Introduction

Childhood and adolescence are times of heightened stress, with the majority of lifetime mental health (MH) disorders developing before adulthood (Kessler et al., 2005). When MH disorders early in life are left untreated and extend into adulthood, results include poorer academic outcomes (Woodward & Fergusson, 2001), higher rates of avoidable hospitalization and emergency department use (Bardach et al., 2014), and substance use (Winstanley et al., 2012). Untreated MH disorders are also major risk factors for suicide, the second leading cause of death for 15- to 24-year-olds (U.S. Department of Health and Human Services, 2017). Fortunately, early intervention has been shown to improve these outcomes (Wolk et al., 2015). Multiple evidence-based treatments have been developed (Weisz et al., 2006), yet, only 30-50% of youth access necessary treatment (Merikangas et al., 2011). The gaps between the need for and access to services disproportionately impact racial and ethnic minority and low-income, Medicaid-enrolled youth – groups found to have higher levels of unmet MH needs than White, higher-income youth (Alegria et al., 2010; Cunningham & Freiman, 1996; Satcher, 2001; Yeh et al., 2003).

A range of barriers to accessing MH care exist, including cost, lack of insurance coverage and acceptance, and long wait times (Carrillo et al., 2011; Schraeder & Reid, 2015; Substance Abuse and Mental Health Services Administration (SAMHSA), 2021). Early evidence suggests that the risk factors related to the COVID-19 pandemic and lockdown (e.g., isolation, grief) may have increased the development and exacerbation of MH symptoms (Golberstein et al., 2020; Panchal et al., 2020; Yao et al., 2020). While the pandemic and the sudden shift to telehealth services may have improved access in some ways, it also brought a reduction in school-based services, as well as privacy and technology constraints associated with telehealth (Golberstein et al., 2020; Svistova et al., 2021). Nonetheless, how exactly the onset of the COVID-19 pandemic, and the resulting shift to telehealth services, impacted the availability and accessibility of outpatient MH, particularly youth services, is not clear (Purtle et al., 2022).

Safety-Net Health Centers

With more than one out of three American youth (39%) insured by Medicaid (Kaiser Family Foundation, 2019), Community Mental Health Centers (CMHCs) and Federally Qualified Health Centers (FQHCs) play a large role in facilitating access to treatment (Panchal et al., 2020). Providers at these health centers offer a range of evidence-based psychosocial treatments for adolescents. Moreover, these sites act as the “de-facto safety net system for youth with MH disorders” (Cummings et al., 2016, p. 718) because the majority accept Medicaid, contrary to MH providers in private practice (Bishop et al., 2014; Rhodes et al., 2009). FQHCs, in particular, have been touted as a key to improving access to MH services. FQHCs are placed in medically underserved areas; mandated to accept Medicaid and prospective clients without insurance; required to offer “enabling services,” such as interpreters and a sliding fee scale; and can offer “integrated” physical and MH services (SAMHSA, 2013). Despite their promise, evidence suggests that FQHCs lack enough staff trained to serve youth with severe MH disorders (Bonilla et al., 2021; Cummings et al., 2013, 2020). As for CMHCs, while they offer a comprehensive array of services for youth with MH disorders, they are not mandated to serve youth enrolled in Medicaid or offer enabling services such as a sliding fee scale (SAMHSA, 2013), potentially reducing their accessibility for low-income youth. As each organizational type (FQHC vs. CMHC) experiences different funding streams, mandates, and incentives (e.g., FQHCs can receive enhanced reimbursement rates for accepting Medicaid clients; Centers for Disease Control, 2014), it is essential to understand what the availability and accessibility of outpatient MH services look like in each organizational type.

Patient-Centered Access to Care Framework

The Patient-Centered Access to Care Framework (Levesque et al., 2013) was selected to inform this study as it synthesizes published literature on the conceptualization of access. Borrowing from this framework, availability is assessed in the following way: (1) whether the health center offers MH services and is accepting new clients, (2) the number of clinicians who can offer services, and (3) the delivery format services are offered in. Accessibility is assessed regarding: (1) timeliness of service delivery (wait time), (2) types of insurance accepted, (3) cost of services, (4) other organizational factors, such as the requiring of referrals before receipt of treatment, and (5) languages services are offered in and the availability of interpreter services. The variables being studied have been adapted from the original framework to ensure they fit the context of safety-net health centers.

