Local Health Departments as Clinical Safety Net in Rural Communities

Local Health Departments as Clinical Safety Net in Rural Communities

Local Health Departments as Clinical Safety Net in Rural Communities Nathan L. Hale, PhD,1 Tamar Klaiman, PhD,2 Kate E. Beatty, PhD,1 Michael B. Meit,...

214KB Sizes 0 Downloads 70 Views

Local Health Departments as Clinical Safety Net in Rural Communities Nathan L. Hale, PhD,1 Tamar Klaiman, PhD,2 Kate E. Beatty, PhD,1 Michael B. Meit, MA, MPH3 Introduction: The appropriate role of local health departments (LHDs) as a clinical service provider remains a salient issue. This study examines differences in clinical service provision among rural/ urban LHDs for early periodic screening, diagnosis, and treatment (EPSDT) and prenatal care services. Methods: Data collected from the 2013 National Association of County and City Health Officials Profile of Local Health Departments Survey was used to conduct a cross-sectional analysis of rural/ urban differences in clinical service provision by LHDs. Profile data were linked with the 2013 Area Health Resource File to derive other county-level measures. Data analysis was conducted in 2015.

Results: Approximately 35% of LHDs in the analysis provided EPSDT services directly and 26% provided prenatal care. LHDs reporting no others providing these services in the community were four times more likely to report providing EPSDT services directly and six times more likely to provide prenatal care services directly. Rural LHDs were more likely to provide EPSDT (OR¼1.46, 95% CI¼1.07, 2.00) and prenatal care (OR¼2.43, 95% CI¼1.70, 3.47) services than urban LHDs. The presence of a Federally Qualified Health Center in the county was associated with reduced clinical service provision by LHDs for EPSDT and prenatal care. Conclusions: Findings suggest that many LHDs in rural communities remain a clinical service provider and a critical component of the healthcare safety net. The unique position of rural LHDs should be considered in national policy discussions around the organization and delivery of public health services, particularly as they relate to clinical services. (Am J Prev Med 2016;](]):]]]–]]]) & 2016 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.

Introduction

H

ealth experts and policymakers have discussed the appropriate role of local health departments (LHDs) as a clinical service provider over the past 25 years, but these discussions have come into sharper focus since the passing of the Affordable Care Act (ACA) in 2010.1–3 Prior to the ACA, a series of reports by the National Academy of Medicine (NAM) drew attention to the need for LHDs to focus on population-based public health services as opposed to From the 1Department of Health Services Management and Policy, East Tennessee State University, Johnson City, Tennessee; 2Health Policy and Public Health, University of the Sciences, Philadelphia, Pennsylvania; and 3 Public Health Research Department, NORC at the University of Chicago, Chicago, Illinois Address correspondence to: Nathan L. Hale, PhD, Department of Health Services Management and Policy, East Tennessee State University, Box 70264, Johnson City TN 37614. E-mail: [email protected]. 0749-3797/$36.00 http://dx.doi.org/10.1016/j.amepre.2016.05.012

providing direct clinical services to the community.4,5 The most recent NAM recommendations moved a step further, suggesting that LHDs should work to develop adequate capacity for clinical service delivery outside the health department and shift clinical care funds to support population-based services.6 Although increasing LHDs’ focus on population-based services is warranted, the recommendation to discontinue LHD clinical service provision in all communities remains a point of contention. Historically, expanding health insurance coverage for pregnant women and children through Medicaid programs provided greater access to private providers, subsequently reducing the demand for LHDs to provide clinical services directly.1 Though this may be true for LHDs operating in urban communities with many local providers, it may not be the case for LHDs operating in rural communities with few clinical providers.7 Rural LHDs remain more engaged in clinical service provision than their urban counterparts, particularly for

& 2016 American Journal of Preventive Medicine. Published by Elsevier Inc. All rights reserved.

