Estimating the Impact of Medicaid Expansion and Federal Funding Cuts on FQHC Staffing and Patient Capacity

Abstract

Policy Points.

• In the preexpansion period, federally qualified health centers (FQHCs) in Medicaid expansion states were significantly different from those in nonexpansion states. This gap widened as revenues in expansion states continued to grow at a faster rate after the expansion.

• If Medicaid expansion had occurred nationwide, FQHCs’ revenue and capacity could have increased substantially. Over time, Medicaid could play a bigger role as it becomes a more stable funding source to allow for capital investments.

• Section 330 grants appear to have a larger impact on access to care. Given the varying levels of reliance on Medicaid, investing through federal grants might be more effective and equitable.

Context

The Health Resources and Services Administration's Health Center Program (HCP) plays a critical role as the national ambulatory safety net, delivering services to patients in medically underserved areas, regardless of their ability to pay. As the program has grown, health policy initiatives may have altered access to care for the underserved population. Understanding how federally qualified health centers (FQHCs) have been affected by past policies is important for anticipating the effects of future policies.

Methods

By analyzing a national data set from the Uniform Data System, we examined, using two sets of random effects regressions, the potential impact of alternative policy actions affecting FQHCs. Our primary equation models the number of full‐time equivalent staff, of patients served, and of visits provided in the subsequent year as a function of Medicaid revenues, Section 330 grants, and other revenues. Our secondary equation is a difference‐in‐differences analysis that models Medicaid revenues as a function of the states’ status of Medicaid expansion.

Findings

The expansion of Medicaid in nonexpansion states could have increased Medicaid revenues by 138%, staffing by 25%, and patients’ visits by 24% in 2017. Compared to the impact of a “repeal” of Medicaid expansion, the percentage of reductions in staffing would be similar to those predicted by a 50% cut in Medicaid revenues or in Section 330 grants. On a dollar‐for‐dollar basis, the effects of one dollar of Section 330 grants were more than double that of one dollar of Medicaid revenue.

Conclusions

Both Medicaid eligibility and Section 330 funding support are important to the HCP, and Section 330 grants are particularly closely related to staffing and the provision of services. States’ decisions not to participate in or to repeal Medicaid expansion, to reduce Medicaid payment rates, and federal funding cuts all could have a negative impact on FQHCs, resulting in thousands of low‐income patients losing access to primary care.

The health center program (hcp), funded by the health Resources and Services Administration's (HRSA) Bureau of Primary Health Care (BPHC), serves as a cornerstone of the ambulatory care safety net in the United States. In 2001, bipartisan health policy initiatives doubled federal funding for the program to $2.1 billion through the Health Center Growth Initiative, with the goal of supporting 1,200 new or expanded health centers sites (we use FQHC in this article to refer to federally qualified Health Center Program awardees, not including FQHC Look‐Alikes). ¹ In 2009, the American Reconstruction and Reinvestment Act added a one‐time appropriation of $2 billion. Then the Affordable Care Act (ACA) of 2010 provided $11 billion over five years through the newly created Community Health Center Fund (CHCF). In addition, efforts to expand access to health insurance via Medicaid (in those states that chose to do so) and private insurance (for those with incomes above Medicaid eligibility limits) through the ACA enhanced the FQHCs’ financial stability. ² Understanding how FQHCs have been affected by past policies is important for anticipating the effects of future policies.

FQHCs’ revenues and operations are tied to three key payment policies: (1) grants provided under Section 330 of the Public Health Service Act (including funds from the CHCF), (2) changes in Medicaid enrollment through Medicaid expansion, and (3) the Medicaid Prospective Payment System (PPS) statute. The recent history of each of these policies reflects some degree of instability.

