Medicaid Reimbursement of Mental Health Peer-Run Organizations: Results of a National Survey

Abstract

This study sought to understand whether knowledge of the Affordable Care Act (ACA) was associated with willingness of mental health peer-run organizations to become Medicaid providers. Through the 2012 National Survey of Peer-Run Organizations, organizational directors reported their organization’s willingness to accept Medicaid reimbursement and knowledge about the ACA. Multinomial logistic regression was used to model the association between willingness to accept Medicaid and the primary predictor of knowledge of the ACA, as well as other predictors at the organizational and state levels. Knowledge of the ACA, Medicaid expansion, and discussions about healthcare reform were not significantly associated with willingness to be a Medicaid provider. Having fewer paid staff was associated with not being willing to be a Medicaid provider, suggesting that current staffing capacity is related to attitudes about becoming a Medicaid provider. Organizations had both ideological and practical concerns about Medicaid reimbursement. Concerns about Medicaid reimbursement can potentially be addressed through alternative financing mechanisms that should be able to meet the needs of peer-run organizations.

Introduction

The Affordable Care Act (ACA) has provided our behavioral healthcare systems with the opportunity for significant reforms in quality of care, cost-containment, and models of service delivery. It will inevitably impact all models of service delivery, but community-based care organizations such as mental health peer-run organizations may be impacted in particular ways because of their non-medical orientation. Peer-run organizations are organizations and programs operated by people with lived experience of the mental health system (Ostrow and Leaf 2014). These organizations use advocacy and social support to promote recovery and empowerment for people with mental disorders (Goldstrom et al. 2006; Ostrow and Hayes 2015).

The ACA includes expansion of Medicaid coverage to a greater proportion of the population, if states elect to do so. Although the effects of full implementation of the Medicaid expansion policy is, as of yet, unclear, it is unquestionable that it will impact peer-run organizations because of the expansion of Medicaid coverage to a large number of people with mental disorders and the financing shifts that may occur in funding mental health services given expanded insurance coverage (Substance Abuse and Mental Health Services Administration 2013). Peer-run mental health programs are an important component in the continuum of care for individuals with mental disorders (New Freedom Commission 2003; Office of the Surgeon General 1999). They have been shown to be effective in supporting recovery outcomes such as community tenure, reduced inpatient use, and gains in empowerment and self-efficacy (Klein et al. 1998; Mental Health Weekly 2011; Min et al. 2007; Nelson et al. 2007; Rogers et al. 2007). In particular, the organizational structure and processes of peer-run organizations contribute to community-building and stigma-reduction (Segal et al. 2013). As an essential part of their mission of being recovery-oriented and non-medical, these organizations often do not comply with the type of documentation required for billing Medicaid.

The study reported here examined factors related to the acceptability of Medicaid reimbursement by mental health peer-run organizations through analyses of the 2012 National Survey of Peer-Run Organizations, and the organizational directors’ concerns about becoming a Medicaid provider.

Medicaid Expansion Under the ACA

The ACA intended to expand Medicaid coverage in all states to single, childless adults with incomes at or below 133 % (effectively 138 % because 5 % of income is not to be considered) of the federal poverty level (FPL). The Supreme Court ruling in 2012 made Medicaid expansion a state option, and not all states are expanding Medicaid at this time. Medicaid expansion as part of the ACA initially affords a 100 % federal medical assistance percentage (FMAP) of the cost of providing Medicaid coverage for those who are newly qualified. The FMAP will decline to 90 % after 2020 (Goldman 2012). The ACA resulted in planned expansion to twenty-seven states’ Medicaid programs as of 2013. Some states are still resisting despite the financial and human benefit incentives, but may alter their policy in the future to take advantage of the federal incentive to expand Medicaid coverage (Kaiser Family Foundation 2013).

According to SAMHSA, Medicaid expansion as part of the ACA could reach 2.7 million people who are currently uninsured and in need of mental health services in the states that chose to expand Medicaid (Substance Abuse and Mental Health Services Administration 2013). The Assistant Secretary for Planning and Evaluation (ASPE) estimated that 5.4 million individuals with a mental health or substance abuse disorder would gain insurance coverage through Medicaid expansion (Donohue et al. 2010) if all states implemented it. This larger figure of potentially covered beneficiaries is still important because the remaining states may still elect to expand Medicaid coverage.

