Study setting

Long Island, a suburb of New York State, faces persistent racial inequities in healthcare access, treatment, and health outcomes. The Nassau and Suffolk counties have a long history of neglecting the health concerns of their Black residents due to structural racism that enables racially segregated housing, school districts, and physical environments. Black and Hispanic residents experience worse outcomes on many health indicators, highlighting the longstanding racial inequality in the area. The NYS Department of Health (2024) reports that 55.4% of Hispanic, 51.2% of Black, and 29.0% of White Nassau County residents die prematurely before age 75. The Black infant death rate in Nassau County is 5.5 per 1000 live births, while the rate for White infants is 2.0 per 1000 live births [14]. Additionally, Nassau and Suffolk Counties had 346,488 and 416,371 Medicaid beneficiaries in 2023, respectively [14]. Long Island accounts for 10% of New York’s Medicaid recipients [14]. Social care integrated with primary care has the potential to improve health outcomes for nearly 763,000 Long Islanders, or 27% of the population. Long Island’s human service system consists of hundreds of standalone agencies with varying capacities and intake procedures. To access the services they need, clients must navigate multiple agencies, intake processes, applications, and follow-ups within this fragmented social care system. New York State’s 1115 Waiver Amendment [15] aims to integrate primary care with a comprehensive social care network and a workforce development program to address health-related social needs. CHWs play a vital role in this initiative, especially on Long Island. By examining burnout among CHWs working in a diverse suburban area, our findings can help organizations build a strong CHW workforce, which is crucial for advancing health equity, empowering communities, and strengthening the U.S. public health system. The study was conducted in Nassau and Suffolk Counties, suburban regions of Long Island, New York. Providing contextual description aligns with qualitative descriptive methodology, which emphasizes situating findings within real-world practice settings to enhance transferability [16].

Study design

The interview data were collected between April 2024 and September 2024, using a qualitative descriptive research design, an approach appropriate for research seeking a comprehensive, minimally theorized account of participants’ experiences in everyday language [16,17]. Semi-structured in-depth interviews and participant observations were conducted to explore experiences of professional development and emotional labor among community health workers (CHWs) serving clients in Nassau and Suffolk Counties, New York. A qualitative approach was used to capture how CHWs interpret their work experiences, construct meaning, and navigate their work environments. Semi-structured interviews allowed for consistency across participants while retaining flexibility, and participant observations provided insight into CHW work in naturalistic settings. The use of multiple data sources supported analytic triangulation.

Participant eligibility

Eligible participants were current community health workers (CHWs), defined as frontline staff or paraprofessionals engaged in outreach, health education, care navigation, or community-based support. Inclusion criteria were: (1) active employment or engagement as a CHW in Nassau or Suffolk County; (2) age ≥ 18 years; (3) ability to communicate in English or Spanish (interviews were conducted in English, with occasional Spanish used by bilingual participants); (4) willingness to provide informed consent and participate in a 45–90-minute interview; and (5) at least six months of CHW experience. Eligibility criteria were defined to ensure participants had direct experiential knowledge of the phenomenon under study. In qualitative descriptive research, purposive inclusion of participants with practical experience is essential for obtaining information-rich accounts [18]. Exclusion criteria included supervisory-only roles without direct CHW duties, volunteers or interns with less than six months of experience, staff with exclusively administrative or clinical roles, and the inability to provide informed consent.

Recruitment strategy

Purposive and snowball sampling were used to recruit CHWs with diverse experiences across counties and service settings. Qualitative descriptive studies often use purposive sampling to ensure variation while focusing on participants who can provide detailed descriptions of the phenomenon [17,18]. Purposive sampling was employed to ensure variation in gender, age, years of experience, primary county of work, and language use. Snowball sampling helped reach CHWs across different community organizations. Initial participants were identified through partner organizations and CHW networks and were invited via email or phone. Participants were also asked to recommend eligible peers. Recruitment continued until enough information was gathered. A brief demographic questionnaire was administered after each interview to record sample characteristics.