Am J Prev Med 2016;](]):]]]–]]] 1

2

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

maternal child health (MCH) services.8 The American Academy of Pediatrics has noted that the supply of pediatricians in rural communities is insufficient to meet existing demand9 and the decline of obstetrics– gynecology providers in rural communities continues to undermine access to important women’s health services, including prenatal care.10,11 LHDs discontinuing clinical MCH services in rural communities where other providers may not have the capacity to absorb the increased demand, if they exist at all, may leave communities without access to necessary health services.7,12 An understanding of the landscape of clinical service provision among rural LHDs, and the context in which these services occur, is critical for informing larger policy discussions about the role of rural LHDs in their communities. This study addresses this issue by examining differences in clinical service provision among rural/ urban LHDs using two historic and longstanding MCH clinical services—early periodic screening, diagnosis, and treatment (EPSDT) and prenatal care. Of particular interest is the role of other public and private providers for these services in these communities. It is anticipated that LHD provision of clinical services for EPSDT and prenatal care would be more common in rural than urban communities and that health system capacity external to LHDs would be an important factor in reported clinical service provision among LHDs.

Methods Study Sample Data collected during the 2013 National Association of County and City Health Officials (NACCHO) Profile of LHDs Survey was used to conduct a cross-sectional analysis examining rural/urban differences for select MCH preventive clinical services among LHDs. The NACCHO Profile survey is distributed to all LHDs in the U.S. and is used to provide a description of LHD infrastructure and practice. In 2013, a total of 2,530 LHDs were eligible to complete the survey, and 2,000 (79%) responded.13 The NACCHO Profile Survey was linked with the 2013 Area Health Resource File to derive other important, county-level, sociodemographic and health system capacity measures.14 The reported LHD ZIP codes from the NACCHO Profile Survey were cross-walked with corresponding county codes and linked with select variables from the Area Health Resource File. Of the 2,000 LHDs responding to the initial profile survey, 89 (4.5%) of the original sample had missing or incomplete ZIP codes and could not be linked. The final sample included 1,911 LHDs. County codes for LHDs with ZIP codes crossing multiple county boundaries were assigned on the basis of the county with the majority of the population. County codes for LHDs responding to the survey as a multicounty jurisdiction were assigned based on the county of the lead LHD within the jurisdiction and applied across all counties served in the multicounty jurisdiction.

Measures The provision of preventive clinical services for EPSDT and prenatal care was of primary interest in this study. LHDs are asked to confirm which services had been provided by the LHD in the past year and how these services were provided. LHDs that indicated these services were performed by the LHD directly were considered as a direct provider of these services. The variable was coded as a dichotomous (yes/no) variable. Of primary interest to this study was the extent to which the provision of preventive MCH clinical services differed geographically. A three-level categorical variable measure of rurality was constructed using the 2013 Urban Influence Codes (UIC). UIC Codes 1 and 2 were used to identify large/small metropolitan areas (urban); UIC Codes 3, 5, and 8 were used to identify micropolitan counties; and the remaining UIC codes (4, 6, 7, 9–12) were used to identify non-core counties (rural).15 The Office of Management and Budget considers both micropolitan and non-core counties as “nonmetropolitan.”16 Micropolitan counties reflect communities with populations between 10,000 and 49,999 and are often conceptualized as larger rural communities, whereas the remaining non-core counties reflect smaller rural communities.16 This analysis focused on smaller rural counties most likely to experience provider shortages. The LHDs were also asked to indicate if services are provided by others in the community independent of LHD funding. This variable was of particular interest to this study as it reflects the extent to which LHDs reported additional capacity for these services within the community. Although this variable was selfreported by LHDs and may be related to the outcome variable of interest, LHDs’ response to this question was theoretically relevant and added context suggestive of community capacity and LHD tendencies toward clinical service provision. This variable was coded as a dichotomous (yes/no) variable reflecting if others also provided the service in the community. County-level measures for income, poverty, unemployment, high school graduation rates, and single-parent homes were used to construct a deprivation index that reflects varying degrees of community vulnerability with regard to social determinants of health. The variables included were standardized using z-scores, with the direction of positive or negative change across each variable in a consistent direction. Scores were summed and results divided into quartiles to derive the index used in the analysis. The index included all U.S. counties, with quartile values assigned to each county in the sample. This approach allowed for collapsing multiple variables into a single measure of community vulnerability. More detailed information regarding the development of the index and selection of variables is available.17 Additional county-level demographic and health system capacity variables from the Area Health Resource File were also linked with the NACCHO Profile data and used in the analysis. These included the presence of a Federally Qualified Health Center (FQHC) in the county and the ratio of primary care physicians to population. The level of Medicaid Managed Care (MMC) enrollment within each state and the decision to expand Medicaid as part of the ACA was also of interest. A flag was created for states with high MMC enrollment, defined as being in the top quartile of MMC enrollment compared with other states. A flag was also created to identify states that chose not to expand Medicaid at the time of this study. www.ajpmonline.org