Medicaid is an important payer nationwide for mental health services (NAMI 2010). Every state has behavioral health provisions in their Medicaid plan. Medicaid plans cover the cost of a wide variety of supportive services, including counseling, recovery supports, and skills training. More intensive services, such as Assertive Community Treatment or inpatient hospitalizations, are also covered. For each of these services, the specifics of coverage vary by state. Some only cover services such as medication management and short-term psychiatric inpatient hospitalization, rather than a broader array of wraparound services, despite evidence for wraparound services such as employment, education, and housing supports (Mechanic 2012). While the ACA mandated that states include mental health parity in their plans for the expansion population, we do not know to what extent similarly limited coverage for behavioral health services might also affect the Medicaid expansion population in particular states.

Potential Effects of Medicaid Expansion on Peer-Run Organizations

Currently, more than 75 % of peer-run organizations receive a majority of their operating funds from governmental sources such as state and local revenues and the SAMHSA Mental Health Block Grant (Ostrow and Hayes 2015). Plans are underway to reduce and substantially change the focus of the SAMHSA Mental Health Block Grant (Department of Health and Human Services 2011). The restructured block grants will focus more on coordinating behavioral health prevention, treatment, and recovery support services with other health and social services to fill the gaps that remain when more individuals have health care coverage for their mental health needs. This means that peer-run organizations’ usual sources of funding may be less available to support their operations (Ostrow and Hayes 2015).

The Centers for Medicare and Medicaid Services (CMS) encourages the use of ‘‘peer support’’ in the states under state Medicaid plan waivers, and has issued guidance for the development and implementation of these supports in recent years (Mann 2010). In many states, Medicaid covers peer support (Landers and Zhou 2014). Peer support encompasses services provided by ‘‘certified peer specialists’’. Many states have a certification process for certified peer specialists and the plurality of these workers are in independent peer-run organizations (Salzer et al. 2010). Additionally, the plurality of the workforce in peer-run organizations is certified peer specialists, although workers may be trained in (and paid for) other types of peer support that do not have a state-sponsored certification process (Ostrow and Hayes 2015). Peer support includes assistance in learning and overcoming challenges in health/wellness, and self-monitoring (IOM 2012).

The first Medicaid-reimbursable certified peer specialist credential was implemented in Georgia in 2001, and many other states have followed (Grant et al. 2010). Certified peer specialists are now Medicaid reimbursable in 31 states and the District of Columbia, as shown in Fig. 1. It is unknown which exact waiver mechanism is being used in each state. Because Medicaid is a federal-state partnership, states vary in what services are reimbursable in their state Medicaid plan and how they set standards for reimbursement.

Fig. 1.

Medicaid expansion and reimbursement of peer specialists, by state in 2014

Considerations in Medicaid Funding of Peer-Run Organizations

CMS and Medicaid managed care companies have made efforts to expand coverage for peer support and involve peer-run organizations in reimbursement (Adler et al. 2010; Daniels et al. 2013). Medicaid reimbursement through a fee-for-service model rather than as part of a managed care contract is seen as largely infeasible for these organizations because of the requirements of fee-for-service billing such as documentation of units of service that could be quite onerous for these programs. If peer-run organizations join a managed care network, it could provide the network with more diverse services to its clients, and also promote the ideas of recovery and resilience within the network. Already, many of the goals of managed care are those of peer support: increasing wellness and recovery, and reducing hospitalization. However, being ‘‘non-medical’’ is one of the essential features of peer support (Mead et al. 2001). For the most part, peer support does not involve using mechanisms such as diagnoses, and peer supporters are not trained to focus on or give diagnoses (Katz and Salzer 2006). Other aspects of Medicaid reimbursement that peer-run organizations may have difficulty with include billing and financial management, and complying with medical necessity criteria (Adler et al. 2010). These may be difficult for peer-run organizations because of their interest in focusing on strengths, not documenting impairments (Holter et al. 2004). Medicaid reimbursement of peer support could mean having to adhere to standards such as using diagnoses. However, while peer-run organizations may risk going out of business because of the challenges they may face in complying with requirements for insurance reimbursement and because of shrinking resources from their usual sources of financing, Medicaid expansion might provide opportunities for more sustainable funding and more reliable revenues that could also broaden the impact of these organizations.

