INTRODUCTION
Youth living with chronic inflammatory conditions such as juvenile idiopathic arthritis (JIA) experience significantly higher rates of anxiety and depression compared to peers without chronic illness.1–4 Psychiatric comorbidities in this population can negatively affect quality of life, treatment adherence, disease activity, and readiness for transition to adult care.3 The relationship between mental and physical health is likely bidirectional: chronic disease may place youth at increased risk for mental health disorders, whereas untreated anxiety or depression can worsen functioning, amplify symptom burden, and undermine disease management.5 Pain intensity and pain interference closely relate to psychological symptoms and contribute to functional and emotional distress in youth with JIA.6 Several studies demonstrate that anxiety and depressive symptoms may persist even when inflammatory disease activity is low or well controlled, underscoring the importance of routine biopsychosocial assessments and care.7 National reports have identified mental health screening as a gap in pediatric chronic disease care, despite increasing emphasis on integrated biopsychosocial models.8 This need has been further underscored by the recent American College of Rheumatology and Childhood Arthritis and Rheumatology Research Alliance guidance statements for addressing mental health concerns in youth with pediatric rheumatologic diseases, which recommend routine, standardized mental health screening as part of comprehensive pediatric rheumatology care.9
Despite this burden, pediatric rheumatology clinics, including at McMaster Children’s Hospital, infrequently perform mental health screening using a standardized tool. As a result, the extent of unmet mental health needs among youth with rheumatic diseases remains largely unknown. Currently, primary care providers typically manage mental health concerns rather than subspecialists.
However, many youth with chronic conditions may have limited or inconsistent contact with primary care providers.10 Medically stable youth often rely primarily on their subspecialist for routine care, and youth with active disease may see their specialist more frequently than their primary care provider. Given evidence that relational continuity facilitates disclosure of psychosocial concerns, embedding mental health screening within subspecialty care may reduce missed opportunities for identification.11
At McMaster Children’s Hospital, no standardized process existed for screening or follow-up of mental health concerns in pediatric rheumatology clinics. Limited access to embedded mental health resources leaves the onus on youth and their families, who are already managing a chronic disease, to recognize symptoms, seek help, and navigate multiple siloed care systems. This gap increases the risk of mental health issues going undetected or poorly managed.
A systematic review of mental health outcomes in youth with JIA demonstrates consistently elevated rates of anxiety and depressive symptoms compared with healthy peers, with reported prevalence ranging from 7% to 36% for depressive symptoms and 7% to 64% for anxiety symptoms, depending on assessment methods and clinical context.3 In a recent Canadian cohort, 40% of youth with JIA screened above clinical thresholds for at least one anxiety or depressive disorder.12 A recent Canadian feasibility study further demonstrated that routine mental health screening using PROMIS anxiety and depression measures can be practically and acceptably integrated into pediatric rheumatology clinic visits, supporting the real-world viability of embedded screening approaches.13 Together, these findings highlight the mental health burden in this population and support the need for routine, standardized mental health screening in pediatric rheumatology care. Recognizing this need, our team aimed to establish a feasible, reliable screening and follow-up care process using quality improvement (QI) methodology. This project seeks to integrate routine screening and timely follow-up pathways directly into subspecialty clinics, improving mental health care assessment and access for youth with chronic inflammatory conditions.
The Patient Health Questionnaire-4 (PHQ-4) is a validated, brief screening questionnaire with four questions that reliably detect anxiety and depression symptoms.14–16 The PHQ-4 may be an effective first-line screening tool to identify youth who may require further assessment or mental health support. Brief screening tools are particularly suited to subspecialty pediatrics, where clinic time is limited and additional questionnaires must not disrupt workflow.
The objectives of this QI initiative were:
METHODS
Setting and Population
We conducted this QI initiative in the pediatric rheumatology outpatient clinics at McMaster Children’s Hospital, a tertiary care center that provides longitudinal care to more than 800 youth with JIA annually. The multidisciplinary clinic team includes 4 full-time pediatric rheumatologists, 1 advanced physiotherapy practitioner, 2 part-time physiotherapists, and 1 registered nurse, with access to child life specialists and social workers as needed. Eligible participants included patients 12–18 years of age with a diagnosis of JIA attending in-person clinic visits.
We excluded patients who could not complete the PHQ-4 in English.