This descriptive study aims to assess the availability and accessibility of outpatient MH services for youth in one major metropolitan county in the United States a year after the COVID-19 pandemic began. Drawing on phone survey data from March and April of 2021, it posed the following research questions:

  1. 1)

    How many health centers listed in prominent online directories (e.g., SAMHSA Treatment Locator) are closed? How many health centers offer outpatient MH services to children and adolescents?

  2. 2)

    What do the availability (e.g., number of clinicians) and accessibility (e.g., wait times) of MH services look like in CMHCs and FQHCs during the COVID-19 pandemic?

Methods

Stage 1: Initial Sample Development from the SAMHSA Treatment Locator

Safety-net health centers in Cook County, Illinois, were identified using the SAMHSA Treatment Locator in November of 2020, a publicly available search tool where health centers can be filtered according to relevant patient factors (e.g., geographic area they are in). Cook County, IL, was selected as it is the second most populous county in the U.S. (Cook County Government, n.d.). The SAMHSA Treatment Locator is populated annually by the National Mental Health Services Survey (N-MHSS). The N-MHSS collects information from all known facilities in the United States that provide services for MH disorders. In Illinois, the response rate for the 2019 N-MHSS was 92% (SAMHSA, 2018). The initial sample was determined by selecting (1) “Cook County, Illinois” for location, (2) “MH” for service and type of care, and (3) the “Health Care Centers” option was selected, which prompts the Treatment Locator to include FQHCs in the search. This search resulted in a sample of 357 health centers. Health centers were considered eligible if they were listed as offering outpatient MH services to children and adolescents (< 18 years old). Health centers were stratified into two categories: FQHCs (n = 249) and CMHCs (n = 108).

Stage 2: Validation of Initial Sample Against Additional Sources

To ensure the completeness and accuracy of the initial sample of safety-net health centers, each list was validated against other directories. The initial sample of FQHCs was validated against the Uniform Data System (UDS, 2018), an online portal operated by the Health Resources and Services Administration that lists up-to-date information on FQHCs. The list of CMHCs was validated against: (1) the Chicago Department of Public Health’s online directory of MH clinicians (CDPH, n.d.) and (2) the Illinois Department of Human Services (IDHS) list of outpatient MH centers (Illinois Department of Human Services, 2021).

Twenty-eight FQHCs (11.2%) were removed from the initial sample as the UDS noted that they do not offer outpatient MH services (e.g., administrative site only, vision services only), and sevenFQHCs (2.8%) were removed because the UDS indicated they were closed. Next, two FQHCs were added from the UDS to total 216 FQHCs. A total of 22 CMHCs (20.4%) were removed since they were listed as only offering MH services to adults, and 23 CMHCs (21.3%) were removed because they only offered inpatient or residential services. Next, 92 CMHCs that were not listed in the SAMHSA Treatment Locator but were listed in the CDPH or IDHS lists were added to the sample, making a total of 155 CMHCs. After validating the initial sample against these additional sources, the analytic sample consisted of 371 health centers (n = 216 FQHCs, n = 155 CMHCs).

Stage 3: Survey Administration and Screening for Eligible Health Centers

The research team called each health center in the final analytic sample between March–April 2021, approximately one year after the pandemic impacted the United States. They asked to speak to the MH director or intake coordinator to administer a voluntary 5-minute survey on the availability and accessibility of MH services their health center offered (see Appendix 1.1 for the survey). The survey was investigator-created in partnership with community members within the Collaborative for Community Wellness. Availability was operationalized as whether health centers offered MH services (binary), the delivery format of services (in-person, telehealth, or both), and the number of clinicians (continuous). Accessibility was operationalized as timeliness of services (continuous), insurance (binary), whether free services or sliding scale were offered (binary), cost (continuous), organizational factors (e.g., whether the agency offered transportation services, binary), whether the health center offers services in Spanish or any other language other than English (binary), and if so, which languages they are offered in (descriptive), as well as how many therapists offer services in Spanish (continuous). To view question wording, see Appendix 1.1 for full survey. At least three call attempts to each health center were made, separated by at least 24 hours. After three call attempts, seven FQHCs (6.0%) and six CMHCs (5.1%) could not be reached but were in operation. For these health centers, the team completed their survey using publicly available information online. Upon attempting to contact them and after an online search, an additional 26 FQHCs (12.0%) and 11 CMHCs (7.1%) were determined to be closed. In total, 37 centers (10.4%) in the initial sample were closed. Regarding response rates, out of agencies that could be reached, no health center declined participation.