The research reported here examines how peer-run organizations view Medicaid reimbursement given the new policy environment. We hypothesized that greater exposure to and knowledge about the ACA and awareness in shifts in the healthcare financing environment including expanded Medicaid coverage would positively influence the organizations’ willingness to become a Medicaid provider. Participants were also asked to note concerns related to many of the issues raised in previous literature on Medicaid funding of peer supports—including medical necessity criteria, administrative capacity for billing, and values related to medical-model treatment.

Methods

This paper uses data obtained from the first national survey of peer-run mental health programs to be published in over a decade—the 2012 National Survey of Peer-Run Organizations; more in-depth discussion of the study and methods is presented in an earlier paper (Ostrow and Leaf 2014). The target population for the study is peer-run organizations excluding independent mutual support groups that do not have a formal organizational structure. This has been a criterion in other peer-run organization research (Brown et al. 2007). Almost 900 organizations/ programs were identified through contacting statewide consumer networks and state offices of consumer affairs from August 2010 to June 2012. Data were collected via a web-based survey, completed online by the directors of the organizations, achieving an 80 % response rate with participants in 49 states and Washington, DC. The survey contained about 83 questions, although not all respondents were asked all questions, as they would ‘‘skip out’’ of sections if subsequent questions were not relevant. The survey was developed and had two rounds of pre- and pilot-testing with the input of mental health services researchers, program directors, advocates, and government representatives. Final inclusion criteria were related to management structure operationalized by a consensus panel. These criteria are described in an earlier manuscript (Ostrow and Leaf 2014). Following data cleaning according to these criteria, there were 380 organizations in the analyses. The research project was deemed not to be human subjects research by the Johns Hopkins Bloomberg School of Public Health IRB.

Outcome

Directors were asked: ‘‘Are you willing to become a Medicaid provider? A Medicaid provider is one that bills or submits claims for individuals served or services provided.’’ Response categories included: ‘‘Yes’’, ‘‘Yes, but we have some concerns’’, ‘‘No’’, ‘‘Don’t know’’, and ‘‘My organization/program is already a Medicaid provider.’’

Predictors

Directors were asked about their knowledge of the ACA: ‘‘How much do you know about the Affordable Care Act (ACA)?’’ with possible response categories ‘‘Nothing,’’ ‘‘A little,’’ ‘‘Some things’’, ‘‘A lot’’, ‘‘Everything’’, and ‘‘Don’t know.’’ The question intentionally did not define the ACA to better gauge respondents’ knowledge without giving them more information about the policy. Response categories ‘‘Don’t know’’ and ‘‘Nothing/A little’’ were collapsed because it was assumed that to not know if one knows the answer is similar to knowing very little about the question.

Participants were asked whether people within their organization had had discussions amongst themselves, within the Statewide Consumer Network, and with a government body about ‘‘health care reform.’’ The term ‘‘health care reform’’ was used instead of ‘‘ACA’’ to identify discussions about policies that may be beyond the scope of the ACA, pre-date it, and to identify more general conversations about how the health system may be changing in ways that affect peer services.

The directors were asked if they knew of plans in their state to expand Medicaid coverage to more people, as this may affect their attitudes toward accepting Medicaid reimbursement based on awareness of shifts in funding and insurance coverage under Medicaid. Actual plans for Medicaid expansion, as of 2013, were obtained from the Kaiser Family Foundation (Kaiser Family Foundation 2013).

Whether peer specialists are reimbursable through Medicaid was also used in the model as a predictor of willingness to become a Medicaid provider, because if organizations are in states where peer specialists are Medicaid reimbursable, it could affect how they perceive Medicaid reimbursement (positively or negatively)—given experiences with this funding mechanism. Information on Medicaid reimbursement of peer-delivered services was provided by SAMHSA (2011 NRI Profiles data and 2011 survey of NASMHPD’s Medicaid and Financing division). Organizational size was operationalized as the program director’s report of the total number of paid staff (Jaskyte, 2011).