Section 330 grants are awarded through a competitive application process, after which the BPHC continues to invest in primary care in the awardee's community. To receive funding under Section 330 grants, FQHCs are required to be governed by a community board composed of a majority of FQHC patients, and they must provide comprehensive primary health care services and supportive services that are available to all, along with a sliding fee scale (fees based on ability to pay). ¹ , ³ , ⁴ , ⁵ These grants allow FQHCs to expand capacity, upgrade infrastructure, and offset operating losses incurred while delivering care to patients who cannot pay. In 2017, Section 330 funding expired during congressional debates over the federal budget ⁶ but was later restored with the Bipartisan Budget Act of 2018. ⁷

The ACA's Medicaid expansion has been a large focus of policy discourse since 2010. A ruling by the Supreme Court in 2012 allowed states the choice of expanding their Medicaid program, resulting in 39 states (including DC) that have done so as of August 2021. The expansion of Medicaid has been shown to reduce disparities in coverage for FQHC patients in expansion states, especially among marginalized populations. ⁸ The Trump administration actively pursued legislative efforts to repeal the ACA, along with multiple legal challenges, including a Supreme Court hearing in June 2021 that eventually left the ACA in place. ⁹ In addition, several states have recently expanded Medicaid, with more contemplating doing so. ¹⁰ , ¹¹ The impact of Medicaid expansion for marginalized populations can thus inform policy discourse in these jurisdictions.

The Medicaid PPS for FQHCs was created as part of the Medicare, Medicaid, and SCHIP Benefits Improvement and Protection Act of 2000. This required FQHCs to be reimbursed at a minimum rate for services provided to Medicaid patients. The PPS sets a per‐visit payment rate for each FQHC based on historical costs of the FQHC, allowing for annual adjustments. States have requested more flexibility in establishing payments to FQHCs and have pointed to PPS rate limits as a barrier to incorporating FQHCs into value‐based payment initiatives. ¹² Overall, both Medicaid reimbursement rates and PPS rate limits have been identified as potential access barriers for FQHC patients in both expansion and nonexpansion states. Accordingly, simulations that examine the potential impact of varying these variables are helpful to state and federal policymakers.

The possible changes to each of these payment policies have important implications for FQHC revenues and their capacity to serve low‐income patients. Previous studies have shown that compared to FQHCs in states that did not expand Medicaid, those in states that did are more financially stable and better able to provide expanded services and affordable care to their patients, whereas FQHCs in nonexpansion states appear to rely more heavily on federal grants than on Medicaid reimbursements. ¹³ , ¹⁴ , ¹⁵ These studies, however, were either based on a survey of FQHCs, or they evaluated the effects of one or two of these policies in isolation.

To help policymakers anticipate the impact of potential changes in key policies on the overall HCP, individual grantees, and the patients they serve, we developed an estimating model of FQHC revenues and operations based on past observations. Our study is the first to investigate the comparative effects of several policy actions over time for multiple outcomes of clinical volume, such as staffing, numbers of visits, and patients seen. In light of President Joe Biden's ambitious plan to build on the ACA and double investment in FQHCs, ¹⁶ our study also provides insights into how best to fund these centers to expand access to care.

Methods

Data Sources and Study Populations

Our analysis sample starts with FQHCs that receive Section 330 federal award funds as our primary unit of analysis. Longitudinal FQHC data come from the Uniform Data System (UDS), 2010‐2017. Patient‐level characteristics include percentages of FQHC patients in each age group, sex/gender, race/ethnicity, income level, and proportions of patients with chronic diseases (diabetes, heart disease, and hypertension). FQHC‐level variables include annual revenues from Medicaid, Medicare, Section 330 grants, and other sources, participation in the 340B Drug Pricing Program (Health Resources and Services Administration. Office of Pharmacy Affairs 340B OPAIS), number of patients served, and number of visits provided.

We supplemented our UDS data with Rural‐Urban Commuting Area (RUCA) codes, county characteristics from CountyHealthRankings.org (CHR), and Small Area Income and Poverty Estimates (SAIPE). County‐level variables include unemployed rate, percentage of uninsured adults, percentage of children in poverty, rate of primary care providers (General Family Medicine, General Practice, General Internal Medicine and General Pediatrics, as in AMA Masterfile) per 100,000 population (from CHR), and median household income (from SAIPE). We further augmented our models with data regarding a center's state participation in Medicaid expansion and alternative payment models (APMs), with amounts of alternative payment reported in the UDS.