Hale et al / Am J Prev Med 2016;](]):]]]–]]] The governance structure of LHDs (state, local, shared); the presence of a local board of health; and the population size served by each LHD was also included the analysis. These variables were derived from the NACCHO Profile. Population size was divided into a five-level categorical variable commonly used in other NACCHO publications.13

3

Table 1. Characteristics of Local Health Departments Included in the Study Sample

Characteristics

Local health departments, n (%) (n¼1,911)

Rural

Statistical Analysis

Urban

490 (25.66)

The characteristics of the study population were described and bivariate analysis for EPSDT and prenatal care services performed directly by LHDs was conducted. Chi-square tests for independence and unadjusted ORs were used to examine bivariate relationships. Multivariate logit models were used to examine adjusted relationships between the variables of interest and the outcome. All analyses were conducted in 2015 using Stata, version 12.

Micropolitan

697 (36.49)

Rural

723 (37.86)

Results Among LHDs responding to the survey, approximately 26% operated in urban counties, 36% were in micropolitan counties, and 39% in rural counties (Table 1). Distribution of LHDs across levels of area deprivation was relatively balanced, whereas the majority of LHDs fell under a county jurisdiction and had a local board of health. Approximately 40% of LHDs included in the sample operated in states that did not expand Medicaid under the ACA (Table 1). Related to health system capacity, 16% of LHDs operated in counties with the lowest physician to population ratio and 35% in counties with the highest ratio. Approximately 61% of LHDs were operating in counties that also had at least one FQHC. Approximately 35% of LHDs included in the analysis provided EPSDT services directly and 26% provided prenatal care (Table 2). Among LHDs that provided EPSDT services, 62% also indicated that no other provider in the community was delivering these services. LHDs reporting no other providers delivering EPSDT services were nearly four times as likely to also report providing these services directly. Similar findings were noted for prenatal care, with 44% of LHDs reporting no other provider delivering the same service in the community. More rural LHDs provided EPSDT and prenatal care services than their micropolitan and urban counterparts (Table 2). When compared with the least deprived communities, clinical service provision for EPSDT and prenatal care were more likely to occur in the most deprived communities. More than half of the LHDs providing EPSDT (53%) and prenatal care (50%) services were doing so in states that did not expand Medicaid. Clinical services for EPSDT and prenatal care were also less common in states with high MMC enrollment. Importantly, 46% of LHDs providing EPSDT services and 37% of those providing prenatal care were doing so in a county with no FQHC ] 2016

Area deprivation Not deprived

544 (28.45)

Somewhat deprived

489 (25.59)

Deprived

469 (24.56)

Most deprived

409 (21.4)

Governance category State

359 (18.77)

Local

1,383 (72.39)

Shared

169 (8.84)

Local board of health Yes

1,351 (70.7)

No

548 (28.68)

Population served r24,999

793 (41.51)

25,000–49,999

388 (20.3)

50,000–99,999

299 (15.65)