A follow up question about specific concerns about Medicaid was asked of those who said they were willing to become a Medicaid provider but had concerns, those who said they were not willing, and those who said they were unsure. The concerns were constructed from key informant interviews with advocates, peer providers, government officials, and insurance company representatives (Ostrow and Leaf 2014) and existing guidance on adaptation of peer-run organizations for insurance reimbursement (Adler et al. 2010).

Statistical Analyses

Multinomial logistic regression was used to analyze the dependent variable, the reported willingness to become a Medicaid provider. Potential within-cluster correlation of responses by the state in which participants were located was controlled for in the final statistical model using a robust variance estimate (Rogers 1993). Respondents who reported they were already a Medicaid provider were excluded from analyses on willingness to become a Medicaid provider since the hypothesis was related to the influence of the ACA on organizations that are facing the choice of becoming a Medicaid provider.

Paid staff is a common metric of organizational size when discussing innovations (Jaskyte 2011); it is used here despite the number of volunteers many of these organizations have because the paid staff are the staff members who are currently using organizational financial resources that might be augmented or substituted by Medicaid reimbursement and would support this innovation in financing.

Concerns about Medicaid reimbursement endorsed by the participants are reported as overall frequencies to show which were most common among any participants who did not indicate they were willing to be a Medicaid provider without concerns, or were already a Medicaid provider. All analyses were conducted in Stata 13 (StataCorp, 2013).

Results

Table 1 presents frequency of director responses to the outcome variable of willingness to accept Medicaid and predictor variables related to exposure to the ACA and state-level characteristics.

Table 1.

Characteristics of exposure to the ACA by willingness to accept Medicaid

	Total (N = 346)		Already (N = 30)		Yes (N = 52)		Yes, concerns (N = 106)		No (N = 87)		Don’t know (N = 71)		χ 2	p
	N	%	N	%	N	%	N	%	N	%	N	%
Knowledge of ACA													27.07	<0.01
Nothing/a little	176	51.0	12	40.0	21	40.4	46	43.4	45	51.7	52	73.2
Some things	104	30.1	7	23.3	18	34.6	38	35.9	30	34.5	11	15.5
A lot	66	19.0	11	36.7	13	25.0	22	20.8	12	13.8	8	11.3
Knowledge of Medicaid expansion	181	52.5	14	48.3	36	69.2	65	61.3	43	49.4	23	32.4	21.19	<0.001
Discussed healthcare reform with…
Organization	187	54.4	20	66.7	30	58.8	65	61.9	43	49.4	29	40.9	10.73	<0.05
Statewide consumer network	147	42.9	12	40.0	29	56.9	55	52.4	30	34.45	21	30.0	15.29	<0.01
Government committee	95	27.7	9	30.0	22	43.1	36	34.3	18	20.7	10	14.3	16.85	<0.01
State medicaid peer specialist reimbursement	241	69.7	24	80.0	37	71.2	66	62.3	58	66.7	56	78.9	7.53	0.11
State plans to expand medicaid	129	37.3	14	46.7	20	38.5	33	31.1	35	40.2	27	38.0	3.22	0.52
	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	Mean	SD	F	p
Mean number of paid staff	12.72	40.51	59.23	125.14	11.2	12.2	10.3	14.1	6.2	7.5	5.8	9.7	12.41	<0.001
Mean number of members	1054	3080	2316	5505	949	1431	1098	3362	1128	3460	467	729	1.94	0.10

Characteristics related to the Affordable Care Act

As shown in Table 1, of the 346 direct service organizations who were included in this analysis, 30 (8.7 %) reported that they were already a Medicaid provider. The remainder said that they were willing to become a Medicaid provider without concerns (n = 52, 16.5 %), were willing but had concerns (n = 106, 33.5 %), were not willing (n = 87, 27.5 %), or did not know if they were willing (n = 71, 22.5 %).

Approximately half (51.9 %) of organization directors reported that they knew ‘‘nothing’’ or only ‘‘a little’’ about the ACA. This was reported most frequently (73 %) by the organizations who did not know if they were willing to become a Medicaid provider. Only 17 % of the sample overall reported that they knew ‘‘a lot’’, and of the ones who were willing without concerns, 25 % reported that they knew ‘‘a lot’’ about the ACA (Table 2).

