murphy JM, Burns BJ. Brief psychosocial screening in outpatient pediatric practice. Journal of Pediatrics. 1986;109:371–378. doi: 10.1016/s0022-3476(86)80408-5. [DOI] [PubMed] [Google Scholar]
  27. Jellinek MS, Murphy JM. Screening for psychosocial disorders in pediatric practice. American. Journal of Diseases of Children. 1988;142:1153–1157. doi: 10.1001/archpedi.1988.02150110031013. [DOI] [PubMed] [Google Scholar]
  28. Jellinek MS, Murphy JM, Robinson J, Feins A, Lamb S, Fenton T. Pediatric Symptom Checklist: screening school-age children for psychosocial dysfunction. Journal of Pediatrics. 1988;112:201–209. doi: 10.1016/s0022-3476(88)80056-8. [DOI] [PubMed] [Google Scholar]
  29. Jellinek MS, Murphy JM. The recognition of psychosocial disorders in pediatric office practice: the current status of the pediatric symptom checklist. Journal of Developmental and Behavioral Pediatrics. 1990;11:273–278. [PubMed] [Google Scholar]
  30. Jellinek MS, Bishop SJ, Murphy JM, Biederman J, Rosenbaum JF. Screening for dysfunction in the children of outpatients at a psychopharmacology clinic. American Journal of Psychiatry. 1991;148:1031–1036. doi: 10.1176/ajp.148.8.1031. [DOI] [PubMed] [Google Scholar]
  31. Jellinek MS, Murphy JM, Little M, Pagano ME, Comer DM, Kelleher KJ. Use of the Pediatric Symptom Checklist to screen for psychosocial problems in pediatric primary care: a national feasibility study. Archives of Pediatric and Adolescent Medicine. 1999;153:254–260. doi: 10.1001/archpedi.153.3.254. [DOI] [PMC free article] [PubMed] [Google Scholar]
  32. Jutte DP, Burgos A, Mendoza F, Ford CB, Huffman LC. Use of the Pediatric Symptom Checklist in a low-income, Mexican American population. Archives of Pediatric and Adolescent Medicine. 2003;157:1169–1176. doi: 10.1001/archpedi.157.12.1169. [DOI] [PubMed] [Google Scholar]
  33. Kelleher KJ, Wolraich ML. Diagnosing psychosocial problems. Pediatrics. 1996;97:899–901. [PubMed] [Google Scholar]
  34. Kleinman RE, Murphy JM, Wieneke KM, Desmond MS, Schiff A, Gapinski JA. Use of a single-question screening tool to detect hunger in families attending a neighborhood health center. Ambulatory Pediatrics. 2007;7:278–284. doi: 10.1016/j.ambp.2007.03.005. [DOI] [PubMed] [Google Scholar]
  35. Kline MW, Matusa RF, Copaciu L, Calles NR, Kline NE, Schwarzwald HL. Comprehensive pediatric human immunodeficiency virus care and treatment in Constanta, Romania: implementation of a program of highly active antiretroviral therapy in a resource-poor setting. Pediatric Infectious Disease Journal. 2004;23:695–700. doi: 10.1097/01.inf.0000135454.46188.83. [DOI] [PubMed] [Google Scholar]
  36. Koot HM, Verhulst FC. Prediction of children’s referral to mental health and special education services from earlier adjustment. Journal of Child Psychology and Psychiatry. 1992;33:717–729. doi: 10.1111/j.1469-7610.1992.tb00907.x. [DOI] [PubMed] [Google Scholar]
  37. Kovacs M. Children’s Depression Inventory. Niagara Falls, NY: Multi-health Systems; 1992. [Google Scholar]
  38. Lee SJ, Detels R, Rotheram-Borus MJ, Duan N. The effect of social support on mental and behavioral outcomes among adolescents with parents with HIV/AIDS. American Journal of Public Health. 2007;97:1820–1826. doi: 10.2105/AJPH.2005.084871. [DOI] [PMC free article] [PubMed] [Google Scholar]
  39. Mellins CA, Brackis-Cott E, Dolezal C, Abrams EJ. The role of psychosocial and family factors in adherence to antiretroviral treatment in human immunodeficiency virus-infected children. Pediatric Infectious Disease Journal. 2004;23:1035–1041. doi: 10.1097/01.inf.0000143646.15240.ac. [DOI] [PubMed] [Google Scholar]
  40. Mellins CA, Brackis-Cott E, Dolezal C, Leu CS, Valentin C, Meyer-Bahlburg HF. Mental health of early adolescents from high-risk neighborhoods: the role of maternal HIV and other contextual, self-regulation, and family factors. Journal of Pediatric Psychology. 2008;33:1065–1075. doi: 10.1093/jpepsy/jsn004. [DOI] [PMC free article] [PubMed] [Google Scholar]