Two screener questions were asked to ensure health centers were eligible to be administered the full survey: (1) Does your health center offer outpatient MH services? and (2) Does your health center offer outpatient MH services to children and adolescents? Upon administering the screener questions, 61 (28.2%) FQHCs and 12 (7.7%) CMHCs reported that they did not offer outpatient MH services over the phone. When health centers reported that they did not offer outpatient MH services, we asked them about the primary service their health center did offer. The majority of FQHCs reported offering only primary care services (n = 48), and CMHCs reported primarily offering only inpatient or residential services (n = 3), substance use services (n = 3), or crisis services (n = 2). Further, 12 (5.6%) FQHCs and 15 (9.7%) CMHCs reported over the phone that they only offered outpatient MH services to adults. These health centers were excluded from the study sample as they did not fit the eligibility criteria. Only health centers that reported offering outpatient MH services to children and adolescents were eligible to be administered the full survey.

Statistical Analysis

Analyses involved descriptive statistics, including frequencies and percentages, to describe the availability and accessibility of outpatient MH services during the COVID-19 pandemic. Results were stratified for FQHCs and CMHCs. Chi-square and t-tests were conducted to determine the relationship between various dependent variables and organizational type. Data were analyzed using SPSS version 28.0. (IBM Corp., 2021). Missing data were handled by omitting cases from analyses using listwise deletion. The University of Chicago Crown Family School of Social Work, Policy, and Practice and Chapin Hall Institutional Review Board determined that this research is exempt, given the minimal risk level to participants. The author has no conflicts of interest to report and certifies responsibility for the manuscript.

Results

Availability of MH Services

After excluding health centers due to ineligibility, the full survey was administered to 117 FQHCs and 117 CMHCs (N = 234). One in four  (26% n = 30) CMHCs were affiliated with hospitals, while 74% (n = 87) were community-based. Overall, less than half of the health centers (42.3%, n = 99) that offered therapy services also offered psychiatry or medication management services. Eighteen CMHCs required that clients receive therapy at their health center before receiving a referral for psychiatry services; no FQHCs reported a similar requirement. Most health centers (96.6% of FQHCs and 82.9% of CMHCs) were accepting new children and adolescent clients (X2(1) = 11.123, p < .001).

Availability – Number of Clinicians

Overall, health centers reported an average of six MH clinicians (SD: 8.19, n = 203), five of which specialized in providing services to children and adolescents (SD: 6.83, n = 185). CMHCs had an average of 8.73 MH clinicians on staff (SD: 9.28, n = 98), seven of whom specialized in providing services to children and adolescents (SD: 8.97, n = 84). FQHCs had an average of 3.26 MH clinicians on staff (SD: 5.92, n = 105), two of whom specialized in providing services to children and adolescents (SD: 2.47, n = 101). There was a statistically significant difference in the mean number of MH clinicians on staff between CMHCs and FQHCs (t162.605 = 4.960, p < .001) and the mean number of MH clinicians who specialized in serving children and adolescents (t93.392 = 5.172, p < .001). On average, FQHCs had 5.47 fewer clinicians on staff than CMHCs, and the average number of MH clinicians who specialized in serving children and adolescents was 5.23-clinicians lower at FQHCs than CMHCs.

Availability – Delivery Format of Services

A year after the COVID-19 pandemic began, only 3% of health centers overall were offering only in-person services (n = 7). Across both organizational types, most health centers reported offering only telehealth services (46.2%, n = 108) or both in-person and telehealth services (44.0%, n = 103). Many health centers that offered both telehealth and in-person services said that in-person services were limited and had to be approved by management (e.g., clients who did not have access to broadband internet).