Table 2.

Multinomial logistic regression of predictors of willingness to become a Medicaid provider (Base category is those who responded “Yes” N = 52)

Predictor variable	Not willing to be a Medicaid provider			Yes, willing to be a Medicaid provider but have concerns			Don’t know if willing
	(N = 87)			(N = 106)			(N = 71)
	OR	p	95 % CI	OR	p	95 % CI	OR	p	95 % CI
Knowledge of the ACA
Nothing/a little (reference)	–	–	–	–	–	–	–	–	–
Some things	1.12	0.78	0.49, 2.57	1.03	0.94	0.46, 2.31	0.48	0.11	0.19, 1.19
A lot	0.88	0.91	0.29, 2.57	0.92	0.86	0.36, 2.32	0.68	0.56	0.19, 2.45
Know of plans to expand Medicaid coverage	0.58	0.18	0.27, 1.28	0.74	0.40	0.36, 1.50	0.38	<0.01	0.18, 0.79
Discussed HCR within organization	1.05	0.93	0.39, 2.83	1.27	0.56	0.57, 2.81	0.89	0.78	0.38, 2.07
Discussed HCR within SCN	0.59	0.23	0.25, 1.41	0.92	0.82	0.46, 1.85	0.64	0.23	0.30, 1.34
Discussed HCR with a government agency or committee	0.49	0.18	0.18, 1.37	0.76	0.57	0.30, 1.94	0.43	0.054	0.19, 1.01
Peer specialists reimbursable in state	0.51	0.20	0.18, 1.41	0.54	0.20	0.21, 1.39	0.90	0.85	0.30, 2.71
State plans to expand medicaid	0.92	0.86	0.35, 2.42	0.72	0.45	0.30, 1.70	0.70	0.50	0.25, 1.97
Number of paid staff	0.95	<0.01	0.93, 0.99	0.99	0.18	0.98, 1.00	0.97	0.14	0.92, 1.01

About half (53 %) of the sample knew of plans to expand Medicaid, but almost 70 % of those who were willing to become a Medicaid provider without concerns reported that they knew of plans to expand Medicaid coverage to more people (χ² = 20.98, p < 0.001). Knowing of plans to expand Medicaid coverage was significantly associated with being in a state that is currently planning to expand Medicaid (70.9 %, χ² = 14.58, p < 0.001). However, whether an organization was in a state that planned to expand Medicaid (64 % overall) was not significantly associated with willingness to accept Medicaid reimbursement for the director’s own program.

Most directors (53 %) reported that people had discussed healthcare reform within the organization, but those who were willing to be a Medicaid provider but had concerns had these internal conversations more frequently (62 %) than the overall sample (χ² = 8.69, p < 0.05). Fewer than half of directors (43 %) reported discussions of health care reform within the Statewide Consumer Network, but 56.9 % of those who were willing to be a Medicaid provider with no concerns said they had discussed with the Statewide Consumer Network (χ² = 15.16, p < 0.05). Slightly more than a quarter (28 %) reported these discussions occurring directly between the organization and a government body, but 43 % of those who were willing to be a Medicaid provider with no concerns said they had discussed with the government (χ² = 16.84, p < 0.05).

According to SAMHSA, 31 states and D.C. included reimbursement of peer specialists in their state Medicaid plan (Hudock 2013). The organizations surveyed in this study were located in 48 states and the District of Columbia (no organizations from Rhode Island and Mississippi were included in the data because no peer-run organizations that met the study criteria could be identified). Close to 69 % of the organizations in the study are in states where peer specialists are reimbursed by Medicaid.

Organizational size was measured by the number of paid staff. On average, there were about 8 paid staff across organizations. The highest mean number of paid staff was in organizations that were willing without concerns to become Medicaid providers (M = 11.22, SD = 12.20). The mean number of members served annually was 1006.2 (SD = 2968.8) across all organizations, which did not differ by category of willingness to be a Medicaid provider.