Starting with 1,440 unique FQHCs from 2010 to 2017, we excluded 32 FQHCs in US territories because Medicaid operates differently in the territories than in the states. We dropped 398 FQHCs that were newly established after 2011 or did not have continuously reported data throughout the years, and 18 FQHCs with missing Medicaid revenue, alternative payment amounts, urban/rural status, and/or county‐level characteristics in any year. Our final sample contained 992 FQHCs with complete data for each year.

Outcomes

To assess the effects of policy changes on FQHC revenues and capacity, our main outcome measures were Medicaid revenues, total number of full‐time equivalent (FTE) staff, number of billable FTEs (physicians, physician assistants, nurse practitioners, clinical psychologists, certified nurse midwives, and clinical social workers, as required by federal law), number of patients served, and number of visits provided. Our estimating models produce predicted outcomes in the subsequent year based on explanatory variables in a current year, as we assumed a one‐year lag structure for changes in key variables to take effect.

Model Specifications

We estimated two sets of regression models. Our primary equation (online Appendix, Stage 2 in Table A2) models each of our capacity outcomes as a function of revenues from Medicaid, Section 330 grants, other revenues, patient characteristics, and county‐level characteristics. Our secondary equation (Stage 1 in Table A2) models Medicaid revenue as a function of a state's Medicaid expansion status and the same patient and county‐level characteristics.

For policy simulations that examine the impact of Medicaid expansion, we first fitted the secondary equation models. Our first step was a difference‐in‐differences (DID) analysis to compare changes in Medicaid revenue after Medicaid expansion for FQHCs in expansion and nonexpansion states, a commonly used design in the context of Medicaid expansion. ¹⁴ , ¹⁵ , ¹⁶ , ¹⁷ By including the binary expansion variable, the indicators for the number of years after expansion, and the interaction of these indicators with the expansion variable, this DID approach addresses secular trends and baseline differences between the two groups while taking into account any differential effects of the expansion in later years. We further adjusted for differences between FQHCs in expansion versus nonexpansion states by including FQHC random effects, rural/urban status, 340B designation, patient characteristics, as well as county‐level socioeconomic factors and access to primary care providers. We used random effects at the FQHC level because we were interested in the effects of time‐invariant variables, which cannot be estimated in fixed effects models. Moreover, a Hausman test failed to reject the preference for the random effects model.

The second step in Medicaid expansion simulations is estimating a panel regression that uses the estimated Medicaid revenues from the first step, together with all other revenue variables and the same set of covariates to predict our three outcome measures. In simulations examining the effects of cuts in per‐visit Medicaid rates and Section 330 grants that do not involve the first step, we used observed Medicaid revenues as the input to our model. In addition to FQHC random effects, we included year fixed effects to account for secular trends not attributable to any other explanatory variables. Furthermore, for both steps, we used bootstrapped (where entire panels were resampled) robust standard errors clustered at the FQHC level to address within‐panel serial correlation and additional variations related to using estimated Medicaid revenues.

For ease of interpretation, we used nontransformed outcome measures in our main models, and presented the results from the same models but with log‐transformed outcomes as sensitivity analyses in the online Appendix.

Policy Scenarios

We evaluated several policy scenarios in which we modified (1) a state's decision to expand Medicaid, (2) the Medicaid PPS statute, and (3) Section 330 grants. An FQHC is considered to be in an expansion state if the state participated in ACA's Medicaid expansion by the end of 2014. To ensure the robustness of our findings, we examined specifications that excluded centers in states that expanded Medicaid eligibility before 2014 (“early expansion”) from the expansion state group as a sensitivity analysis. We also examined specifications that excluded centers in states that participated in the expansion with an effective date between 2015 and 2017 (“late expansion”). The results of these analyses are in the online Appendix.

Health Center Policy Simulation 1: Nationwide Medicaid Expansion

After fitting the first‐stage DID model, to simulate the scenario in which nonexpansion states had instead expanded, we set the expansion variable to 1 for all FQHCs and estimated Medicaid revenues again using the same regression coefficients. This resulted in an estimate of additional Medicaid revenues that would have accrued to FQHCs had they been located in expansion states, controlling for differences in other characteristics. Using these estimated Medicaid revenues, the second‐stage random effects models predicted FTEs, visits provided, and patients served in the subsequent year.

