The Pediatric Symptom Checklist-17 (PSC-17; Gardner et al., 1999) is a 17-item parent-report measure of emotional and behavioural problems in children aged 4 to 16. It is a shortened version of the original PSC and is one of the most widely used tools for efficiently assessing overall psychosocial functioning across emotional, behavioural, and attentional domains.
Parents or guardians rate how often specific behaviours or emotional concerns apply to their child, providing a structured snapshot of functioning across key areas of daily life. The PSC-17 is intended as both a screening tool to help clinicians recognise children who may be experiencing difficulties that warrant further evaluation and as a measurement-based care tool to monitor client progress.
Example PSC-17 Item
Children’s psychosocial functioning encompasses emotional, behavioural, and attentional domains that, when impaired, have been found to be associated with poorer academic outcomes, interpersonal difficulties, and increased likelihood of later mental health concerns (Gao et al., 2025). Brief screening tools such as the PSC-17 increase identification rates relative to clinical judgement alone (Murphy et al., 2016) and allow clinicians to triage which children warrant more comprehensive assessment.
In addition to a total score, the PSC-17 includes three subscales:
The PSC-17 is well-suited to primary care, paediatric clinics, schools, and child-welfare settings. It can be administered as a screener at intake or at regular intervals throughout care to support ongoing monitoring. When administered repeatedly, the PSC-17 supports progress monitoring during care, making the measure useful for tracking response to intervention. In practice, it is typically completed prior to or at the start of a visit and reviewed by the clinician to guide follow-up questions, clinical decision-making, and whether additional assessment or referral is indicated.
Its three-subscale structure helps clinicians orient toward likely treatment targets. For example, elevations on Internalising suggest anxiety or mood-focused follow-up, elevations on Attention raise questions about attentional or developmental concerns, and elevations on Externalising point toward conduct or behavioural follow-up. Importantly, positive screens on the PSC-17 indicate that further assessment is warranted, not that a diagnosis is established. A negative screen does not rule out difficulties, particularly when concerns are emerging, underreported, or better captured through alternative informants (e.g., child self-report).
The PSC-17 was derived from the 35-item Pediatric Symptom Checklist (PSC-35; Jellinek et al., 1988) by Gardner and colleagues (1999) using an exploratory factor analysis on a large primary care sample of children aged 4 to 15. Parental reports were collected on 18,045 children seen across the Pediatric Research in Office Settings (PROS) and Ambulatory Sentinel Practice Network (ASPN) practice-based research networks, representing all regions of the United States and parts of Canada. Promax rotation produced three conceptually coherent factors, with 17 items retained from the original 35.
Internal consistency for the PSC-17 is strong across multiple validation samples. In the original derivation, Cronbach’s alpha values were reported as α = 0.89 (Total), α = 0.79 (Internalising), α = 0.83 (Attention), and α = 0.83 (Externalising) (Gardner et al., 1999; N = 18,045 primary care). These values were closely replicated in a subsequent national sample of 80,680 paediatric outpatients (α = 0.87 Total, 0.78 Internalising, 0.82 Attention, 0.80 Externalising; Murphy et al., 2016).
Short-term test-retest reliability was assessed by Murphy and colleagues (2016) using a subsample of 84 children whose parents completed the PSC-17 a second time 8 to 14 days after the first administration. The intraclass correlation for the Total score was strong (ICC = 0.85), with subscale ICCs of 0.76 (Internalising), 0.83 (Attention), and 0.82 (Externalising), supporting the temporal stability of scores over short retest intervals.
The three-factor structure of the PSC-17 has been consistently supported across multiple validation studies. Confirmatory factor analyses by Stoppelbein and colleagues (2012; N = 723) supported the three-factor solution over one-factor and two-factor alternatives (CFI = 0.95, RMSEA = 0.05, SRMSR = 0.06), with strong factor loadings ranging from 0.44 to 0.82 across the primary factor for each item. Murphy and colleagues (2016) confirmed the three-factor model in their large national sample (N = 59,836 in the CFA subsample), reporting moderately strong factor loadings (R² range 0.40 to 0.81). Gao and colleagues (2025) replicated the three-factor solution in a younger parent-rated sample (N = 1,305 children aged 3 to 6; CFI = 0.948, RMSEA = 0.054, loading range 0.45 to 0.89). Meinert and colleagues (2025) reported configural fit indices of CFI = 0.975, TLI = 0.971, RMSEA = 0.044, and SRMR = 0.048 in their sample of 3,596 children, providing support for the three-factor structure in primary care settings.