Improvement Framework
We used iterative plan-do-study-act (PDSA) cycles within the Model for Improvement framework to test and refine workflow changes. A key driver diagram guided the development of interventions targeting care team awareness and engagement, standardized workflows, accessible screening tools, visible data feedback, and patient- and family-centered design. A fishbone analysis identified barriers related to people, processes, environment, technology, and policy.
We developed both the key driver diagram (Fig. 1) and the fishbone analysis (Supplemental Digital Content 1, https://links.lww.com/PQ9/A830) with feedback from frontline staff.
Intervention Components
Screening Tool
We conducted mental health screening using the PHQ-4, a brief, validated instrument that combines two items assessing anxiety and two assessing depressive symptoms. Each item is scored on a 4-point Likert scale (0–3), yielding a total score ranging from 0 to 12. The rooming health care provider (eg, physician, registered nurse, physiotherapist, or advanced practice physiotherapist) administered the PHQ-4 either on paper or verbally.
We categorized total scores as 0–2 (normal), 3–5 (mild), 6–8 (moderate), and 9–12 (severe), consistent with established scoring thresholds.14
Follow-up Algorithm
A tiered response pathway (Fig. 2) outlined algorithm-concordant actions, including brief education and self-management resources for minimal or mild symptoms; referral to social work or primary care for moderate symptoms; and an urgent safety assessment for severe symptoms or suicidality. We adapted the algorithm from an existing PHQ-9 and GAD-7 review pathway used in the pediatric systemic lupus erythematosus clinic at SickKids Hospital.
We refined it with input from frontline staff, including social work and child life specialists.17
Workflow Integration
A swimlane diagram defined standardized roles for registration staff, nursing, allied health, and physicians. Visual prompts and designated electronic medical record (EMR) fields supported consistent documentation of PHQ-4 scores and follow-up actions.
Data Feedback and Training
We supported adoption and sustainment through data feedback and training, including biweekly check-ins by a research assistant, monthly reviews of statistical process control charts, and training sessions.
Measures
QI Measures
In accordance with the Model for Improvement, this initiative assessed process, outcome, and balancing measures. Process measures were the percentage of eligible visits with PHQ-4 completion (objective 1), and the percentage of screened visits with documented algorithm-concordant follow-up actions (objective 2). Outcome measures were the percentage of patients screening positive on the PHQ-4 (score ≥v3).
Balancing measures consisted of patient and healthcare provider feedback surveys assessing satisfaction and perceived burden, with optional free-text feedback. We administered feedback surveys to health care providers who conducted the screening. The surveys used a 5-point Likert scale, with 1 indicating “strongly disagree” and 5 indicating “strongly agree.”
Analysis
We used statistical process control rules to detect nonrandom variation and assess the intervention’s impact over time, using P-charts when appropriate.
Ethical Considerations
As this project met institutional criteria for QI and involved no collection of identifiable data, our institutional review board did not require formal research ethics review.
RESULTS
Process Measure: Objective 1: Screening Reliability
Three iterative PDSA cycles (May 2024–August 2025) increased PHQ-4 completion from 0% to a sustained ≥75% across eligible visits. Across 266 consecutive audited visits, screening reliability remained consistently above target (Fig. 3).
PDSA Cycles
We conducted 3 PDSA cycles to test and refine the screening workflow iteratively.
Cycle 1 focused on foundational design work. The team developed a driver diagram, completed a fishbone analysis, and drafted the initial PHQ-4 workflow, including EMR shortcuts such as smart phrases and flow sheets. We used one-on-one educational sessions to test feasibility, introduce EMR documentation, and clarify how health care providers should administer and document the screening tool.
Cycle 2 aimed to strengthen staff engagement and standardize administration. We delivered a brief educational session to all clinic staff outlining the importance of mental health screening for youth with JIA. Based on feedback from early tests, we placed readily accessible paper copies in clinical work areas to support consistent use.
Cycle 3 targeted reliability and sustainment. We introduced biweekly check-ins to identify missed screens, troubleshoot barriers in real time, and reinforce documentation expectations across the team. These refinements contributed to sustained improvements in screening reliability over time.
These cycles produced stepwise improvements and supported sustained screening reliability above target.