Data collection

Data were collected through semi-structured interviews (n = 10) and non-participant observations (n = 5). Semi-structured interviews align with qualitative descriptive designs because they allow participants to describe experiences in their own words while covering core topics (17). The interviews were conducted via Zoom, lasted approximately 45–90 minutes, and were audio-recorded with participant consent. In addition to interviews, five observation sessions were conducted to contextualize the interview data and enrich the descriptive depth. Combining interviews with observations enhances descriptive validity and supports triangulation in qualitative research [18]. Observations took place over 2–4-hour periods in naturalistic CHW work settings, including community-based organizations, outreach events, team meetings, and case conferences. The main focus of the observations was on CHWs’ daily routines, their interpersonal interactions with clients and colleagues, emotional expressions during service delivery, boundary-setting practices, and the use of organizational supports such as supervision and peer consultation. Interviews were conducted via Zoom to accommodate participants’ schedules and geographic locations. While virtual interviews increased accessibility and comfort for participants, we recognize that this format may limit some contextual and embodied cues present in in-person interactions. To address this, interviewers paid close attention to vocal tone, pauses, and emotional expressions and supplemented interview data with detailed field notes and participant observations made in work settings.

Observations were non-participant and unobtrusive. Detailed field notes were recorded during and immediately after each observation, capturing contextual details, interactional dynamics, and reflexive insights related to emotional labor and workplace support. Following interviews, participants completed a brief demographic survey.

Maslach’s multidimensional burnout framework guided the development of the interview guide. Following the guidelines outlined by Kallio et al. [19], questions were intentionally crafted to explore experiences of emotional exhaustion (e.g., emotional demands and fatigue), depersonalization (e.g., emotional distancing or strain in client relationships), and professional accomplishment (e.g., perceived impact and meaning of CHW work), while giving participants flexibility to introduce unexpected themes.

Sample size justification and saturation

The sample size was determined based on the principle of information power, which suggests that the more relevant information participants provide regarding a specific study goal, the fewer participants are needed [20]. Given the precise aim, the relatively homogeneous professional group, and the use of an established theoretical framework, a modest sample size was suitable. Empirical research shows that code saturation often occurs after about 9 interviews in focused qualitative studies [21], with systematic reviews supporting small samples once analytical depth is reached [22]. After the tenth interview and fifth observation, no new substantive codes or themes appeared, and further analysis confirmed stability in the coding framework, indicating thematic saturation.

Ethical considerations

Ethical approval was obtained from Hofstra University’s Institutional Review Board (Reference number: 20240320-PH-HPHS-ACH-1). Written informed consent was obtained from all interview participants. For participant observations, verbal and/or written consent was obtained from CHWs and organizational supervisors prior to observation. When observations occurred in group settings (e.g., team meetings or case conferences), all individuals present were informed of the researcher’s role and the purpose of the observation. No identifying information about clients or non-consenting individuals was recorded. The study adhered to the Declaration of Helsinki.

Reflexivity

Reflexivity was maintained throughout the research process. The authors acknowledged their positionality as public health researchers with both insider familiarity and potential power asymmetries. Reflexive journaling was used after interviews to document assumptions, emerging insights, and potential biases, which were revisited during team analysis meetings to support interpretive transparency. This is consistent with recommendations that qualitative researchers explicitly articulate methodological positioning and analytic decisions [18].

Data analysis

Audio recordings were transcribed verbatim using an AI-powered transcription service and reviewed for accuracy by two research assistants. All transcripts and field notes were de-identified prior to analysis. Data were analyzed thematically using Braun and Clarke’s [23] six-phase framework and managed with ATLAS.ti (version 18). An initial codebook was developed inductively from early transcripts and refined iteratively through team discussion. All transcripts were independently coded by the three authors, with consensus reached through negotiated agreement. Discrepant cases were retained to refine conceptual boundaries. Observation field notes were coded using the final codebook and integrated into theme development to support triangulation. All transcripts were independently coded by three authors. Coding discrepancies were discussed in analytic meetings and resolved through negotiated consensus. This collaborative process minimized premature closure and deepened interpretation [24].

Although coding was inductive, Maslach’s burnout framework was applied deductively during later stages of analysis to support interpretation. Higher-order themes were examined in relation to emotional exhaustion, depersonalization, and professional accomplishment. This analytic strategy reflects qualitative description in integrating existing conceptual frameworks while maintaining descriptive fidelity [17].

The study adhered to the Consolidated Criteria for Reporting Qualitative Research (COREQ), thereby ensuring transparency across the study design, participant selection, reflexivity, data collection, and analysis [20].