Medicaid expansion and peer specialist reimbursement

Figure 1 reveals that many states (N = 17) currently reimburse for peer support and also were planning Medicaid expansion in 2013. Fewer states reimburse for peer-delivered services but do not plan for Medicaid expansion (N = 15), and even fewer states plan for Medicaid expansion but have no provisions for reimbursement of peer-delivered services are shown in light grey (N = 11). Finally, a minority of states (N = 7) had neither plans for Medicaid expansion nor do they reimburse for peer-delivered services (Fig. 2).

Fig. 2.

Value-based concerns about Medicaid reimbursement by organizations that were not willing, had concerns, or did not know if they would be willing to bill Medicaid

Analytic Results of Willingness to Become a Medicaid Provider

Knowledge of the ACA was not associated with willingness to become a Medicaid provider. Knowledge of plans to expand Medicaid was significantly associated with those who did not know if they were willing compared to those who were willing without concerns (OR = 0.38, p < 0.01). If directors knew about Medicaid expansion they were 60 % less likely to report that they ‘‘didn’t know’’ if they were willing to become a Medicaid provider. The only other significant predictor was the size of the organization. For every paid staff member, an organization was 5 % less likely to say they were ‘‘not willing’’ to become a Medicaid provider.

Concerns about Medicaid Reimbursement

Of the organizations that reported specific concerns about Medicaid reimbursement (those who answered to the willingness question: ‘‘Yes, but we have concerns,’’ ‘‘No’’, or ‘‘Don’t know’’), 79 % reported ‘‘conforming to the medical model’’ to be a concern. Other common value-based concerns included: Medicaid reimbursement detracting from the mission (76 %), medical necessity criteria (71 %), and Medicaid reimbursement interfering with their commitment to advocacy (67 %). Close to 80 % of the respondents (n = 152; 78 %) reported all three of these concerns as problems (Fig. 3).

Fig. 3.

Practical concerns about Medicaid reimbursement by organizations that were not willing, had concerns, or did not know if they would be willing to bill Medicaid

The most frequently reported practical concern was not having financial staff to manage billing (72 %). Other practical concerns were related to keeping records for claims (60 %) and sophistication of computer systems (57 %). The application process (48 %), auditing by an insurance company (42 %), and quality and performance measurement (37 %) were the least frequently reported concerns, although 42 % of organizations said that they ‘‘didn’t know’’ if quality and performance measurement would be a problem—indicating a potential lack of understanding of these requirements or their ramifications.

Discussion

At a time when more behavioral healthcare providers and state agencies see Medicaid reimbursement on the horizon as a sustainable source of revenue, this study demonstrated that mental health peer-run organizations had concerns about accepting Medicaid reimbursement, or were not currently willing to become Medicaid providers. Although we hypothesized that understanding the ACA would positively impact willingness of peer-run organizations to participate as a Medicaid provider, the results here indicate that it is not. We do not yet know whether these attitudes will predict how many peer-run organizations decide to accept Medicaid reimbursement in the future.

Educational endeavors about the ACA seem unlikely to address the organizations’ unwillingness to bill Medicaid and their ideological and practical concerns. The current organizational structures were not created to manage the practical requirements that might accompany Medicaid. However, deciding to bill Medicaid may be more dependent on norms for maintaining the values of peer support as being mutual and non-professional, organizations’ sense of control over the environment, and developing skills around billing Medicaid. There will also be a need to develop reimbursement documentation procedures that don’t conflict with their values.

Although currently many organizations rely on governmental sources of funding, many of these streams (such as the block grant) do not have billing or financial reporting that would require knowledge of the ACA or Medicaid. Peer-run organizations may not have had the motivation to learn about the ACA or be in information networks that could provide this knowledge. Knowledge of the ACA is important at this time because it could help inform organizational planning for sustainability within a shifting environment. The directors of most of the organizations in this study (83 %) did not feel that they had a lot of knowledge about the ACA, although about half (53 %) were aware of plans to expand Medicaid.

There was no relationship between peer-run organizations’ willingness to bill Medicaid and state policies related to reimbursement of peer specialists or state plans to expand Medicaid. The peer specialist reimbursement policy of the state in which a peer-run organization is located did not have an effect on willingness of these organizations to bill Medicaid. Similarly, the policy of the state regarding Medicaid expansion had no effect on their willingness to bill Medicaid. We may see changes in attitudes related to the intersection of peer specialist reimbursement and Medicaid coverage expansion in the future as these policies are implemented.