100,000–249,999

217 (11.33)

Z250,000

214 (11.22)

State expansion of Medicaid Yes

1,136 (59.47)

No

775 (40.53)

High MMC enrollment Top quartile (488.4%) Bottom quartiles (o88.3%)

430 (22.53) 1,481 (77.47)

Physicians per 1,000 population Quartile 1 (0–0.76)

306 (16.01)

Quartile 2 (0.77–1.22)

441 (23.08)

Quartile 3 (1.23–1.82)

492 (25.74)

Quartile 4 (41.83)

672 (35.18)

Federally qualified health center Yes

1,172 (61.31)

No

739 (38.69)

MMC, Medicaid Managed Care.

4

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

on record. Bivariate analysis also suggested a gradient of increasing primary care capacity and a decreasing likelihood of LHDs providing clinical services. After adjusting for additional LHD characteristics, community factors, and health services capacity, whether or not others in the community were providing the same service remained a strong predictor of LHDs providing these services directly. As noted in Table 3, LHDs reporting no others providing these services in the community were 44.5 times more likely to also report providing EPSDT services directly (AOR¼4.67, 95% CI¼3.74, 5.84) and 6.5 times more likely to provide prenatal care services directly (AOR¼6.59, 95% CI¼4.99, 8.69). Rural communities were also more likely to provide both EPSDT (AOR¼1.46, 95% CI¼1.07, 2.00) and prenatal care (AOR¼2.43, 95% CI¼1.70, 3.47) than their urban counterparts. Clinical service provision for EPSDT (AOR¼1.84, 95% CI¼1.45, 2.32) and prenatal care (AOR¼1.39, 95% CI¼1.08, 1.78) was also more likely to occur in states that did not expand Medicaid. LHDs operating in states with high MMC enrollment were less likely to provide EPSDT services (AOR¼0.66, 95% CI¼0.49, 0.89), but this was not the case for prenatal care. Adjusted analysis also suggested an association between the presence of an FQHC and reduced clinical service provision by LHDs. In counties with at least one FQHC also operating, LHDs were less likely to provide EPSDT (AOR¼0.70, 95% CI¼0.54, 0.90) and prenatal care (AOR¼0.56, 95% CI¼0.43, 0.75) services. The relationship between increasing primary care physician capacity and reduced clinical service provision by LHDs was attenuated in the adjusted analysis and no longer significant.

Discussion Study findings suggest that rural LHDs are more likely to provide direct clinical services for EPSDT and prenatal care. This finding is intuitive, as many LHDs operating in rural and underserved communities have historically provided these gap-filling services to compensate for deficiencies in primary care infrastructure and health insurance coverage more common in rural communities.1,3 About half of all LHDs providing clinical services also indicated there are no other community-based providers providing this service too. This study further found that additional health system capacity within these communities—such as the existence of an FQHC—to be an important factor associated with reduced clinical service provision among LHDs. Taken together, these findings suggest that many LHDs in rural communities remain a clinical service provider, and a critical component of the healthcare