Process Measure: Objective 2: Follow-up Reliability
A standardized follow-up algorithm (Fig. 2) operationalized care steps and expectations. By the end of the implementation period, ≥75% of those screened had a documented algorithm-concordant action (education, referral, or safety assessment) (Fig.
4). This component required no additional PDSA cycles because screening volume remained high and follow-up uptake was consistent.
Outcome Measure: Positive Screens and Symptom Severity Distribution
Among 266 visits, 17% of participants had a PHQ-4 score ≥3. Severity distribution was: 83% normal (0–2), 12% mild (3–5), 3% moderate (6–8), and 2% severe (9–12) (Fig. 5).
Balancing Measures
Balancing measures included postscreening healthcare provider and patient feedback surveys (Supplemental Digital Content 2, https://links.lww.com/PQ9/A831).
Five healthcare providers who administered the screening completed feedback surveys. Overall, healthcare providers reported low perceived burden, with strong disagreement that the screening was time-consuming (mean 1.2/5), difficult to interpret (mean 1.0/5), or delayed clinic flow (mean 1.6/5). Most also disagreed that screening made no difference to patient care (mean 1.4/5).
Healthcare providers agreed that the screening helped identify patient needs (mean 4.0/5), improved communication with patients and caregivers (mean 3.6/5), and improved overall quality of care (mean 3.8/5). All healthcare providers indicated willingness to continue using the tool as part of routine care.
Postscreening feedback from 32 patients showed that patients felt comfortable with mental health screening during clinic visits and perceived it as valuable. Because we collected these data in a de-identified manner, we could not calculate the exact proportion of survey responses relative to the number of unique patients screened. However, we collected survey responses over approximately 2 months, well within a routine follow-up interval of approximately 6 months, so the 32 respondents likely represent unique patients.
Most respondents reported feeling either very comfortable (46.9%) or comfortable (28.1%) completing the screening, with 25.0% reporting neutral comfort and no respondents indicating discomfort. Most also felt that asking mental health questions during clinic visits was very important (56.3%) or important (34.4%), with few respondents expressing neutral views (9.4%) and none indicating it was not important. Nearly half of respondents (46.9%) reported that a healthcare provider reviewed or mentioned the screening results with them during the visit (Fig.
6). Overall, patients showed strong acceptability of screening, perceived the relevance of mental health discussions in subspecialty care, and identified opportunities to further standardize follow-up communication.
DISCUSSION
Interpretation
In this initiative, PHQ-4 screening increased rapidly after implementing standardized workflows and iterative PDSA cycles, with several months achieving screening rates above 85%–90% and sustained performance at or above the 75% target over time (Fig. 3). In parallel, documentation of algorithm-concordant follow-up actions improved quickly after introduction of the structured follow-up pathway, reaching and maintaining percentages above 80%, with peaks approaching 100% (Fig.
4). These findings show that a multidisciplinary pediatric rheumatology clinic could achieve high reliability, even without fully automated previsit electronic screening, through deliberate workflow integration and team-based task sharing.
These results compare favorably with some prior reports from subspecialty clinics that relied on longer screening instruments, in which median screening rates frequently ranged from 50% to 70%.17–19 At the same time, the observed performance was comparable to higher completion rates reported in tablet-based or electronically completed patient screening, which often reach 85%–90%.18 Notably, the present initiative achieved sustained reliability without full automation or previsit electronic completion, suggesting that high screening performance may also be attainable through deliberate workflow integration and team engagement in resource-constrained settings. Similarly, QI initiatives incorporating structured workflow redesign, staff training, and practice facilitation have meaningfully increased depression and anxiety screening completion in primary care and other pediatric settings,20–22 supporting the broader applicability of this approach to mental health screening across pediatric care contexts.
Several features of this initiative distinguish it from prior work and may help explain its sustained performance. First, the brief screening tool minimized added burden for both patients and health care providers. Unlike longer instruments, the PHQ-4 is designed to rapidly flag anxiety and depressive symptoms rather than provide diagnostic assessment.
This brevity likely facilitated consistent administration during routine clinic flow and reduced fatigue. In addition to its brevity, the PHQ-4 enhances the generalizability of this approach to other subspecialty clinics seeking to efficiently address multiple mental health conditions.