The size of the organization was associated with willingness to become a Medicaid provider, and having staff to manage billing was a frequent concern. Those who have more paid staff were more likely to say they were willing to accept Medicaid reimbursement than those with fewer staff. The association between larger organizational size and willingness to bill Medicaid indicates the importance of capacity, at least in terms of influencing perceptions of the directors. Larger organizational size has been shown to influence the ability of non-profits to innovate (Jaskyte 2011; Jaskyte and Dressler 2005). Adapting to new financing mechanisms such as Medicaid may be an important innovation for these organizations in the near future. It will be important to assure that appropriate procedures used for dissemination of knowledge and training are in place when more peer-run or other organizations are to be integrated into the system. Directors may be knowledgeable about the specifics of Medicaid reimbursement—as demonstrated in the concerns they identified in this study—without having enough information to know whether the organization could adapt or how they can adapt.

Concerns About Medicaid Reimbursement of Peer Support

This study documented both ideological and practical concerns of peer-run organizations about billing Medicaid. The most frequent concerns were in the value-based categories. All of the ideologically-based concerns were reported by at least 70 % of the participants except for the ‘‘commitment to advocacy.’’ It is important to understand that the value-based concerns reflect compromises these organizations would have to make in their missions and the way in which they engage service users and their communities, rather than merely day-to-day operations. Changing the mission of mission-driven nonprofits could compromise their sustainability equally, if not more so, than funding instability as it could change the very nature of their supports and activities. The frequency of reporting value-based concerns as the most significant barriers may be observed because the organizations see these potential conflicts more clearly at this time, given that more concrete, practical concerns are further in the distance or not as well understood. For the practical concerns, the only one that exceeded more than 70 % of respondents reporting it as a concern was the financial staff to manage billing. This is consistent with the finding that organizations with more staff were more willing to become a Medicaid provider. The perception that having staffing capacity would make Medicaid reimbursement more palatable is clear in both the analysis of willingness and the analysis of specific concerns. Going forward, we may see that as organizations explore Medicaid reimbursement, practical concerns other than staffing capacity become more salient. However, the first barrier may be overcoming significant concerns about the values on which these organizations were founded.

Limitations

Although paid staff is a relatively standard measure of organizational size in more traditional agencies, it may have limited practical use in this context. We cannot expect that organizations will scale-up their staffing in preparation for Medicaid reimbursement. Here we have interpreted it as a measure of perceived organizational capacity to innovate and change revenue streams, but participating in Medicaid reimbursement would also likely provide the opportunity to hire more paid staff. Overcoming other perceived barriers—in particular the value-based concerns reported by the organizations—may be a more important first step in determining organizational readiness. Organizational leaders and policymakers should attend to these measures of perceived barriers, even in the presence of a large, well-staffed organization.

A related limitation is that we did not measure the particulars of the funding environment or Medicaid policy in the states and localities where these organizations operate. The availability of Medicaid reimbursement for peer services alone does not necessarily indicate an funding or policy environment where other, more customary funding streams are shrinking, nor did we model the exact reimbursement mechanism for peer services such as the requirements for particular waivers or a state where beneficiaries receive behavioral health services through a fee-for-service rather than managed care system. Additionally, we did not measure the perceived adequacy of reimbursement rates for peer specialists, which may have an impact on willingness to participate in Medicaid. These specifics of Medicaid policy in the states should be studied further in the future. However, we believe the findings reported here are largely generalizable to organizations facing a dramatically changing landscape of funding requirements, and can be interpreted in the context of having to shift to more medically-oriented reimbursement mechanisms as Medicaid is increasingly used to sustain peer support services.

Conclusion

Peer-run organizations, while still small in number, provide essential services to communities (Chinman et al. 2014; Goldstrom et al. 2004; Ostrow and Hayes 2015). As demonstrated in earlier research documenting the supports and activities of peer-run organizations, these programs are providing valuable resources to people with psychiatric disabilities to support social networks and community-building (Ostrow and Hayes 2015). In addition, the Pillars of Peer Support conference attendees—who have historically set many of the national policy priorities for peer support—believe that the peer workforce is strengthened when there are consumer-run organizations that serve to involve and support consumers in service development and delivery (Daniels et al. 2010).