Accessibility – Timeliness of Services

When health centers reported that they offered both in-person and telehealth services, the research team asked for wait time estimates for both. Across both organizational types, the overall median wait time was 28 days for telehealth services (SD: 66.48, n = 171) and 28 days for in-person services (SD: 55.01, n = 103). The median reported wait time for telehealth services was 60 days at CMHCs (SD: 82.24, n = 88) compared to 15 days at FQHCs (SD: 26.41, n = 83). For in-person services, the median reported wait time was 60 days at CMHCs (SD: 70.00, n = 45) compared to 21 days at FQHCs (SD: 21.83, n = 58). There was a statistically significant difference in mean wait time for in-person services between CMHCs and FQHCs (t50.662 = 4.710, p < .001) and mean wait time for telehealth services (t105.671= 5.334, p < .001). Within CMHCs, centers affiliated with hospitals reported longer wait times for telehealth services than centers not affiliated with hospitals (t20.113= 1.881, p < .037), with average wait times at hospital-affiliated CMHCs being 57.52 days longer. Eight CMHCs reported that their waitlist was closed, with no estimate of when it would reopen. For medication management/psychiatry services, there were no statistical differences in wait times between organizational types, with a median of 21 days (SD: 43.25, n = 34) at FQHCs and 30 days (SD: 42.93, n = 30) at CMHCs (t62.000= 0.831, p = .409). Eight FQHCs and four CMHCs reported their waitlist for new psychiatry clients was closed.

Accessibility – Insurance

Given that we purposively sampled safety-net health centers, it is no surprise that 93.2% of CMHCs (n = 108) and 100.0% of FQHCs (N = 117) accepted at least one Medicaid managed care organization (MCO) plan (X2(1) = 7.343, p = .007). Only two-thirds (65.8%, n = 77) of CMHCs accepted at least one private insurance plan compared to 100% of FQHCs (N = 117; X2(1) = 41.071, p < .001). CMHCs were less likely to accept uninsured clients (75.2%, n = 88) than FQHCs (94.9%, n = 111; X2(1) = 25.658, p < .001).

Accessibility – Cost

Approximately 27.4% (n = 32) of CMHCs offered free MH services, while only 4.3% (n = 5) of FQHCs offered free MH services (X2(1) = 24.492, p < .001). On the other hand, most FQHCs (98.3%, n = 115) offered a sliding fee scale, while only 59.0% (n = 69) of CMHCs did (X2(1) = 47.143, p < .001). On average, the lowest rate on FQHCs sliding scales was $21 per session (SD: 9.81). At CMHCs, the lowest rate on the sliding fee scale was $35 per session (SD: 46.95). There was a statistically significant difference between the lowest rate on the sliding fee between FQHCs and CMHCs (t54.240= 2.132, p = .038). Additionally, many CMHCs put restrictions on their sliding scale and free services. For example, some centers stated their sliding fee scale could only be offered to individuals without insurance or if they lived in a specific catchment area. Further, most CMHCs noted their free services were only available for specific populations (such as domestic assault survivors), given restrictions from grant funding.

Accessibility – Organizational Factors

Approximately two-thirds (65.0%, n = 76) of FQHCs required prospective clients to have a referral before scheduling a MH appointment; only 7.7% (n = 9) of CMHCs had the same requirement (X2(1) = 84.649, p < .001). Further, more than half (54.7%, n = 64) of FQHCs required prospective clients to switch their primary care provider (PCP) into their health center network before they would schedule a MH appointment, compared to 4.5% of CMHCs (n = 5), all of which were hospital-based (X2(1) = 69.780, p < .001). Approximately one-third (34.2%, n = 40) of FQHCs reported the ability to assist clients with at least one form of transportation to their appointments (i.e., bus pass, rideshare, or private transportation) compared to 22.2% (n = 26) of CMHCs (X2(1) = 4.180, p = .041).

Accessibility – Language

Across all health centers, around half (53.8%, n = 126) had at least one Spanish-speaking MH clinician, with an average of three Spanish-speaking clinicians on staff (SD: 3.61). Approximately 9.8% (n = 23) of all health centers offered MH services in a language other than English or Spanish, and around half could offer language interpreter services (53.8%, n = 126). Differences between organizational types were noted for language-based accessibility factors. Around two-thirds of CMHCs (64.1%, n = 75) had at least one MH clinician on staff who could provide services in Spanish, whereas less than half (43.6%, n = 51) of FQHCs had at least one Spanish-speaking MH clinician (X2(1) = 5.389, p = .020). On average, CMHCs had four MH clinicians who could offer MH services in Spanish (SD: 4.42), whereas FQHCs had one (SD: 0.60; t62.423= 3.577, p < .001). Further, 17.1% of CMHCs (n = 20) had MH clinician(s) who could speak a language other than English and Spanish. Only three FQHCs were able to offer MH services in other languages. Instead, most FQHCs (97.4%, n = 114) offered language interpretation services to accommodate individuals who speak other languages. Few CMHCs offered language interpretation services (10.3%, n = 12).