Rigor and trustworthiness

Trustworthiness was established using Lincoln and Guba’s [25] criteria of credibility, transferability, dependability, confirmability, and authenticity, as well as contemporary guidance on methodological coherence [26]. Credibility was improved through triangulation of interviews and observations, independent coding by multiple researchers, and negotiated consensus during analysis meetings. Discrepant cases were kept, refining the thematic boundaries. Transferability was supported through detailed descriptions of the suburban context, participant characteristics, and organizational settings. Dependability and confirmability were strengthened by maintaining an audit trail of coding decisions and codebook updates, as well as through reflexive journaling to reveal assumptions and reduce bias. Findings were grounded in verbatim quotations to ensure transparency in analysis. Authenticity was supported by depicting both shared and differing participant experiences.

Participant demographics

Ten (10) community health workers (CHWs) participated in this study, representing diverse experiences and client populations across Nassau and Suffolk Counties, New York. Participants were predominantly women (approximately two-thirds) and ranged in age from their late twenties to mid-forties, with most (86%) between 35 and 44 years. The majority (85.8%) had between two and five years of experience as CHWs, and nearly all (86%) worked primarily in Nassau County, serving low-income families, mothers, older adults, and Medicare beneficiaries. Most participants were bilingual, with Spanish being the most frequently used language other than English. All participants reported working within their communities of residence, underscoring the relational proximity between CHWs and their clients. Three participants did not complete the post-interview demographic questionnaire due to time or technical limitations; however, their narratives were included in the thematic analysis, which focused on the full set of qualitative data.

Although ten CHWs participated in interviews, only seven completed the brief post-interview demographic survey (Table 1). The remaining three participants declined to complete the form due to time constraints immediately after their interviews or technical difficulties with the electronic link. All ten interviews, however, were fully recorded and transcribed, providing rich qualitative data for analysis. The missing demographic information did not affect the integrity of the findings because thematic saturation was achieved across interviews, and themes were derived from the full qualitative dataset rather than from demographic variables. Demographic summaries were used only for descriptive context, not as analytic strata influencing coding or theme generation.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Demographic characteristics of study participants.

https://doi.org/10.1371/journal.pone.0346430.t001

Overview of themes and sub-themes and coding structure

Analysis followed Braun and Clarke’s six-phase framework. Initial open coding generated 132 codes, which were iteratively grouped into six subthemes and two overarching themes through team-based analysis. Subthemes represent patterned responses across participants, while themes capture higher-order organizing concepts grounded in participants’ accounts (Table 2). When interpreted through Maslach’s burnout framework, Theme 1 primarily reflects strategies to address emotional exhaustion, whereas Theme 2 captures organizational resources that mitigate depersonalization and reinforce professional accomplishment. Participant observations corroborated interview findings by documenting real-time boundary-setting practices, informal peer support exchanges, and supervisory interactions, and in some cases, revealed emotional strain and task-sharing behaviors that participants normalized or minimized in interviews.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Identified interview themes and sub-themes.

https://doi.org/10.1371/journal.pone.0346430.t002

• Theme 1: Creating a Healthy Work–Life Balance (individual strategies)
  • Subtheme 1.1: Establishing boundaries
  • Subtheme 1.2: Practicing self-care
• Theme 2: Building Support Structures in the Workplace (organizational strategies)
  • Subtheme 2.1: Case conferences provide support
  • Subtheme 2.2: Active supervision
  • Subtheme 2.3: Coworker collaboration and task-sharing
  • Subtheme 2.4: Access to additional educational opportunities

To enhance analytic transparency, the process of theme development followed an inductive–deductive approach. Initial open codes were generated line by line from each transcript to capture participants’ language and meanings. Similar codes were clustered into categories that represented broader patterns across interviews (e.g., “emotional depletion,” “work–home spillover,” “peer learning,” “training access”). These categories were then compared and refined through iterative team discussions to identify subthemes, defined as coherent clusters of related codes that reflected distinct aspects of the participants’ coping strategies. Subthemes were then organized under higher-order themes that captured the overall conceptual structure of burnout prevention among CHWs. Field notes and observation data were used to confirm or elaborate subthemes, ensuring analytic triangulation. A reflexive audit trail documenting coding decisions and revisions was maintained throughout analysis.

