Universal health-related social needs screening in a paediatric rheumatology clinic

Abstract

Objectives

Health-related social needs (HRSN) have a significant impact on health outcomes. While screening is prevalent in paediatric primary care settings, little evidence exists regarding the feasibility of HRSN screening in paediatric rheumatology clinics. Individuals with paediatric rheumatic disease have frequent longitudinal visits, therefore it is important to understand social needs that may hinder optimal health outcomes in this population. The objectives of this study were to understand the burden of HRSN and to contextualize the feasibility of universal HRSN screening in a busy paediatric rheumatology clinic.

Methods

The study team secured funding for a full-time social worker and developed a needs assessment questionnaire (NAQ) through an iterative process. The paper-based NAQ was distributed by administrative clinical staff to all patients at each in-person visit. Medical providers returned completed NAQs to a designated collection tray. The responses were reviewed and analysed. The social worker addressed positive screens among families who opted in for contact.

Results

More than 1/3 of patients returned an NAQ (985/2819), 604/985 (61%) of which revealed one or more concerns. The most frequent social concern was school or work related. The social worker responded by phone or patient electronic portal message to returned NAQs that identified unmet needs.

Conclusion

A school or work concern was a frequent health-related social need in the paediatric rheumatology clinic. The prevalence of unmet HRSN, despite primary care access, shows the importance of screening in rheumatology clinics. Sustainable funding is needed to scale efforts to assess and address them.

Introduction

The American Academy of Pediatrics and the Joint Commission have encouraged paediatricians and required healthcare organizations to screen for social determinates of health (SDoH), hereafter referred to as health-related social needs (HRSN), and to collaborate with community partners to address social needs as a routine part of care [1–3]. Paediatric rheumatology is the branch of medicine that specializes in the study, diagnosis and treatment of a heterogeneous group of inflammatory and autoimmune conditions affecting the musculoskeletal, connective tissue and various organ systems [4]. As with other childhood conditions, unmet HRSN in rheumatic disease are associated with inequities in health outcomes compared with those without HRSN [5]. One study found that publicly vs. privately insured children with juvenile idiopathic arthritis (JIA) were more likely to have more disability, a lower health-related quality of life and a polyarticular or systemic disease course [5]. Most paediatric rheumatic conditions are chronic and require frequent follow-up, therefore unmet needs such as missed school, health-related costs and transportation issues are potentially amplified in this population.

Although International Classification of Disease, 10th Revision (ICD-10) Z-codes, a form of structured documentation of HRSN, may be leveraged to understand the social needs of a given population, their use remains low in paediatric electronic health records (EHRs) [6, 7]. Limited data report successful universal HRSN screening in paediatric specialty care settings [8, 9]. To date, we are unaware of any studies reporting the burden of HRSN in an outpatient paediatric rheumatology setting. Therefore, we aimed to develop a process for universal screening and understand the unmet HRSN of our patients and families in a large academic outpatient paediatric rheumatology clinic. The resulting study arose out of a quality improvement (QI) initiative.

Methods

Setting

This study was conducted at the main campus of a 477-bed free-standing children’s hospital and tertiary care centre outpatient paediatric rheumatology clinic in a major US city. The paediatric rheumatology program evaluates and treats ≈2800 patients annually.

Study team and study population

The primary study team included a full-time social worker, quality improvement specialists and paediatric rheumatologists. All patients who attended an in-person visit between 1 January 2022 and 31 December 2022 received a Needs Assessment Questionnaire (NAQ) at every visit. There were no exclusions. The medical staff included 17 faculty and 4 trainees (paediatric rheumatology fellows) whose responsibility was to return NAQs to a designated collection tray in a clinician dictation room. One to six providers staffed any given clinic session.

History of the integration of the social worker into the specialty clinic

Historically, the rheumatology clinic did not employ an embedded full-time social worker. HSRN were identified and addressed on an ad hoc basis when a patient or guardian raised an acute or serious concern. Physician members of the rheumatology department secured funding via philanthropic and institutional support to hire a part-time social worker who initiated a formal assessment of HRSN soon thereafter. Due to the pervasive HRSN identified, funding was increased for a full-time social worker from the same philanthropist who committed funding for half the full-time salary with a matched contribution from divisional resources. During the study period, the social worker was available for the outpatient clinic in person or by pager 35 h/week divided over the standard 5-day work week. This study focuses on one calendar year during which a full-time social worker (M.T.) was integrated in the rheumatology clinic.