safety net. As such, acting on the NAM recommendations to develop external capacity for clinical services5,6 could prove to be challenging for some LHDs, particularly in the current health policy environment. Previous research suggests LHDs operating in more-urban areas may be better situated to make this transition, but the impact of these transitions in rural communities on access to important MCH services could be substantial.7 These findings also raise additional concerns in the context of the ACA and continued health system reforms. It is unlikely the existing demand for EPSDT and prenatal care services provided by LHDs will change substantially with the implementation of the ACA. Most states already extend coverage to children and pregnant women at or beyond 138% of the federal poverty line targeted in the ACA Medicaid reforms. Although there are rural/urban differences in the duration, entry, and exit of Medicaid coverage,18 existing coverage of children through Medicaid and Child Health Insurance Programs has translated to comparable levels of insurance coverage for rural and urban children.19 Additionally, pregnant women remain eligible for Medicaid across states through presumptive eligibility. However, constrained provider supply in rural communities—which this study suggests is a key predictor of clinical service provision by LHDs—remains an ongoing challenge.9–11 These findings also raise important questions for public health practice that warrant further discussion and investigation. As the national focus on population health and health system reform continues, the role of LHDs within these larger systems is coming into sharper focus. Provisions in the ACA require nonprofit hospitals to conduct Community Health Needs Assessments every 3 years, which must include individuals with expertise in public health.20,21 Furthermore, the national movement toward accreditation of LHDs also emphasizes community health assessments and health improvement plans, which aim to advance population health initiatives.22 Although some rural LHDs have demonstrated the ability to leverage policy shifts and act as a catalyst for improving population health,23 evidence also suggests the capacity for most rural LHDs to engage in a wide array of population-based health services and pursue voluntary accreditation may be limited.24–27 Furthermore, the ACA may create financial incentives for LHDs to provide clinical services that run counter to the NAM recommendations. Though some evidence suggests high input costs from clinical services provision may disproportionately consume scarce financial resources, potentially at the expense of more population–based services,28,29 expanding insurance stemming from the ACA may provide LHDs with the opportunity to develop or enhance revenue streams from billing for clinical www.ajpmonline.org

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

5

Table 2. Proportion of LHDs Providing Maternal and Child Health Services by Covariates of Interest (2013) EPSDT Study variables

Yes, n (%) (n¼665, 34.80%)

Prenatal Care

No, n (%) (n¼1,246, 65.20%)

Unadjusted OR (LBL-UBL)

Yes, n (%) (n¼498, 26.04%)

No, n (%) (n¼1,413, 73.96%)

Unadjusted OR (LBL-UBL)

Others providing same service Yes

253 (38.06%)

881 (70.73%)

ref

277 (55.73%)

1,197 (84,70%)

ref

No

412 (61.94%)

364 (29.27%)

3.93 (3.22, 4.80)

220 (44.27%)

216 (15.30%)

4.40 (3.49, 5.54)

Metropolitan

139 (20.93%)

351 (28.18%)

ref

100 (20.19%)

390 (27.58%)

ref

Micropolitan

229 (34.42%)

468 (37.59%)

1.23 (0.96, 1.51)

175 (35.11%)

552 (36.97%)

1.30 (0.99, 1.70)

Rural

297 (44.65%)

426 (34.23%)

1.76 (1.37, 2.25)

223 (44.70%)

501 (35.45%)

1.72 (1.31, 2.25)

Not deprived

176 (26.51%)

367 (29.49%)

ref

128 (25.81%)

415 (29.38%)

ref

Somewhat deprived

141 (21.24%)

347 (27.91%)

0.84 (0.65, 1.11)

116 (23.28%)

373 (26.40%)

1.00 (0.75, 1.34)

Deprived

171 (25.77%)

297 (23.91%)

1.20 (0.92, 1.56)

123 (24.81%)

346 (24.47%)

1.15 (0.87, 1.54)

Most deprived

176 (26.48%)

232 (18.68%)

1.57 (1.20, 2.06)

130 (26.10%)

279 (19.74%)

1.51 (1.13, 2.01)

Rural

Area deprivation

Governance category State

164 (24.71%)

194 (15.60%)

ref

114 (22.87%)

245 (17.32%)

ref

Local

393 (59.00%)

991 (79.54%)

0.47 (0.37, 0.60)

320 (64.27%)

1,063 (75.25%)

0.65 (0.50, 0.84)

Shared

108 (16.29%)

61 (4.86%)

2.11 (1.45, 3.08)

64 (12.86%)

105 (7.43%)

1.31 (0.89, 1.92)

Local board of health Yes

500 (75.30%)

850 (68.90%)

1.38 (1.11, 1.70)

340 (68.38%)

1,011 (72.11%)

0.84 (0.67, 1.04)

No

164 (24.70%)