Second, we paired screening with a simple, clearly operationalized follow-up algorithm that translated screening results into predefined actions. By clarifying expectations for education, referral, or safety assessment, the algorithm reduced reliance on ad hoc clinical decision-making and minimized the need for real-time consultation or escalation. This standardization supported timely, consistent responses across providers while limiting additional workflow burden.
A third, and particularly important, contribution of this project is its emphasis on co-design with frontline clinic staff. Unlike some prior initiatives that were primarily investigator-driven or dependent on individual champions, we iteratively refined the workflow, documentation processes, and follow-up expectations in this study with direct input from nurses, allied health professionals, physicians, and social work. This collaborative approach likely enhanced feasibility, acceptability, and shared ownership of the process.
Health care provider feedback supports this interpretation, as staff reported low perceived burden, minimal workflow disruption, and willingness to continue using the screening tool as part of routine care. The literature supports that staff engagement and perceived value are critical determinants of sustainability in QI.23–25
The symptom severity distribution observed in this initiative also warrants careful interpretation. Most positive screens fell within the mild range, with relatively few moderate or severe scores, a pattern that differs from rates reported in some studies of youth with JIA.12,26 This difference likely reflects both population and measurement factors. The clinic population in this study included many youth with well-controlled disease and access to psychosocial support, which may reduce symptom burden.
We did not systematically collect measures of disease activity and baseline mental health diagnoses as part of this initiative, and we included all eligible patients regardless of disease activity or preexisting mental health conditions. In addition, the PHQ-4 identifies current anxiety and depressive symptoms rather than establishing psychiatric diagnoses and may capture a narrower spectrum of distress than longer or disorder-specific instruments. Furthermore, the mode of screening administration, including verbal delivery or completion in the presence of caregivers, may have influenced symptom disclosure and contributed to underreporting, as confidentiality has been shown to affect youth willingness to report mental health concerns.
Taken together, the existing literature suggests that successful implementation of mental health screening in pediatric subspecialty care depends less on the choice of screening instrument alone and more on how screening is embedded into routine clinical practice. This project adds to the literature by demonstrating that sustained, reliable screening and follow-up are achievable using a brief tool when supported by frontline co-design, standardized yet flexible workflows, visible data feedback, and clearly defined follow-up expectations. These findings complement prior studies emphasizing automation and electronic completion by showing that alternative, lower-resource approaches may also be effective when implementation is thoughtfully tailored to local context.
Implications
Reliable mental health screening and timely linkage to appropriate follow-up care may improve treatment adherence, symptom management, and transition readiness for youth with chronic inflammatory disease.27–30 Integrating screening within a subspecialty clinic, the primary point of contact for many youth, can reduce reliance on fragmented primary care referrals and supports a more patient-centered model. The workflow and tools developed here are scalable to other chronic disease programs with similar challenges.
Limitations
As part of a QI initiative, we did not collect patient-level demographics, socioeconomic factors, and disease activity indices, limiting opportunities for subgroup analyses and equity-related assessments. The PHQ-4, although highly feasible for routine use, may underdetect certain mental health presentations such as somatic symptom expression, trauma-related concerns, or mild yet functionally impairing distress. Implementation occurred at a Canadian tertiary care center with established psychosocial supports, which may limit generalizability to settings with fewer resources or different clinic structures.
Next Steps
The primary aim of this initiative was to demonstrate that routine mental health screening and a standardized follow-up pathway can be feasibly integrated into pediatric rheumatology care. Future efforts will focus on implementing confidential screening, sustaining screening reliability within rheumatology, and extending the workflow to other pediatric subspecialty clinics at the same center. Although this project does not include extensive downstream outcome analyses, the workflow, tools, and implementation insights generated here may facilitate replication in other chronic disease programs.
By sharing this approach and early results, this QI initiative contributes to broader efforts to promote routine mental health screening across pediatric subspecialty settings.
ACKNOWLEDGMENTS
The pediatric rheumatology clinical team at McMaster Children’s Hospital is gratefully acknowledged for its contributions to workflow development, pilot testing, and ongoing feedback throughout this QI initiative. This project received financial support through an investigator-initiated grant from Pfizer Canada. The authors declared no conflicts of interest.
We presented results from this initiative at the Children’s Healthcare Canada Annual Conference (2025) and the Canadian Rheumatology Association Annual Scientific Meeting (2026).
Supplementary Material