NAQ development

The study team developed the NAQ a priori based on the SDoH framework domains of Healthy People 2030 [10] and other screening tools [11] through an iterative process. The final version of the NAQ had 11 items that were printed on neon pink paper, one side in English and the other side in Spanish (Fig. 1). Domains included emotional/behavioural, adjustment to illness, housing, utilities, living conditions, transportation, food insecurity, medical/health-related costs, health insurance coverage and education/work concerns. Education or work concerns were not collected as separate domains until 25 October 2022. Additional features of the NAQ included a telephone number for families to contact the rheumatology social worker and a hotline with an associated QR code that linked patients and families to local community and statewide health resources (Mass211) [12]. The first question on the screening provided the opportunity for patients or parents to opt out of contact.

Implementation process

The study team drafted a process map that outlined the implementation of universal HRSN assessment in the clinic. The final NAQ and process map were reviewed with medical providers at one official training session at its launch. The process map underwent several iterations based on the informal feedback of the clinical administrative and medical staff until finalized (Supplementary Fig. S1, available at Rheumatology Advances in Practice online). Either a medical provider, nurse or a positive finding on the NAQ initiated a referral to the social worker. Providers were educated about the implementation process in two training sessions and encouraged to page the social worker for acute needs or safety concerns.

The NAQ was distributed in paper format by front desk staff to every patient/family for every in-person clinic visit. The study team set a goal for ≥80% of all NAQs to be returned to a single receptacle. They were collected by the social worker within 1–2 business days. The social worker then reviewed and logged positive screens and tracked interventions into an Excel file (Microsoft, Redmond, WA, USA) stored electronically behind an institutional firewall. A quality improvement (QI) administrative assistant logged all screening results into an internal instance of the Research Electronic Data Capture (REDCap) [13, 14] platform.

Social worker interventions

Each positive NAQ screen resulted in a social worker consult unless the respondent opted out of contact. Resources and communication templates compiled by the social worker were transmitted to patients with identified HRSN via the EHR-based patient portal or a phone call. Standardized information about food assistance programs, family resource centres, housing assistance programs, medical cost assistance, behavioural health, school accommodations, bullying prevention, school advocacy, home tutoring, transportation and utility assistance were included in the communication templates.

A new community partnership was developed between the paediatric rheumatology program and Mass211 [12], a statewide non-profit organization, to help address gaps in the available resources. Mass211 builds and maintains a web platform and telephone hotline largely sponsored by the Council of Massachusetts United Ways that provides information to address social and health-related needs. At no cost to users, the referral hotline is available 24 h/day, 7 days/week. A web-based listing of resources ranging from, but not limited to, adolescent and youth services, food and nutrition, housing, financial assistance, legal aid, medical care and insurance, mental health, transportation, utilities and employment is also available. Paediatric rheumatology-specific content was added to the Mass211 website in 2022, funded by additional philanthropic support.

Ethical statement

This project was conducted according to local institutional standards for QI initiatives and therefore is exempted from official institutional review board review.

Results

Characteristics of the clinic population

During the study period there were 3788 eligible visit encounters (Table 1), representing 2819 distinct patients. Approximately one-quarter of all patients were publicly insured. English was the preferred language for 94% of families. The most frequently spoken languages besides English were Spanish [n = 87 (3%)], Brazilian Portuguese [n = 26 (1%)] and Arabic [n = 16 (0.5%)]. A total of 67% of patients identified as White, 4% Black, 3% Asian and 5% Hispanic or Latino (not mutually exclusive with race). Approximately 25% either declined or did not specify race or ethnicity. There was no statistically significant difference in the proportion of demographic characteristics among those who returned the NAQ compared with the entire clinic population.

Implementation

Initially, 16 (94%) faculty providers consistently participated in this project, however, this increased to 100% participation by the end of the study observation period. A total of 2819 individual patients presented for care during the study period and 985 patients returned NAQs (35%). Among those who returned NAQs, 138 individuals opted out of being contacted. NAQs were given out at each visit, therefore some patients responded to the NAQ multiple times over the course of the study period. There were 3788 visits to the rheumatology clinic, translating to 3788 NAQs distributed and 2355 (62%) of them collected.

Identified needs

Among the 985 patients who returned NAQs (Table 2), 604 (61%) indicated at least one need. Educational and work concerns (48%) were the most prevalent concerns identified, followed by adjustment to illness (32%) and emotional or behavioural concerns (21%). The social worker responded to all positive NAQs via the patient portal or a phone call. No calls to the Mass211 hotline were received by patrons who cited a paediatric rheumatologic condition as the primary concern or reason for engaging with the hotline. However, in 2023, the year following the study period, rheumatology-specific content on the website was viewed 11 times (affiliations unknown).

