Brief Psychosocial Screening at Health Supervision and Acute Care Visits

METHODS

The study was conducted at 8 out-patient pediatric practices in the Minneapolis-St Paul metropolitan area. The practices included 2 urban community clinics, 2 urban private practices, and 4 suburban private practices. During defined data collection periods, all consecutive English-speaking parents of children aged 7 to 15 seen for a medical visit were invited to complete the PSC-17 before their visit with the primary care clinician. Research assistants explained the purposes and procedures of the study to the parents. Participation was voluntary. Procedures and consent forms were approved by the Institutional Review Board of the University of Minnesota. Parents also provided demographic information and the reason for their child’s visit to the clinic.

The PSC-17 consists of 17 symptoms that parents rate as “often,” “sometimes,” or “never” present in their child. The PSC-17 is scored by assigning 2, 1, or 0 points, respectively, to these ratings. The points are then added to obtain PSC-17 subscale and total scores. The internalizing subscale consists of 5 items: “feels sad, unhappy,” “feels hopeless,” “is down on self,” “worries a lot,” “seems to be having less fun”; the externalizing subscale consists of 7 items: “fights with other children,” “does not listen to rules,” “does not understand other people’s feelings,” “teases others,” “blames others for his/her troubles,” “refuses to share,” “takes things that do not belong to him/her”; the attention subscale consists of 5 items: “fidgety, unable to sit,” “daydreams too much,” “distracted easily,” “has trouble concentrating,” “acts as if driven by a motor.” A positive score for each of the subscales was as follows: 5 or greater for the internalizing subscale, 7 or greater for the externalizing subscale, and 7 or greater for the attention subscale. A positive score on the screen was defined as at least 1 positive PSC-17 subscale or a PSC-17 total score of ≥15.¹⁶

The data were initially examined using bivariate descriptive statistics. We used c² tests to examine the significance of differences in having a positive or negative score on the screen when data were stratified by other factors, such as patient gender, age, race/ethnicity, family structure, welfare status, and visit type. To assess the overall effect of visit type on scoring positive on the screen after controlling for potential confounding factors listed above, we used a generalized linear models procedure based on generalized estimating equations, PROC GENMOD, which allowed us to control for clustering of participants by pediatric practices. Because the dependent variable was dichotomous, we specified the binomial distribution for the dependent variable in the PROC GENMOD code.

RESULTS

Ninety-one percent of the 2315 parents who were approached agreed to complete the questionnaire. Overall, 22% of youth (451/2028) scored positive on the screen. As shown in Table 1, 250 (12%) of the youth scored positive on the internalizing subscale, 202 (10%) scored positive on the externalizing subscale, and 144 (7%) scored positive on the attention subscale. Nearly half of the 22% of youth scoring positive on the screen (224/451) had a positive PSC-17 total score. Of the 451 youth with a positive score, 307 (68%) had 1 positive subscale, 107 (24%) had 2 positive subscales, and 25 (6%) had 3 positive subscales. The remaining 12 patients (3%) scoring positive had a positive PSC-17 total score, but did not score positive on any of the subscales. Cronbach α coefficient, a measure of the internal consistency of a multi-item scale, ranged from 0.67 to 0.82 for the PSC-17 and its subscales (Table 1).

Demographic and visit-related characteristics of the sample are presented in Table 2. Most screens were completed by the child’s mother. Routine well-care characterized the reason for 46% of visits. The remainder presented for a non-routine visit, such as for illness, follow-up on illness, or injury.

Youth with a positive score on the screen were significantly more likely than youth with a negative score to be male, non-white, not living with both biological parents, and receiving welfare (Table 2). Visit type also differed between those scoring positive and negative on the screen, with youth scoring positive more likely to be presenting to clinic for a non-routine visit than those scoring negative (61% vs 52%, P < .001). Likewise, for each of the subscales and the PSC-17 total, a greater percentage of youth who scored positive presented to clinic for a non-routine visit than those who scored negative. This difference was statistically significant for the PSC-17 total only (63% vs 54%, P = .02) and approached significance for the internalizing subscale (61% vs 54%, P = .07). Although boys were more likely than girls to score positive on the attention subscale (10% vs 4%, P < .001), the aggression subscale (12% vs 9%, P = .025), and the PSC-17 total (13% vs 9%, P = .001), boys and girls were equally likely to have a positive score on the depression subscale (12% of both boys and girls).

We used multivariate analysis to assess the importance of visit type on having a positive score on the screen while controlling for possible confounders (Table 3). After adjusting for patient gender, age, race/ethnicity, family structure, and welfare status, patients presenting for an illness-related or injury visit were more likely to score positive on the screen than those presenting for a routine visit (odds ratio: 1.46; 95% confidence interval: 1.07–1.98).