Criterion validity was established by Gardner and colleagues (1999) using receiver operating characteristic curves against the Iowa Connors Rating Scale (aggression and inattention-overactivity subscales) and the Screen for Child Anxiety Related Emotional Disorders (SCARED). The Total score showed strong agreement with a composite criterion (AUC = 0.88, sensitivity = 0.82, specificity = 0.81), and subscale agreement was similarly strong (Internalising AUC = 0.82, Attention AUC = 0.90, Externalising AUC = 0.87). Stoppelbein and colleagues (2012) provided independent criterion validation against the Child Behavior Checklist (CBCL) using a clinical-range threshold (T-score equal to or greater than 70). Receiver operating characteristic analyses produced high area-under-curve values across all subscales (Total AUC = 0.95, Internalising AUC = 0.88, Attention AUC = 0.91, Externalising AUC = 0.92).
Convergent and discriminant validity were further supported by Jacobson and colleagues (2019) in a sample of 6,492 foster children. The PSC-17 Internalising subscale showed the strongest correlation with the SCARED Anxiety subscale (r = 0.51), with weaker correlations for Attention (r = 0.27) and Externalising (r = 0.19). A similar pattern was found for the SCARED PTSD subscale, with the PSC-17 Internalising subscale showing the strongest correlation (r = 0.44). Point-biserial correlations between PSC-17 subscale scores and concordant psychiatric diagnoses (attention diagnosis with Attention, internalising diagnosis with Internalising, externalising diagnosis with Externalising) were generally higher than correlations with non-concordant diagnoses, supporting the construct-specific interpretation of the subscales.
Sensitivity-to-change evidence for the PSC-17 is well established. Short-term test-retest reliability from Murphy and colleagues (2016) is strong (ICC = 0.85 over 8 to 14 days), supporting use across short retest intervals. Long-term test-retest evidence from Jacobson and colleagues (2019; six-month ICCs of 0.55 Total, 0.41 Internalising, 0.54 Attention, 0.48 Externalising in foster-care children) shows that scores are stable enough across multi-month intervals to support continuous use, while remaining responsive to true change in clinical state. The combination of strong short-term reliability and moderate long-term reliability supports the use of the PSC-17 for both initial screening and progress monitoring.
These reliability values support reliable change indices (RCIs) that identify the smallest change in score unlikely to reflect measurement error alone. The Murphy and colleagues (2016) short-term ICC results in a Total RCI of 6 points, calculated using the Jacobson and Truax (1991) formula. At the subscale level, Kamin and colleagues (2015) established a 2-point RCI for the PSC Internalising subscale, and McCarthy and colleagues (2016) established a 2-point RCI for the PSC Attention subscale. Both studies administered the 35-item PSC, but the 5-item subscales they analysed are derived from Gardner and colleagues (1999) and are identical to the corresponding PSC-17 subscales, so these values transfer directly. Note: Neither paper derived an RCI for the Externalising subscale; the Massachusetts General Hospital PSC institutional page (Murphy & Jellinek, n.d.) extends the 2-point threshold to Externalising as the institutional default, which NovoPsych adopts pending future validation work specific to that subscale.
NovoPsych reports change as the raw score difference between the respondent’s initial (baseline) administration and the current administration, and applies the published reliable-change thresholds in the interpretive text. Two concepts are surfaced: (i) a reliable change is one whose magnitude exceeds the reliable-change threshold; and (ii) a clinically significant change is a reliable change that also crosses the clinical cut-off (Gardner et al., 1999), indicating a change in risk status.
The percentile rankings used in the report are derived from a non-clinical community sample of 322 children (aged 6-16) attending routine paediatric medical appointments, reported by Stoppelbein and colleagues (2012). Average raw scores in this sample are 6.74 for the Total score (SD = 5.62), 1.27 for Internalising (SD = 1.71), 2.67 for Attention (SD = 2.43), and 2.78 for Externalising (SD = 2.78).
The PSC-17 cut-off scores used by NovoPsych follow the original Gardner and colleagues (1999) recommendations: a Total score equal to or greater than 15, an Internalising score equal to or greater than 5, an Attention score equal to or greater than 7, and an Externalising score equal to or greater than 7 each indicate a positive screen warranting further assessment. These thresholds were derived from the receiver operating characteristic analyses (summarised in the Validity section above) and are widely adopted in the published PSC-17 literature, including the Murphy and colleagues (2016) replication in 80,680 children, which found that observed prevalence rates at these cut-offs were consistent with the rates observed in the original derivation sample.