Discussion

This report describes process and context for universal HRSN screening and reveals the estimated prevalence of unmet HRSN in an outpatient paediatric rheumatology clinic within a large tertiary care academic free-standing children’s hospital in a major urban US city. The clinic serves a population of primarily commercially insured (72%), White (67%) patients, among whom English (94%) is the primary language. The findings revealed a substantial burden of HRSN, as 61% of screens (n = 604) reported one or more HRSN. The prevalence of at least one in five individuals with unmet needs who present for specialty care in a clinic population that is skewed heavily toward the middle class, has minimal language barriers, in the only state in the USA with near universal access to primary care is contextually very significant. The reported process of universal HRSN screening was sustained for ≥1 year, however, it required extensive administrative effort and thus poses a threat to the sustainability and scalability of the process described.

Challenges with implementation

The paper-based NAQ was easily integrated into the flow of the clinic check-in process. They were included in the distribution packs of other paper-based clinical assessment forms that screen for disease activity and functional status. The average return rate of 62% of completed NAQs over 12 months was satisfactory; however, 75% of NAQs were returned during the final month of the observation period. The goal of 80% completion was not achieved. The 62% completed return rate is similar to another study of HRSN screening in a paediatric haematology clinic at a different medical centre in the surrounding metro area [8]. Failure to meet the target rate in our study could be attributed to at least three factors. First, the order of the forms may influence their rates of completion. Families are given at least two to three other forms to complete at check-in while they wait to be evaluated by their provider. Second, the paper-based process of in-clinic data collection from the NAQ relies on the medical provider to manually return the completed paper-based form to the single tray designated by the social worker. Some NAQs were lost at this step, in part because several providers initially discarded questionnaires that indicated no unmet HRSN. Third, having only one place to return the NAQs was a significant change in workflow for some providers. To address this challenge, providers were reminded of the process map. Study staff also reviewed implications about the ethics of families having reported HRSNs that could potentially go unaddressed. Lastly, one provider initially passively declined to participate in the process but later engaged once a study team member met with the provider individually to discuss the project. Ultimately the rate of returned NAQs increased to 75% during the final quarter of the study period.

High burden of need and sustainability

Long-term sustainability relies on several factors. Integrating dedicated, full-time social work staff in the clinic is crucial for managing unmet needs. The volume of referrals severely limited the ability of the social worker to engage intensively with families that needed focused specialized support. To scale the ability to address the high burden of needs identified by the NAQ, community health workers or patient navigators could be hired to review the screenings and provide standardized resource sheets to the families [15]. This would allow the social worker to engage in personalized interventions at the highest level of their skill set. Specialized social work interventions would include referrals to specific mental health providers and other community resources based on the unique complexities of each situation [16, 17]. They also include in-depth collaboration with primary care providers, agencies or schools on section 504 accommodations (Rehabilitation Act of 1973) to ensure full access to academics and social enrichment [16, 17]. Other interventions would include social worker–led patient support groups, grief debriefings and consultations for patients and families at the onset of a diagnosis of complex multisystem diseases such as SLE, ANCA-associated vasculitis, RF-positive polyarticular JIA, juvenile dermatomyositis, systemic scleroderma or systemic JIA (adult-onset Still’s disease) [18, 19].

While challenges related to implementing EMR changes may be manually and fiscally burdensome initially, the use of technology to automate a universal screening process could reduce administrative burden and improve analytical efficiency to address unmet HRSN. Additionally, we could further leverage technologies such as artificial intelligence to identify unmet HRSN in free-text EHR notes and connect families with the appropriate resources [20]. Ultimately, continued support and investment from health administrators and clinical teams are vital for ongoing sustainability.

Strengths

This study has several strengths, the first of which is the inclusion of educational concerns on the NAQ. Education status is a predictor of morbidity and mortality, even when controlling for income status in the USA [21–23]. Therefore, educational concerns in childhood could have compound effects on future achievement. Most HRSN questionnaires screen for issues that have more immediate health-related implications such as financial insecurity, food, housing and transportation. Educational concerns for the child are not included in many HRSN screenings [24–26].

Second, the nature of the process of identifying resources for families fostered the organic development of a sustained partnership with Mass211 [12]. This collaboration has the potential to improve rheumatology-specific health literacy and access to related support for residents of the state. This model is adaptable to other contexts to develop sustainable partnerships that extend beyond the walls of a single clinic or medical centre.

Limitations

This study has limitations. First, it was conducted at a single clinic site within a single hospital system. Despite the limited generalizability of the quantitative results of this study, an attempt was made to describe the context of this study since features of implementation and the process for universal screening may be transferable where certain contextual factors are similar.

The timing of the study coincided with the final months of the COVID-19 pandemic state of emergency, during which there were major disruptions for students. It is unclear whether the COVID-related school disruptions influenced the reporting of concerns among this population. Since few screenings routinely assess education or school-related concerns for the child in the context of an underlying health condition [25, 26], we have a limited pre-COVID basis for comparison. The question about educational concerns was framed to relate specifically to health conditions. Responses to educational or work concerns were aggregated until October 2022, however, educational concerns outnumbered work concerns 7:2 after they were assessed in separate questions.