Discussion

Due to various barriers (e.g., wait times, insurance status, family income), youth who need MH services often do not receive them. Given school disruptions and increasing demand for services, the COVID-19 pandemic likely exacerbated this disparity. This study’s findings provide insight into the availability and accessibility of outpatient MH services for youth in safety-net health centers one year after the COVID-19 pandemic began.

This study contributes to the literature by clearly identifying inaccuracies and gaps in information in the SAMHSA Treatment Locator, a prominant online, freely accessible directory of MH centers across the US used by researchers, policy makers, public health practioners, and families seeking care. In the current study, approximately 10% of safety-net health centers listed in the SAMHSA Treatment Locator were closed or could not be located online. This finding is balanced against the finding that, in validating the initial sample against additional sources, another 92 CMHCs not listed on the SAMHSA Treatment Locator were identified. Further, about 20% of the final sample (28.2% of FQHCs and 7.1% of CMHCs) reported not offering outpatient MH services, despite multiple online directories listing them as being offered. A significant number of peer-reviewed studies use the SAMHSA Treatment Locator as their primary source for identifying and characterizing the mental health system in the US (e.g., Cantor et al., 2021; Cummings, Wen, Ko, & Druss, 2013; Cummings et al., 2022; Guerrero & Kao, 2013; Shrader et al., 2023). The findings from this study suggest that online directories of MH centers (e.g., SAMHSA Treatment Locator), when used for research purposes or otherwise, should be validated against multiple sources to ensure the most accurate sample. While recent literature has found mixed results on the accuracy of the SAMHSA Treatment Locator for substance use services (Anyanwu et al., 2022; Flavin et al., 2020), this study represents one of the first of its kind to highlight inaccuracies for MH services.

Around 10% of health centers that offered outpatient MH services only served adults. Aligned with their mission to specialize in MH services, CMHCs had significantly more MH clinicians on staff than FQHCs (mean of 9 vs. 3 clinicians). CMHCs also had more MH clinicians who could provide services in Spanish or another language other than English than FQHCs. Complying with a federal mandate to offer “language assistance services” (Denson & Graves, 2022, p. 34) the majority of FQHCs offered a service where clinicians call a professional language interpretation line, increasing their ability to serve individuals who speak any language. As professional interpreters are considered “best practice” compared to “ad hoc” interpreters (i.e., interpreters without training), CMHCs should consider offering professional interpreters to expand access to individuals with limited English proficiency (Karliner et al., 2007).

Despite CMHCs having more specialty MH clinicians on staff (indicating higher capacity), their reported wait times were significantly longer FQHCs (median of 60 days vs. 15 days). One interpretation of this finding is that demand for MH services may be higher at CMHCs than FQHCs. Parents might be more likely to intuitively seek MH services for their children at specialty health centers such as CMHCs rather than FQHCs, as FQHCs tend to be better known in the community for the medical services they provide. Increasing community knowledge that FQHCs offer MH services may increase demand. That said, more than half FQHCs required a referral and/or an internal PCP before they would schedule a MH appointment. A similar finding was reported in a mystery shopper of medical clinics in Illinois conducted in 2019 as a part of an external quality review (Illinois Department of Healthcare and Family Services, 2020). The report noted that requiring families to designate their PCP with the FQHC through insurance before scheduling an appointment acted as a “barrier to all Medicaid enrollees trying to schedule appointments” (Illinois Department of Healthcare and Family Services, 2020, p. 60). These administrative tasks may prevent families from being added to the waitlist, possibly keeping waitlists shorter at FQHCs. While these administrative requirements are well-intentioned tasks meant to better integrate physical and MH care, they could pose an additional barrier for families seeking treatment. Future research should explore how demand differs at each organizational type and how families experience administrative requirements implemented by FQHCs.