Health Center Policy Simulation 2: “Repeal” of Medicaid Expansion

Our second policy simulation employed a similar method as that used in the first simulation, except that we set the expansion variable to 0 for all FQHCs to simulate the scenario that no state had expanded Medicaid eligibility by the end of 2014. This led to an estimate of reduced Medicaid revenues for FQHCs in expansion states, as well as resulting changes in FTEs, visits provided, and patients served in the subsequent year.

Health Center Policy Simulation 3: Cuts in Medicaid Revenues

Our third policy scenario assumes that the per‐visit Medicaid rates were cut by 25%, 50%, or 75%. As a result, we assumed that the Medicaid revenues would be effectively reduced by the same percentages. Because the actual PPS rates were not reported, we simulated the cuts by substituting observed Medicaid revenues with 25%, 50%, or 75% of the observed revenues in the aforementioned random effects panel regression to predict subsequent‐year FTEs, visits, and patients.

Health Center Policy Simulation 4: Cuts in Section 330 Grants

In the fourth policy scenario, keeping everything else the same, we reduced the observed Section 330 grants by 25%, 50%, 75%, or 100% in order to estimate the corresponding effects on subsequent‐year FQHC staffing and clinical volume.

Limitations

Our findings should be interpreted in light of several study limitations. First, for better comparability across time and to fit our DID analytical approach, we simplified the categorization of expansion versus nonexpansion states as binary and time‐invariant. Although we examined alternative specifications that excluded early‐ and late‐expansion states, our DID analysis offers only an estimated effect of Medicaid expansion averaged across all expansion states compared to all nonexpansion states. Our simulation results reveal what could have happened if this averaged effect had also taken place in nonexpansion states, or had never taken place, assuming parallel trends between FQHCs in expansion and nonexpansion states in the absence of the expansion. We tested the validity of this assumption by visually inspecting the preexpansion trends in outcomes for FQHCs in expansion versus nonexpansion states (see Descriptive Analysis in the online Appendix).

Second, to avoid compositional changes, we excluded FQHCs that were not continuously reported during the study period, which limited our ability to study the patterns of newly established or closed FQHCs. We found that revenues and capacity were slightly smaller when these FQHCs were included, and the effect size of Medicaid expansion also seemed to be smaller.

Third, because each FQHC has a patient‐majority governing board that makes decisions about services provided, it is likely that the full impact of funding or policy changes may not be realized in one year, especially in regard to increasing the scope of services available to patients. But when we assumed a one‐year lag, we were able to detect meaningful changes in FQHC staffing and capacity in the short term, which is likely to be more accurate than estimating with a longer lag.

Fourth, for policy simulation 3, we assumed that the per‐visit payment rates were cut while patient volume remained the same, so that a 25% reduction in Medicaid revenues would translate into a hypothesized 25% cut in per‐visit rates. We based this assumption on the fact that FQHCs are mandated to treat patients, regardless of their ability to pay.

Fifth, we based our estimating model on UDS data starting from 2010. In future analyses, we might consider including earlier data that preceded the ACA. We also acknowledge the limitations of these self‐reported data, whose accuracy depends on each FQHC's skill in collecting and reporting the required information.

Finally, although our analysis took into account all sources of FQHC revenues and grants, it did not consider the combined effects of multiple policies that may simultaneously affect FQHCs. As an initial exercise, we believed that a formal assessment of individual policies while holding others constant was more likely to produce valid predictions than would assessing combined policy changes. Accordingly, one area of future inquiry would be to consider combined policy changes that might have countervailing effects.