  41. Miller E, Wieneke KM, Murphy JM, Desmond S, Schiff A, Canenguez KM, Kleinman RE. Child and parental poor health among families at risk for hunger attending a community health center. Journal of Health Care for the Poor and Underserved. 2008;19:550–561. doi: 10.1353/hpu.0.0008. [DOI] [PubMed] [Google Scholar]
  42. Moffitt TE, Caspi A. Childhood predictors differentiate life-course persistent and adolescence-limited antisocial pathways among males and females. Developmental Psychopathology. 2001;13:355–375. doi: 10.1017/s0954579401002097. [DOI] [PubMed] [Google Scholar]
  43. Murphy JM, Reede J, Jellinek MS, Bishop SJ. Screening for psychosocial dysfunction in inner-city children: further validation of the Pediatric Symptom checklist. Journal of the American Academy of Child and Adolescent Psychiatry. 1992;31:1105–1111. doi: 10.1097/00004583-199211000-00019. [DOI] [PubMed] [Google Scholar]
  44. Murphy JM, Ichinose C, Hicks RC, Kingdon D, Crist-Whitzel J, Jordan P, Feldman G, Jellinek MS. Utility of the Pediatric Symptom Checklist as a psychosocial screen to meet the federal Early and Periodic Screening, Diagnosis, and Treatment (EPSDT) standards: a pilot study. Journal of Pediatrics. 1996;129:864–869. doi: 10.1016/s0022-3476(96)70030-6. [DOI] [PubMed] [Google Scholar]
  45. Navon M, Nelson D, Pagano M, Murphy M. Use of the pediatric symptom checklist in strategies to improve preventive behavioral health care. Psychiatric Services. 2001;52:800–804. doi: 10.1176/appi.ps.52.6.800. [DOI] [PMC free article] [PubMed] [Google Scholar]
  46. Ndetei DMKL, Mutiso VN, Mbwayo A. Recognition of depression in children in general hospital-based paediatric units in Kenya: practice and policy implications. Annals of General Psychiatry. 2009;28:25–30. doi: 10.1186/1744-859X-8-25. [DOI] [PMC free article] [PubMed] [Google Scholar]
  47. Pagano M, Murphy JM, Pedersen M, Mosbacher D, Crist-Whitzel J, Jordan P, Rodas C, Jellinek MS. Screening for psychosocial problems in 4–5-year-olds during routine EPSDT examinations: validity and reliability in a Mexican-American sample. Clinical Pediatrics. 1996;35:139–146. doi: 10.1177/000992289603500305. [DOI] [PMC free article] [PubMed] [Google Scholar]
  48. Pao M, Lyon M, D’Angelo LJ, Schuman WB, Tipnis T, Mrazek DA. Psychiatric diagnoses in adolescents seropositive for the human immunodeficiency virus. Archives of Pediatric and Adolescent Medicine. 2000;154:240–244. doi: 10.1001/archpedi.154.3.240. [DOI] [PubMed] [Google Scholar]
  49. Patel V, Araya R, Chatterjee S, Chisholm D, Cohen A, De Silva M, Hosman C, McGuire H, Rojas G, van Ommeren M. Treatment and prevention of mental disorders in low-income and middle-income countries. Lancet. 2007;370:991–1005. doi: 10.1016/S0140-6736(07)61240-9. [DOI] [PubMed] [Google Scholar]
  50. Pela OA, Reynolds CR. Cross-cultural application of the Revised-Children’s Manifest Anxiety Scale: normative and reliability data for Nigerian primary school children. Psychological Reports. 1982;51:1135–1138. doi: 10.2466/pr0.1982.51.3f.1135. [DOI] [PubMed] [Google Scholar]
  51. Prince M, Patel V, Saxena S, Maj M, Maselko J, Phillips MR, Rahman A. No health without mental health. Lancet. 2007;370:859–877. doi: 10.1016/S0140-6736(07)61238-0. [DOI] [PubMed] [Google Scholar]
  52. Rao R, Sagar R, Kabra SK, Lodha R. Psychiatric morbidity in HIV-infected children. AIDS Care. 2007;19:828–833. doi: 10.1080/09540120601133659. [DOI] [PubMed] [Google Scholar]
  53. Reijneveld SA, Vogels AG, Hoekstra F, Crone MR. Use of the Pediatric Symptom Checklist for the detection of psychosocial problems in preventive child healthcare. BMC Public Health. 2006;6:197. doi: 10.1186/1471-2458-6-197. [DOI] [PMC free article] [PubMed] [Google Scholar]
  54. Reynolds CR, Richmond BO. What I think and feel: a revised measure of children’s manifest anxiety. Journal of Abnormal Child Psychology. 1978;6:271–280. doi: 10.1007/BF00919131. [DOI] [PubMed] [Google Scholar]