An alternative set of receiver operating characteristic-derived cut-offs was reported by Stoppelbein and colleagues (2012) against the Child Behavior Checklist (T-score equal to or greater than 70) as the criterion (Total cut-off equal to or greater than 12, Internalising cut-off equal to or greater than 2, Attention cut-off equal to or greater than 5, Externalising cut-off equal to or greater than 6). These alternative cut-offs are more sensitive but less specific than the cut-offs proposed by Gardner et al. (1999) and were derived against a different criterion. NovoPsych retains the Gardner cut-offs as the primary interpretive framework, consistent with the broader literature and clinical convention.
Both threshold sets are integrated on NovoPsych under a three-tier descriptor framework. The Gardner cut-offs serve as the primary screening threshold, marking the boundary between the Mild and the above-clinical-screening-threshold ranges. The Stoppelbein cut-offs serve as the lower bound of the Mild range, flagging subthreshold scores worth clinical monitoring without meeting the threshold for a positive screen.
The PSC-17 items administered on NovoPsych use singular “they / their / them” pronouns in the three items where the original Gardner and colleagues (1999) wording referenced the child’s gender. This is a NovoPsych-localised wording adaptation intended to broaden inclusion of non-binary and gender-diverse respondents while preserving the construct content of each item.
The three affected items are:
External benchmarks used in NovoPsych’s PSC-17 scoring (the percentile rankings from Stoppelbein et al., 2012; the cut-offs from Gardner et al., 1999; the reliable-change thresholds from Murphy et al., 2016, Kamin et al., 2015, and McCarthy et al., 2016) were all derived using the original Gardner item wording. Interpretation of NovoPsych scores against those benchmarks relies upon an assumption of equivalence between versions.
The PSC-17 results on NovoPsych use a three-tier descriptor framework, with a “Mild” band sitting between “Within typical range” and “Above clinical screening threshold”. The Mild range is anchored to an alternative empirically-derived cut-off reported by Stoppelbein and colleagues (2012), and is intended as a monitoring flag rather than a positive screen. Scores in this range fall below the primary clinical screening cut-off but may still be worth tracking across administrations, particularly if the score is rising over time, if other informants report similar concerns, or if contextual factors (e.g., recent stressors, transitions, or developmental shifts) might explain the elevation. A reliable increase from the typical range into the Mild range, or from the Mild range into the above-clinical-screening-threshold range, may warrant additional clinical attention even when the absolute score remains modest.
The PSC-17 captures one informant’s perspective: the parent. Children’s psychosocial concerns often present differently across settings (home, school, with peers), so triangulating parent-reported scores with other informants strengthens clinical decision-making. For attentional and behavioural concerns in particular, teacher reports are often valuable because difficulties frequently produce school-setting impairments that may not be visible at home; the Vanderbilt ADHD Diagnostic Teacher Rating Scale (VADTRS) captures this directly, and the Swanson, Nolan and Pelham Rating Scale (SNAP-IV) can be administered to either parents or teachers. Older children can also self-report on related domains using measures such as the Depression Anxiety Stress Scales – Youth Version (DASS-Y) or the Spence Children’s Anxiety Scale – Child (SCAS-Child) for internalising symptoms specifically. Discordance between informants is itself clinically informative: a parent reporting subthreshold attention difficulties while a teacher reports significant classroom impairment, for example, may point to setting-specific factors worth exploring.
The PSC-17 supports two distinct uses, and administration frequency depends on which is intended. As a screening tool, it is typically administered once at intake to identify children who may benefit from further psychosocial assessment. As a measurement-based care (MBC) tool to monitor progress during treatment, it is usually administered at regular intervals (commonly every three months in routine outpatient care). Both short-term and long-term test-retest reliability are well established (Murphy et al., 2016; Jacobson et al., 2019), supporting use across both short retest intervals and multi-month treatment timelines. The brevity of the measure (around 2.5 minutes to complete) makes repeated administration practical without imposing significant respondent burden.
A positive PSC-17 screen indicates that further psychosocial assessment is warranted; it does not constitute a diagnosis. The next step is typically a more detailed clinical evaluation, which may include diagnostic interviews, gathering reports from additional informants (teachers, the child themselves), reviewing developmental and family history, and considering domain-specific measures based on which subscale is elevated. For example, a positive Internalising screen might prompt administration of an anxiety-specific measure such as the Spence Children’s Anxiety Scale – Child (SCAS-Child) or a broader internalising measure such as the Depression Anxiety Stress Scales – Youth Version (DASS-Y) for older children; a positive Attention screen often prompts more detailed attention-related assessment using the Vanderbilt ADHD Diagnostic Teacher Rating Scale (VADTRS) or the Swanson, Nolan and Pelham Rating Scale (SNAP-IV) to incorporate teacher input; a positive Externalising screen typically prompts further evaluation across multiple settings, including teacher and school perspectives. The brevity and breadth of the PSC-17 make it well-suited as a triage tool that points clinicians toward the most relevant follow-up assessment.