Results

Health Center Characteristics and Trends in Revenues

In our sample, 55% of the 992 FQHCs were in expansion states, and 70% were located in urban areas. On average, in the counties with FQHCs, 17% of adults were uninsured, 6% were unemployed, 21% of children were in poverty, 95 primary care providers were available per 100,000 population, and median household income was $46,952 at baseline (2010). The mean number of staff FTEs increased from 125 in 2010 to 196 in 2017. The mean number of visits that FQHCs provided rose from 72,745 to 97,844, and patient volume increased from 18,329 to 24,054. In 2010, only 18% of the FQHCs received payment under any Medicaid APM, but participation went up to 37% in 2017, with 28% of the centers receiving $10 or more per Medicaid patient per year. To be able to purchase 340B‐discounted drugs, 91% of FQHCs were enrolled with the drug‐pricing program in 2010, which increased to 97% in 2017. In all centers, Medicaid revenue went up from an average of $5.15 million to $10.10 million, and Section 330 grants grew from $2.07 million to $3.96 million (online Appendix, Table A1), both effects indicating a near doubling of support.

We observed that in the preexpansion period (2010‐2013), FQHCs in Medicaid expansion states were already receiving larger shares from Medicaid and relied less on Section 330 funding at baseline (Figure 1), with higher per‐patient revenues (Figure 2). The gap in total revenues widened as revenues of FQHCs in expansion states continued to grow at a faster rate after the expansion, whereas revenues in nonexpansion states went up only marginally.

Figure 1.

Payer Mix of FQHC Revenues by State Medicaid Expansion Status, 2010‐2017^a

    

^a State Medicaid expansion status is defined as a time‐invariant binary variable that captures whether an FQHC is located in a state that had expanded Medicaid eligibility by the end of 2014. There were 27 expansion states (MN, CT, DC, CA, WA, NJ, AZ, AR, CO, DE, HI, IL, IA, KY, MD, MA, NM, ND, NV, NY, OH, OR, RI, VT, WV, MI, NH) and 24 nonexpansion states (PA, IN, AK, MT, LA, ID, ME, NE, UT, VA, AL, FL, GA, KS, MS, MO, NC, OK, SC, SD, TN, TX, WI, WY). Sources of “other patient related revenue” include other public payers, private payers, and self‐pay. “Section 330 grants” include four types of funds under the Health Center Program authorized by Section 330 of the Public Health Service Act: for migrant health centers, for community health centers, for health care for the homeless, and for public housing primary care. “Other grants” include other BPHC grants, other federal grants, nonfederal grants, and contracts. “Other revenue” includes non‐patient‐related revenue not reported elsewhere. All revenues have been adjusted for inflation.

Figure 2.

Per‐Patient Dollar Amounts of FQHC Revenues by State Medicaid Expansion Status, 2010‐2017

In the postexpansion period (2014‐2017), Medicaid accounted for more than 40% of the total revenue among FQHCs in expansion states, but its share in nonexpansion states maintained a preexpansion level of around 25%. In contrast, Section 330 funding represented a large and growing share of revenue for nonexpansion state FQHCs throughout the period (26% to 34%) compared to expansion state FQHCs (17% to 22%).

Health Center Policy Simulation 1: Nationwide Medicaid Expansion

In our main analysis, the expansion states group includes 27 states (including DC) that expanded Medicaid eligibility by the end of 2014. The other 24 states are considered nonexpansion states.

Within our first‐stage difference‐in‐differences model, we found that FQHCs in Medicaid expansion states experienced markedly greater increases in Medicaid revenue between preexpansion (2010‐2013) and postexpansion (2014‐2017) than did otherwise comparable FQHCs in nonexpansion states, with differences of $2.15 million in 2014, $3.81 million in 2015, $3.97 million in 2016, and $4.40 million in 2017 (online Appendix, Table A2). Under the simulated Medicaid expansion, the predicted incremental growth in Medicaid revenue is almost exponential (Simulation 1 in Table A2).

Our simulation suggests that when adjusting for covariates, if all the nonexpansion states actually participated in the Medicaid expansion by the end of 2014, we would expect each FQHC in those states to have $6.40 million more (137% higher) Medicaid revenue in 2016, which would allow a center to add 36 (25%) total FTEs or 16 (8%) billable FTEs, provide 16,837 (24%) more visits, and serve 2,258 (11%) more patients in 2017.