DISCUSSION

First proposed by Haggerty et al²⁴ in 1975 as the “new morbidity,” psychosocial problems have become the most common chronic condition for pediatric visits.²⁵ In this sample of youth ages 7 to 15, 12% scored positive on the internalizing subscale, 10% on the externalizing subscale, and 7% on the attention subscale. Overall, 11% had a positive PSC-17 total score, and 22% had at least one positive PSC-17 subscale score or a positive total score. These rates are comparable to those found in other studies using parent-completed measures including the PSC-17¹⁶ and the Diagnostic Interview Schedule for Children.²⁶ However, these studies observed a higher rate of externalizing symptoms than internalizing symptoms among the samples of youth studied, whereas more youth scored positive on the internalizing than the externalizing subscale in our study. Other findings that are consistent with previous work include the higher rate of psychosocial problems seen in boys, youth who are not living with both biological parents, and those with a household member receiving public assistance.^13,27 The current study was the first to demonstrate an increased risk of psychosocial dysfunction for youth presenting to clinic for a sick visit as compared with those presenting for a health supervision visit.

It should be noted that a positive score on the PSC-17 is not a diagnosis, but rather an indication for further assessment of the child and family, with subscale scores serving as a guide for the assessment.¹⁶ Although the PSC-17 subscales show good agreement with previously validated parent-report instruments of internalizing, externalizing, and attention problems, they have not yet been validated with confirmatory diagnoses.¹⁶ Case classifications on the 35-item PSC, from which the PSC-17 is derived, have shown good agreement with case classifications on the Children’s Behavior Checklist, clinicians’ Global Assessment Scale ratings of impairment, and the presence of psychiatric disorder in a variety of pediatric and subspecialty settings representing diverse socioeconomic backgrounds.^{11,12,14,15,28,29}

The greater likelihood of scoring positive on the PSC-17 among patients presenting for a sick visit as compared with those presenting for a health supervision visit may be associated with acute care visits for symptoms related to emotional and behavioral problems. Campo et al³⁰ found that children and adolescents with frequent complaints of pain were significantly more likely than unaffected children to show signs of an emotional or behavioral problem based on the PSC-17 subscales and clinician impression. Recurrent pain was also significantly associated with greater health service use and visits to the doctor for medically unexplained symptoms.

Attention to psychosocial and behavioral problems has long been recommended as a critical component of well-child care.^20,24 Guidelines for incorporating screening and assessment of emotional and social development and behavior into health supervision have been published.^17–19 Given the high rates of unrecognized, untreated psychosocial problems in children and adolescents, using every contact with a family, including acute care visits, as an opportunity to screen for psychosocial problems, will likely increase detection of children with emotional and/or behavioral disorders. This strategy has been used successfully to increase immunization rates among children.^31–34 Although immunizations are an essential part of health supervision, acute care visits provide an opportunity to review the child’s or adolescent’s immunization status, emphasize the importance of being appropriately immunized, and encourage catch-up immunization at a current or follow-up visit. Evidence suggests that immunization levels can be improved by up to 30% by eliminating missed opportunities for vaccination.^31,34 Similarly, screening for psychosocial problems at any visit, whether for an acute illness or well care, if not done within the past 6 to 12 months, will increase the chances that annual screening is achieved and take advantage of opportunities to identify problems that would otherwise be missed. In addition, use of a brief psychosocial screen can help limit inefficient use of professional time, guiding pediatricians on which patients to pursue psychological questioning.²⁷

Use of the PSC in pediatric settings has been shown to significantly increase rates of mental health referrals.^15,35 Furthermore, there is evidence that psychosocial screening followed by referral to mental health services for youth identified with psychosocial dysfunction results in improved outcomes.³⁶ Recognition and appropriate management of emotional and behavioral problems in children has been identified as one of the best ways to prevent delinquency, violence, and other high-risk behaviors.^20,37 Effective or promising treatments for many mental disorders exist in children, including cognitive-behavioral therapy and selective serotonin reuptake inhibitors for depression,^38,39 parent training and multisystemic therapy for conduct disorder,⁴⁰ and psychostimulants and behavioral training of teachers for attention-deficit/hyperactivity disorder.^41,42 Improving the diagnosis and treatment of mental health problems among children and adolescents in primary care settings requires addressing important environmental, physician, and patient factors such as: attitudinal and procedural barriers to implementing and sustaining a clinic system for psychosocial screening with the PSC-17 or other tool^35,43; inadequate provider knowledge about mental and behavioral issues and mental health treatment resources in the community; insufficient reimbursement to primary care providers for their efforts in identifying, referring, and treating children with mental health disorders; short supply of mental health services and providers for children and adolescents; fragmented mental health service systems; financial barriers to accessing mental health treatment; and the stigma surrounding mental illness.⁴⁴ Pediatrician training about emotional and behavioral issues, primary care-friendly guides to assessment and diagnosis of psychosocial problems, and strengthening collaborative relationships between pediatricians and mental health professionals will likely improve mental health services in primary care.^25,26,45,46 The evaluation of screening and intervention for mental health problems on outcomes for youth will advance knowledge of effective evidence-based practice in primary care settings. A recent study demonstrated the dearth of research attention and federal funding for understanding the identification and treatment of children’s behavioral and emotional issues within primary medical care settings, despite considerable attention to this issue in the planning documents of many federal agencies.⁴⁷ Given the magnitude and consequences of emotional and behavioral problems in children and adolescents coupled with the feasibility of screening and the potential for effective intervention, recognition of psychosocial problems is an important issue in primary care pediatric practice.