  55. Rudy BJ, Murphy DA, Harris DR, Muenz L, Ellen J Adolescent Trials Network for HIV/AIDS Interventions. Patient-related risks for nonadherence to antiretroviral therapy among HIV-infected youth in the United States: A study of prevalence and interactions. AIDS Patient Care and STDS. 2009;23:185–194. doi: 10.1089/apc.2008.0162. [DOI] [PMC free article] [PubMed] [Google Scholar]
  56. Ruiz-Casares M, Thombs BD, Rousseau C. The association between single and double orphanhood with symptoms of depression among children and adolescent in Namibia. European Child and Adolescent Psychiatry. 2009;18:369–376. doi: 10.1007/s00787-009-0739-7. [DOI] [PubMed] [Google Scholar]
  57. Saini A, Chandra J, Goswami U, Singh V, Dutta AK. Case control study of psychosocial morbidity in beta thalassemia major. Journal of Pediatrics. 2007;150:516–520. doi: 10.1016/j.jpeds.2007.01.025. [DOI] [PubMed] [Google Scholar]
  58. Scharko AM. DSM psychiatric disorders in the context of pediatric HIV/AIDS. AIDS Care. 2006;18:441–445. doi: 10.1080/09540120500213487. [DOI] [PubMed] [Google Scholar]
  59. Simonian SJ, Tarnowski KJ. Utility of the Pediatric Symptom Checklist for behavioral screening of disadvantaged children. Child Psychiatry and Human Development. 2001;31:269–278. doi: 10.1023/a:1010213221811. [DOI] [PubMed] [Google Scholar]
  60. Stoppelbein L, Greening L, Jordan SS, Elkin TD, Moll G, Pullen J. Factor analysis of the Pediatric Symptom Checklist with a chronically ill pediatric population. Journal of Developmental and Behavioral Pediatrics. 2005;26:349–355. doi: 10.1097/00004703-200510000-00002. [DOI] [PubMed] [Google Scholar]
  61. Stover J, Fidzani B, Molomo BC, Moeti T, Musuka G. Estimated HIV trends and program effects in Botswana. PLoS One. 2008;3:e3729. doi: 10.1371/journal.pone.0003729. [DOI] [PMC free article] [PubMed] [Google Scholar]
  62. Sturner RA. Parent questionnaires: basic office equipment? Journal of Developmental and Behavioral Pediatrics. 1991;12:51–54. [PubMed] [Google Scholar]
  63. Thun-Hohenstein L, Herzog S. The predictive value of the pediatric symptom checklist in 5-year-old Austrian children. European Journal of Pediatrics. 2008;167:323–329. doi: 10.1007/s00431-007-0494-z. [DOI] [PubMed] [Google Scholar]
  64. UNAIDS. (Joint United Nations Programme on HIV/AIDS) Country Spotlight Botswana: 2011. [accessed 23 January 2011]. Available at http://www.unaids.org/en/regionscountries/countries/botswana/. [Google Scholar]
  65. UNICEF (United Nations Children’s Fund) Children and AIDS: 5th Stocktaking Report. Geneva: UNICEF; 2010. [Google Scholar]
  66. Visser JH, van der Ende J, Koot HM, Verhulst FC. Continuity of psychopathology in youths referred to mental health services. Journal of the American Academy of Child and Adolescent Psychiatry. 1999;38:1560–1568. doi: 10.1097/00004583-199912000-00018. [DOI] [PubMed] [Google Scholar]
  67. Walker WO, Jr, LaGrone RG, Atkinson AW. Psychosocial screening in pediatric practice: identifying high-risk children. Journal of Developmental and Behavioral Pediatrics. 1989;10:134–138. [PubMed] [Google Scholar]
  68. Weissman MM, Orvaschel H, Padian N. Children’s symptom and social functioning self-report scales. Comparison of mothers’ and children’s reports. The Journal of Nervous and Mental Disease. 1980;168:736–740. doi: 10.1097/00005053-198012000-00005. [DOI] [PubMed] [Google Scholar]
  69. Williams PL, Storm D, Montepiedra G, Nichols S, Kammerer B, Sirois PA, Farley J, Malee K PACTG 219C Team. Predictors of adherence to antiretroviral medications in children and adolescents with HIV infection. Pediatrics. 2006;118:e1745–e1757. doi: 10.1542/peds.2006-0493. [DOI] [PubMed] [Google Scholar]
  70. Wren FJ, Bridge JA, Birmaher B. Screening for childhood anxiety symptoms in primary care: integrating child and parent reports. Journal of the American Academy of Child and Adolescent Psychiatry. 2004;43:1364–1371. doi: 10.1097/01.chi.0000138350.60487.d3. [DOI] [PubMed] [Google Scholar]

RESOURCES