Our second‐stage model estimates suggest that when controlling for covariates, $1 million in Medicaid revenue would translate into 5.6 total FTEs or 0.8 billable FTEs, 2,633 visits, and 353 patients in the subsequent year. Revenues from Medicare, other public and private payers, and self‐pay all had greater predictive power for clinical volume than did Medicaid revenue (Table A2).

Using our data from 2010 through 2017, we estimated that in 2018, if FQHCs in nonexpansion states had located in expansion states, the annual increments in capacity could have accumulated 220 total FTEs or 29 billable FTEs, 117,066 visits, and 26,207 patients, all of which are around twice as many as those FQHCs initially had in 2010 (Simulation 1 in Table A2).

Health Center Policy Simulation 2: “Repeal” of Medicaid Expansion

If Medicaid had not expanded in the expansion states, the predicted $6.40 million change associated with the expansion would represent a 48% reduction in Medicaid revenue in 2016 (as opposed to a 137% increase in Simulation 1). This difference in percentages can be attributed to the preexisting difference in Medicaid revenues between FQHCs in expansion states versus those in nonexpansion states. This predicted revenue loss among FQHCs in expansion states would in turn have led to a reduction of 36 FTEs (now 15%) in staffing, a 16,837 (now 14%) drop in visits provided, and a 2,258 (now 8%) drop in patients served in 2017 (Figure 3). If we counted all FQHCs in the nation, this counterfactual “repeal” would have resulted in 37% less Medicaid revenue in 2016, leading to 10% fewer FTEs, 10% fewer visits, and 5% fewer patients in 2017 (Simulation 2 in Table A2).

Figure 3.

Estimated Impact of Four Policy Simulations on FQHC Medicaid Revenue, Full‐Time Equivalents, Visits Provided, and Patients Served, Featuring Year 2017^a

    

^a Simulation 1 assumes that the effect of the 2014 Medicaid expansion was applied to all states. Simulation 2 assumes that the 2014 Medicaid expansion did not take place in any state. Simulation 3 assumes that Medicaid revenues were cut by 50% for all FQHCs. Simulation 4 assumes that Section 330 grants were cut by 50% for all FQHCs.

These estimated aggregate effects of a “repeal,” averaged across all FQHCs in the nation, are more pronounced than are the effects under the scenario of a nationwide Medicaid expansion (30%, 8%, 8%, and 4% in Simulation 1 in Table A2), which suggests that from a national point of view, a repeal of Medicaid expansion may have more detrimental consequences when the observed gains in FQHC capacity are reversed.

Health Center Policy Simulation 3: Cuts in Medicaid Revenues

Our panel regression analysis suggests that $1 million in Medicaid revenue is associated with 6 more FTEs, 3,666 more visits, and 599 more patients in the subsequent year (Table A2). Our simulation results show that a 50% reduction in Medicaid revenue would lead to an average nationwide drop of 15% in FTEs, 18% in visits, and 12% in patients in 2017 (Simulation 3 in Table A2). The anticipated effects would be larger among FQHCs in expansion states: 18%, 21%, and 14% (Figure 3). If Medicaid revenues were reduced by 25% instead of 50%, we would expect the drops in staffing and capacity outcomes to be half as large.

Health Center Policy Simulation 4: Cuts in Section 330 Grants

Based on our model, $1 million in Section 330 grants translate into 17 more staff FTEs, 7,406 more visits provided, and 2,152 more patients seen (Table A2). These estimated effects are more than double the effects associated with the same increment in Medicaid revenue. It is worth noting that the average dollar amount of Medicaid revenue for each FQHC in expansion states is more than twice as much as that from the Section 330 grants. Therefore, it is not surprising that a percentage funding cut in either Medicaid revenues or Section 330 grants would lead to comparable changes in expansion states (Figure 3), despite the disproportionate impact on a dollar‐for‐dollar basis from these two funding sources.