Policies such as the passage of the Mental Health Parity and Addiction Equity Act in 2008 and the Affordable Care Act in Congress in 2010 have provided many opportunities for financing of innovations in our mental health system (Glied and Frank 2008; Mechanic 2012). There are potential advantages of Medicaid reimbursement of peer support. It could make more resources available to peer-run organizations and allow for diversifying the organizations’ funding streams, which could potentially expand operating budgets. Peer-run organizations that pursue Medicaid funding may have increased organizational revenue, sustainability, and ability to reach more people in distress. Some Medicaid managed care companies are interested in creating partnerships with peer-run organizations to enlist peer supporters to provide recovery-oriented supports to insurance plan members (Daniels et al. 2013). If peer-run organizations join managed care company networks, the peer-run organizations can provide beneficiaries with a more diverse array of services, and also promote the ideas of recovery and resilience within the network. Including peer-run organizations could mainstream both their services and values, which can influence the culture of the whole system (Daniels et al. 2013). While peer-run organizations reject conceptualizations of life problems as medical illnesses (Mead et al. 2001), this may hold the organizations back from joining managed care networks. However, participating in these networks could broaden opportunities for recovery-oriented services within the managed care network and other network providers (Davidson and Guy 2012). Peer-run organizations can be targeted by managed care to capitalize on their current activities and funding streams, help develop the organizations, and broaden community support programs for people with mental disorders.

The concerns reported about receiving Medicaid funds reflect the organizations’ perceived barriers to Medicaid reimbursement. Reimbursing peer support through Medicaid risks compromising some of the core values of peer support, including mutuality, equality, non-coercion, and the lack of a power imbalance (Mead et al. 2001). Promoting empowerment and self-direction through non-hierarchical relationships has been the guiding vision of the consumer/survivor movement (Morrison 2013). Our findings on specific concerns indicate that the population of peer-run organizations is considering values of self-direction and non-hierarchical relationships in relation to Medicaid reimbursement. They also had more practical concerns—such as not having financial staff to manage billing or keeping records for claims.

Financing for innovative mental health services is a challenge for many states and providers because of standardization and billing procedures (Magnabosco 2006). Medical necessity criteria involve documenting diagnoses and functional deficits in clients. These criteria may be more difficult for peer-run organizations to document than other providers because they are not only unqualified to document diagnoses, but generally want to focus on non-medical approaches to members and the members’ individual strengths (Holter et al. 2004). The lack of adaptation of billing procedures and capacity for peer-run organizations to meet these requirements was also of concern.

Implications

Addressing the very real concerns of peer-run organizations by providing more flexible billing mechanisms that use global payments or self-directed care funds to receive Medicaid reimbursement may be a potential solution. Global payments provide a per-person payment to a provider (or group of providers) to provide services, rather than relying on fee-for-service billing. Self-directed care gives individual service users direct control over service delivery dollars so that they can purchase goods, services, and supports—including peer support—and use the services of a financial broker, which would alleviate burden from smaller peer-run organizations of needing financial staff (Cook et al. 2008). Emerging Medicaid financing options, such as self-directed care, address many of the peer-run organizations’ concerns, since the organizations are concerned about documenting medical necessity and having internal staffing capacity (Alakeson 2008). These mechanisms are becoming more widely researched and developed (Alakeson 2010) and are currently involving peer-run organizations in their implementation research (Cook et al. 2010). However, self-directed care model has not been widely implemented for people with psychiatric disabilities. More widely implementing financing mechanism like this one could be important to maintaining the sustainability of peer-run organizations, while preserving the values and unique supports that these organizations offer their communities and members.

This study addressed some potential challenges, opportunities, and data to support concerns of peer-run organizations nationwide about Medicaid reimbursement. If we anticipate and respond to their concerns with concrete and acceptable compromises and creative financing mechanisms, we could ensure the sustainability of both the peer-run model and the individual organizations.