Finally, the reported needs may not reflect the true burden and distribution of needs.

NAQs were collected from 35% of individual patients. While there was no statistically significant difference in demographic characteristics between individuals that returned at least one NAQ compared with the entire clinic population, it is possible that there is overreporting among individuals with lower levels of disease control who may be more likely to return a completed NAQ. Alternatively, there could be bias resulting in underreporting, due to stigma associated with disclosing other needs (food, housing, etc.). Although the proportion of HRSN reported on our paper-based NAQ was consistent with the face-to-face disclosure rate of needs in another study [27], we may not be capturing the burden of unmet HRSN compared with what could be captured if HRSN were screened for electronically, which offers more privacy [27].

Certain analyses were beyond the scope of this study, including validation of the survey responses or psychometric testing of the NAQ or analysis of differences in response rates or concerns by demographic or specific disease factors. We have included Supplementary Table S1, available at Rheumatology Advances in Practice online, to illustrate the top 10 diagnoses ever coded for all individuals who presented to the clinic during the study period. The supplementary table represents aggregated data not analysed at the individual level to examine associations with the presence or absence of HRSN. A formal cost analysis was not directly assessed; however, sustainability could be threatened because the availability of a full-time social worker is based on philanthropic donations and not a standard line item in the institutional budget.

Future directions

There are many possible future directions for this study, some of which have been mentioned in previous sections. They include automating the screening process and introducing methods to reduce the administrative burden of the social worker to allow for more dedicated interventions specific to his/her skill set [15, 28]. Identification of factors associated with unmet HRSN could facilitate the design of targeted interventions. Systematically tracking patient engagement in the resources to which they were referred, including the Mass211 hotline and website, is important for understanding the impact and effectiveness of interventions. Since this study revealed the highest burden of health-related social needs in the domain of education/work, the nature of educational concerns warrants further investigation [29]. Future iterations of the NAQ will no longer include an assessment of behavioural health needs, as they are direct health needs rather than health-related social needs. Behavioural health screenings were incorporated in the EHR system’s patient portal institution-wide at the time this report was drafted.

In conclusion, broader uptake in HRSN screening is being mandated in various clinical settings, therefore the process of screening and context described herein may be adapted to other centres. The volume of patients with concerns for unmet HRSN in a population with reliable access to primary care highlights the importance of universal screening in a subspecialty clinic. The prevalence of needs resulted in a substantial administrative burden for one full-time social worker. Therefore, we assert that clinical social services warrant reimbursement or care management support through insurance payers and consideration as a standing item in institutional budgets. Finally, the number of individuals with educational concerns identified in this population suggests that the domain may need to be considered as a standard feature of paediatric HRSN screening tools.

Supplementary material

Supplementary material is available at Rheumatology Advances in Practice online.

Data availability

The data underlying this article cannot be shared publicly to protect the privacy of individuals who participated in the study. The data will be shared upon reasonable request to the corresponding author.

Authors’ contributions

M.T.C. conceptualized and designed the study, designed the data collection instrument, collected data, carried out the initial analysis, drafted the initial manuscript and critically reviewed and revised the manuscript. M.A.B. designed the study, designed the data collection instrument, carried out the initial analysis and critically reviewed and revised the manuscript. A.D. carried out the revised analysis and critically reviewed and revised the manuscript. M.T. designed the data collection instrument, collected data, carried out the initial analysis and critically reviewed the manuscript. B.G.E. collected data, carried out data analysis and critically reviewed the manuscript. J.C.C. collected data and critically reviewed and revised the manuscript for important intellectual content. B.E. and M.B.S. coordinated and supervised data collection, collected data and critically reviewed and revised the manuscript. M.M.H. conceptualized and designed the study, designed the data collection instrument, coordinated and supervised data collection, collected data, carried out the initial analysis and critically reviewed and revised the manuscript. All authors approved of the final manuscript as submitted and agree to be accountable for all aspects of the work.

Funding

M.T.C. receives salary support from the Childhood Arthritis and Rheumatology Research Alliance (CARRA). M.T. received salary support from the Lang Family Fund for Rheumatology and the Samara Jan Turkel Clinical Center for Pediatric Autoimmune Disease.

Disclosure statement: The authors have declared no conflicts of interest.

Acknowledgements

The authors wish to acknowledge CARRA and the ongoing Arthritis Foundation financial support of CARRA. The authors would like to thank Caitlyn Dorsey, MSW for her contribution of ideas, the Lang Family Fund for Rheumatology for philanthropic support of the social work program and the patients and families who continue to choose our clinic to deliver their care. The content is solely the responsibility of the authors and does not necessarily represent the official views of the funding agencies.

References