Our analysis suggests that for FQHCs in expansion states, similar to the effect of a 50% cut in Medicaid revenues (representing an average of $5.07 million per FQHC), a 50% cut in Section 330 grants (representing an average of $1.48 million per FQHC) would lead to a drop of 14% in FTEs, 13% in visits, and 16% in patients in 2017. Moreover, although a 50% cut in Section 330 grants would have the same impact on all FQHCs across the country, the anticipated negative effects are also comparable to that of a Medicaid expansion “repeal.” As mentioned in Simulation 2, in the scenario in which states reversed the decision on expanding Medicaid in expansion states, the associated 48% drop in Medicaid revenues for FQHCs in expansion states would be very close to 50%. For FQHCs in nonexpansion states, the effects on clinical volume associated with a cut in Section 330 grants would be more pronounced than that associated with a cut in Medicaid revenues because of their much lower Medicaid revenues and greater reliance on Section 330 grants (Figure 3). Notably, if Section 330 grants were suspended, staffing would drop by 33%, visits would drop by 29%, and patients would drop by 34% in 2017 (Simulation 4 in Table A2).

Discussion

We constructed an estimating model of FQHC revenue and clinical volume based on observations over the past decade, a period of strong growth in FQHCs. An important observation from our findings is that in the preexpansion period, FQHCs in Medicaid expansion states were already significantly different from FQHCs in nonexpansion states. In 2010, using patient counts as a proxy for FQHC size, there was an average of 21,178 patients in FQHCs in expansion states, 43% more than the average in nonexpansion states (14,783). At the same time, FQHCs in expansion states also had more overall revenues per patient and a larger proportion of Medicaid revenues. Building on baseline differences, Medicaid expansion in 2014 was associated with incremental increases in Medicaid revenues. Our estimating model captures both these baseline differences and incremental changes in Medicaid revenues in the years following the expansion. If Medicaid had expanded throughout the country, the FQHCs’ revenue and clinical volume could have risen significantly. Our findings thus underscore the beneficial impact of Medicaid expansion on expanding treatment capacity. Indeed, policymakers’ awareness of such patterns may have been a factor undergirding the vocal opposition of both Democratic and Republican governors to the repeal of Medicaid expansion in 2017.

The dynamics of growing more capacity to serve new patients in an FQHC is complicated, and there are several possible explanations for the differences in the predicted percentage increases in Medicaid revenue, staffing, visits, and patients. First, it is important to note that Medicaid revenue accounted for only 26% of total revenue in nonexpansion state FQHCs. Changes in a single revenue source among many would not have as large an impact as one would expect if it were the dominant source. Despite the relatively modest share of baseline revenue from Medicaid, our analysis showed that Medicaid expansion would lead to increases in staffing by 25% and patients by 11%, representing a fairly large elasticity.

Another possible reason to see somewhat modest effects on patient numbers is that, with Medicaid expansion, existing patients who seek care at FQHCs may have shifted from being uninsured to insured. We also do not know the relative clinical complexity of the newly added patients with Medicaid who may have required more care for unaddressed chronic conditions. In order to shed light on the intensity as well as the variety of care, we examined both the number of patients served and the number of visits provided as key outcomes. In Simulations 1, 2, and 3, the estimated impact of those policy changes was larger on visits than on patients, indicating that Medicaid revenues may allow FQHCs to take care of existing patients more often. In contrast, Simulation 4 reveals that cuts in Section 330 grants would reduce the number of patients more than the number of visits.

It is also important to view clinical volume from the perspective of FQHC leadership. Each FQHC may make different strategic decisions regarding how to engage the newly gained resources to serve patients who have access to FQHCs after the expansion of Medicaid. Some of the impact would concern existing patients, which would stabilize the FQHCs financially. In addition, it also takes time (i.e., years) for FQHCs to bring in new personnel and patients.

Finally, the political uncertainty of the permanence of Medicaid expansion and other organizational factors may have influenced the FQHCs’ decisions. Our data may not have allowed us to observe the full effect of Medicaid expansion if stakeholders had reacted to it as a permanent change in the policy landscape rather than as an immediate reaction to a transient change. Over time, Medicaid could have a larger impact as it becomes a more stable source of funding, thereby allowing FQHCs to make capital investments and expand their services.

Among the reduction policy scenarios we explored, the repeal of Medicaid expansion would have large negative financial and capacity effects on FQHCs in expansion states, similar to those resulting from 50% reductions in Medicaid revenues or Section 330 grants.