Theme 1: Creating a healthy work–life balance

Participants consistently described burnout as emerging when emotional demands exceeded their capacity to disengage from clients’ needs. In response, CHWs developed intentional strategies to regulate emotional labor, primarily through boundary-setting and self-care. These practices were framed not as personal preferences but as professional necessities learned over time.

Theme 2: Building support structures in the workplace

While individual strategies were critical, participants consistently emphasized that burnout mitigation depended on organizational conditions that distributed emotional labor and normalized support-seeking. Workplace structures shaped how CHWs processed emotionally complex cases, navigated ethical dilemmas, and sustained professional motivation.

Study recommendations

The findings of this study underscore that while individual coping strategies are important for mitigating emotional exhaustion, burnout among community health workers (CHWs) is fundamentally shaped by organizational and structural conditions. Accordingly, effective burnout prevention requires action across multiple levels, including individual practice, organizational accountability, and broader policy and funding structures.

Individual-level strategies

CHWs described boundary-setting, self-care, and peer advocacy as important strategies for managing emotional strain. These practices, such as setting limits on availability, seeking peer support, and prioritizing personal well-being, can help CHWs regulate emotional labor and sustain engagement in demanding roles. However, these strategies should be understood as supportive coping mechanisms rather than substitutes for institutional responsibility. Expecting CHWs to manage burnout primarily through individual resilience risks normalizing structural inadequacies and obscuring the role of organizations and systems in producing emotional overload.

Organizational-level obligations

Organizations employing CHWs bear primary responsibility for creating conditions that prevent burnout. The findings highlight the importance of active and emotionally attuned supervision, structured case conferences, and formal peer support mechanisms that redistribute emotional labor rather than individualize it. Organizations should institutionalize protected time for reflection and debriefing, establish clear role boundaries, and integrate self-care and boundary-setting into formal training and supervision rather than treating them as personal choices. Additionally, organizations should ensure equitable access to ongoing professional development and recognize emotional labor as a core component of CHW work within job descriptions and performance expectations.

Structural and policy-level interventions

At the structural level, burnout prevention requires sustained investment in the CHW workforce and reforms that address fragmented service systems, particularly in suburban contexts. Policymakers and funders should prioritize stable, long-term funding streams that support adequate staffing, supervision, and training rather than short-term, grant-dependent models that exacerbate job insecurity and emotional strain. Workforce policies should explicitly fund supervisory roles, peer support infrastructure, and continuing education as essential components of CHW programs. In addition, efforts to improve inter-agency coordination and reduce service silos, such as shared referral platforms or integrated care networks, can alleviate the coordination burden currently absorbed by CHWs. Addressing burnout at this level shifts responsibility from individual workers to the systems that structure their labor.

Study limitations

Several limitations should be acknowledged when interpreting the findings of this study. First, as a qualitative study relying on self-reported data, participants’ accounts may be influenced by social desirability or recall bias, despite efforts to build rapport and ensure confidentiality. Second, the use of purposive and snowball sampling may have resulted in a relatively homogeneous sample drawn from overlapping professional networks, potentially limiting the range of perspectives captured, particularly those of CHWs working in less supportive or more precarious organizational settings. Third, the heavy concentration of participants in Nassau County may have shaped the findings toward experiences specific to that service and organizational context, underrepresenting CHWs working in more rural or resource-constrained areas of Suffolk County. This study did not include standardized measures of mental health conditions; therefore, references to emotional distress reflect participants’ self-described experiences rather than clinically diagnosed anxiety or depression. Lastly, many participants were embedded in organizations that offered active supervision, peer collaboration, and professional development opportunities. While these contexts provided important insight into resilience-building practices, they may also have introduced a positive bias toward organizational support strategies, potentially amplifying their prominence in the findings.

Taken together, these factors underscore the context-specific nature of the results and suggest that the identified themes reflect how burnout is navigated within relatively supportive suburban CHW environments rather than a comprehensive depiction of all CHW working conditions. Nevertheless, triangulation through field observations and multi-coder analysis enhanced the credibility and trustworthiness of the results. Future research should incorporate larger and more socioeconomically diverse CHW samples and mixed-methods designs to validate and extend these findings.