Two concepts apply to interpreting change between administrations:
In practice, a reliable change without a band shift (e.g., a Total score moving from 8 to 14, both within the typical and Mild ranges) suggests a meaningful change in symptoms worth discussing with the family but does not constitute a positive screen.
A reliable change that crosses the primary screening cut-off indicates a clinically significant transition and typically prompts more detailed clinical follow-up, regardless of whether the direction is improvement or deterioration. Sub-reliable changes (e.g., a 1-point drop on a subscale) are best interpreted with caution; they may reflect measurement variability rather than meaningful change.
Gardner, W., Murphy, M., Childs, G., Kelleher, K., Pagano, M., Jellinek, M., McInerny, T. K., Wasserman, R. C., Nutting, P., & Chiappetta, L. (1999). The PSC-17: a brief Pediatric Symptom Checklist with psychosocial problem subscales. A report from PROS and ASPN. Ambulatory Child Health, 5(3), 225–236.
Gardner, W., Murphy, M., Childs, G., Kelleher, K., Pagano, M., Jellinek, M., McInerny, T. K., Wasserman, R. C., Nutting, P., & Chiappetta, L. (1999). The PSC-17: a brief Pediatric Symptom Checklist with psychosocial problem subscales. A report from PROS and ASPN. Ambulatory Child Health, 5(3), 225–236.
Gao, R., Distefano, C., Liu, J., Jiang, N., Greer, F., & Dowdy, E. (2025). Uniform differential item functioning across gender, grade level and racial groups: A MIMIC investigation of the non-clinical parent ratings of the Pediatric Symptom Checklist-17. Journal of Psychopathology and Behavioral Assessment, 47, 15. https://doi.org/10.1007/s10862-024-10180-x
Jacobson, J. H., Pullmann, M. D., Parker, E. M., & Kerns, S. E. U. (2019). Measurement based care in child welfare-involved children and youth: Reliability and validity of the PSC-17. Child Psychiatry & Human Development, 50, 332–345. https://doi.org/10.1007/s10578-018-0845-1
Jacobson, N. S., & Truax, P. (1991). Clinical significance: A statistical approach to defining meaningful change in psychotherapy research. Journal of Consulting and Clinical Psychology, 59(1), 12–19. https://doi.org/10.1037/0022-006X.59.1.12
Jellinek, M. S., Murphy, J. M., Robinson, J., Feins, A., Lamb, S., & Fenton, T. (1988). Pediatric Symptom Checklist: screening school-age children for psychosocial dysfunction. The Journal of Pediatrics, 112(2), 201–209. https://doi.org/10.1016/s0022-3476(88)80056-8
Kamin, H. S., McCarthy, A. E., Abel, M. R., Jellinek, M. S., Baer, L., & Murphy, J. M. (2015). Using a brief parent-report measure to track outcomes for children and teens with internalizing disorders. Child Psychiatry & Human Development, 46(6), 851–862. https://doi.org/10.1007/s10578-014-0525-8
McCarthy, A., Asghar, S., Wilens, T., Romo, S., Kamin, H. Jellinek, M., & Murphy, J. (2016). Using a brief parent-report measure to track outcomes for children and teens with ADHD. Child Psychiatry & Human Development, 47(3), 407–416. https://doi.org/10.1007/s10578-015-0575-6
Meinert, A. C., Mire, S. S., Kim, H. J., Shellman, A. B., Keller-Margulis, M. A., & Curtis, D. F. (2025). A study of the psychometric properties of the Pediatric Symptom Checklist-17 for children with developmental delays and disorders. Clinical Pediatrics, 64(5), 656–664. https://doi.org/10.1177/00099228241284095
Murphy, J. M., Bergmann, P., Chiang, C., Sturner, R., Howard, B., Abel, M. R., & Jellinek, M. (2016). The PSC-17: Subscale scores, reliability, and factor structure in a new national sample. Pediatrics, 138(3), e20160038. https://doi.org/10.1542/peds.2016-0038
Murphy, J. M., & Jellinek, M. (n.d.). Pediatric Symptom Checklist (PSC). Massachusetts General Hospital. Retrieved May 7, 2026, from https://www.massgeneral.org/psychiatry/treatments-and-services/pediatric-symptom-checklist
Stoppelbein, L., Greening, L., Moll, G., Jordan, S., & Suozzi, A. (2012). Factor analyses of the Pediatric Symptom Checklist-17 with African-American and Caucasian pediatric populations. Journal of Pediatric Psychology, 37(3), 348–357. https://doi.org/10.1093/jpepsy/jsr103