Another potential reason for this extreme difference in reported wait times between organizational types is that CMHCs were more often hospital-based than FQHCs. Previous research has found that hospital-based clinics had longer wait times than community-based clinics (Olin et al., 2016). Finally, this disparity in reported wait times could be due to different treatment modalities offered by each organizational type. For example, FQHCs provide routine and preventative MH treatment that is often short-term in nature (Jacobs & Steiner, 2016). By providing shorter-term therapy, FQHCs may be able to see more clients than CMHCs, in turn, shortening wait times. They may also be less equipped to serve youth with severe MH disorders, referring out clients who present with these symptoms (Cummings et al., 2020). Combined, these factors may help keep clinician caseloads low at FQHCs and increase their ability to move through waitlists faster. Results indicate that the pandemic may have had an impact on wait times for outpatient MH services. A survey conducted in 2016 by the CDPH found that the average wait time for individual psychotherapy at community-based clinicians was 57 days (range 0-365 days) (Chicago Department of Public Health, 2016), compared to an average of 73 days (range 0-365 days) at CMHCs in this study. While a comparison between these wait times should be interpreted lightly, the wait time estimates seen in this study are a cause for alarm.

Results indicate that the cost of outpatient MH services varies by organizational type. Aligned with funder mandates, FQHCs offered sliding fee scales, interpreter services, and transportation assistance more often than CMHCs. They were also more likely to accept Medicaid, private insurance, and uninsured individuals than CMHCs. These results are encouraging as they all allow FQHCs to meet their stated mission of making MH services more accessible for marginalized populations. In contrast, CMHCs are under no such mandates from the federal government to accept any particular insurance types or to offer enabling services. CMHCs were less likely to offer enabling services such as sliding fee scales and less likely to accept certain insurance types or individuals without insurance.

Results signal intervention points to increase access for each organizational type. At CMHCs, wait times were very long. One intervention shown to reduce wait times in FQHCs is the collaborative care model, an evidence-based approach that includes case management, care coordination, and measurement-based care (Kinnan et al., 2019). Future research should assess the feasibility and acceptability of implementing collaborative care models in CMHCs, particularly hospital-based CMHCs, which tend to have the longest waiting times. While wait times at FQHCs were shorter than CMHCs, they had fewer MH clinicians overall, reducing their promise to increase MH access to the most marginalized communities. Increasing the capacity of FQHCs by adding more full-time equivalent MH clinicians on staff has been shown to increase access to MH treatment (Bonilla et al., 2021). Policymakers should also consider expanding loan repayment programs, easing the administrative burden of participating in the Medicaid program, and increasing Medicaid reimbursement rates to help health centers hire and retain staff, increasing their capacity to treat prospective clients in a timely manner (Adams et al., 2019; Long, 2013; Olfson et al., 2014).

Limitations

Several limitations should be noted. First, the survey was self-report (i.e., not subject to independent verification) and could have been subject to social desirability and recall bias. In future studies, wait time should be assessed more rigorously through mystery shopper methodologies to provide more accurate, real-time estimates (Olin et al., 2016). Second, we cannot determine just how much the pandemic has impacted the availability and accessibility of services as data is cross-sectional and only collected at one point in time. We did not ask what the availability and accessibility of services looked like prior to the pandemic. Notably, demand for MH services increased during the COVID-19 pandemic when this study was conducted (Byrne et al., 2021). It is difficult to disentangle the impact of the pandemic on the adverse-care patterns noted in this study, such as long wait times. We only asked about a point in time estimates of availability and accessibility. Finally, while Cook County is the second most populous county in the U.S. (Cook County Government, n.d.), reducing this concern, this study is limited to one county and may not generalize to regions with different behavioral health financing or policies.

Despite these limitations, this study represents the first to assess the availability and accessibility of outpatient MH services for youth during the COVID-19 pandemic. In this study, the availability and accessibility of outpatient MH services varied by organizational type, with longer wait times at CMHCs and fewer MH clinicians at FQHCs. Further, findings from this study indicate that gold-standard online directories such as the SAMHSA Treatment Locator are often inaccurate or out-of-date. The data from this study provide a meaningful benchmark for policymakers and safety-net health centers that aim to increase access to MH care coming out of the pandemic.