For FQHCs in nonexpansion states, Section 330 grants appear to play a larger role relative to Medicaid revenue, a finding consistent with those of previous studies. As FQHCs report using federal grants as the backbone on which staffing decisions are largely based, losing these funds would significantly impact all of FQHCs’ operations. Larger cuts in Medicaid rates and Section 330 grants would have more harmful effects on the FQHCs, and in the face of a more limited budget for Section 330 grants, FQHCs in nonexpansion states would suffer more.

These projections were designed to inform the downstream effects of decisions surrounding the HCP and highlight the value of HRSA‐funded health centers. Policy simulations outline the indirect impacts that Medicaid's growth, repeal, and cuts could have on FQHCs’ capacity for patient care (i.e., FTE, number of patients and visits) by way of Medicaid revenue. Fluctuations in Section 330 grants are directly tied to supporting FQHCs’ capacity and services rendered. Our policy simulations illustrate the larger impact of Section 330 grants on improving access to health services at HRSA‐funded health centers. Therefore, given FQHCs’ varying levels of reliance on Medicaid revenue, investing in these centers through Section 330 grants might be both more effective and equitable.

In 2017, FQHCs collected approximately $11.5 billion in total Medicaid revenue and received $4.7 billion from Section 330 grants. Comparing $1 million in Medicaid revenue with that in Section 330 grant funding, we found that the latter was associated with more staff (17 vs 6 FTEs), double the number of visits (7,406 vs 3,666 patient visits), and more than triple the number of patients (2,152 vs 599 patients).

These differences in the marginal effect of revenue from two funding sources should be interpreted in the context of different sets of expectations, requirements, timelines, and other constraints imposed by the funders. Reimbursements from Medicaid are less predictable and, in a sense, less durable in the long term; only the net profits could be invested in the following years. Grant funding has a fixed cycle, and FQHCs, as grantees, are accountable for showing newly developed clinical volume. In addition, it is important to consider the relative share of total revenue from each funding stream at baseline.

Section 330 grant funding has been essential to providing ongoing care to uninsured and underinsured FQHC patients, and the growing funding support through the CHCF has facilitated the establishment of new sites, expanded services, and greater capacity to serve patients. Furthermore, nonreimbursable services that facilitate patients’ access to care like enabling services rely heavily on the support of Section 330 grants. Because of challenges in recruiting and retaining providers and nonclinical support staff, to ensure FQHCs’ ability to plan ahead and support operations, federal investments should be predictable, sustained, and stable. The CHCF expired in September 2019, was reauthorized by the CARES Act through November 2020, was extended by a continuing resolution through December 2020, and was finally reauthorized by Congress for 2021 through 2023.

In summary, our study provides an opportunity to understand how alternative state and federal policies might impact the HCP. The ability to provide ambulatory care to patients regardless of their ability to pay requires that FQHCs have adequate financial resources, and policy actions determine whether or not these financial resources will be available.

Conclusions

Our simulations have provided important insights into the expected financial and service capacity outcomes of alternative policies. States’ decisions not to participate in or to repeal Medicaid expansion, to lower Medicaid's payment rates, or any federal funding cuts to the HCP could have a negative impact on FQHCs, resulting in tens of thousands of low‐income patients losing access to primary care. The expansion of Medicaid augmented FQHCs’ staffing and services, and further widened preexisting disparities between expansion and nonexpansion states. Conversely, states’ decisions to embrace Medicaid expansion may reduce these disparities. Our findings emphasize the disproportionate impact that cuts in Section 330 funds would have on the HCP's ability to serve America's most marginalized and historically underserved patients.

Funding/Support: This article was funded by the US Department of Health and Human Services (HHS), Health Resources and Services Administration (HRSA), under HRSA contract number HHSH250201300025I. The views expressed in this publication are solely the opinions of the authors and do not necessarily reflect the official policies of the US Department of Health and Human Services or the Health Resources and Services Administration, nor does mention of the department or agency names imply endorsement by the US government.

Conflict of Interest Disclosures: All authors have completed the ICMJE Form for Disclosure of Potential Conflicts of Interest. No conflicts were reported.

Supporting information

Online Appendix