384 (31.10%)

ref

157 (31.62%)

391 (27.89%)

ref

r24,999

277 (41.72%)

515 (41.39%)

0.98 (0.72, 1.33)

163 (32.76%)

630 (44.59%)

0.59 (0.42, 0.81)

25,000– 49,999

135 (20.37%)

252 (20.26%)

0.98 (0.70, 1.37)

101 (20.33%)

287 (20.29%)

0.80 (0.56, 1.14)

50,000– 99,999

102 (15.46%)

196 (15.74%)

0.96 (0.67, 1.36)

97 (19.51%)

202 (14.28%)

1.09 (0.76, 1.57)

100,000– 249,999

73 (11.03%)

143 (11.49%)

0.93 (0.64, 1.36)

71 (14.22%)

146 (10.32%)

1.10 (0.75, 1.61)

Z250,000

76 (11.42%)

138 (11.11%)

ref

66 (13.18%)

149 (10.52%)

ref

Population served

State expansion of Medicaid Yes

310 (46.66%)

826 (66.31%)

ref

249 (50.09%)

887 (62.77%)

ref

No

355 (53.34%)

420 (33.69%)

2.25 (1.85, 2.72)

248 (49.9%)

526 (37.23%)

1.68 (1.37, 2.06)

325 (26.13%)

0.53 (0.42, 0.68)

95 (19.09%)

335 (23.74%)

0.76 (0.59, 0.98)

High MMC enrollment Top quartile (488.4%)

105 (15.79%)

(continued on next page)

] 2016

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

6

Table 2. Proportion of LHDs Providing Maternal and Child Health Services by Covariates of Interest (2013) (continued) EPSDT Study variables Bottom quartiles (o88.3%)

Prenatal Care

Yes, n (%) (n¼665, 34.80%)

No, n (%) (n¼1,246, 65.20%)

Unadjusted OR (LBL-UBL)

Yes, n (%) (n¼498, 26.04%)

No, n (%) (n¼1,413, 73.96%)

Unadjusted OR (LBL-UBL)

560 (84.21%)

920 (73.87%)

ref

402 (80.91%)

1,078 (76.26%)

ref

Physicians per 1,000 population Quartile 1 (0–0.76)

128 (19.21%)

178 (14.92%)

ref

91 (18.35%)

215 (15.18%)

ref

Quartile 2 (0.77–1.22)

197 (29.71%)

243 (19.53%)

1.13 (0.84, 1.53)

105 (21.02%)

336 (23.80%)

0.73 (0.52, 1.02)

Quartile 3 (1.23–1.82)

166 (25.02%)

325 (26.12%)

0.71 (0.53, 0.96)

147 (29.53%)

345 (24.40%)

1.00 (0.73, 1.37)

Quartile 4 (41.83)

173 (26.06%)

499 (40.05%)

0.48 (0.36, 0.65)

155 (31.10%)

517 (36.61%)

0.72 (0.52, 0.96)

Federally qualified health center Yes

359 (55.00%)

812 (65.21%)

0.63 (0.52, 0.76)

275 (55.28%)

896 (63.43%)

0.71 (0.58, 0.88)

No

306 (46.00%)

433 (34.79%)

ref

222 (44.72%)

517 (36.57%)

ref

Note: Boldface indicates statistical significance (po0.05). EPSDT, early periodic screening, diagnosis, and treatment; LBL, lower bound limit; LHD, local health department; MMC, Medicaid Managed Care; UBL, upper bound limit.

services they do provide. Revenue from clinical services can be an important source of funds for rural LHDs,12 and potentially serve as a link to other more population– based services provided by LHDs. These incentives could draw rural LHDs deeper into clinical service provision, creating an operational divide between LHDs serving rural and non-rural jurisdictions. Ultimately, the nature of the relationship between clinical service provision and engagement in population-based services remains the salient issue, and one that warrants further investigation.

Limitations This study is not without limitations. A cross-sectional study design was used to examine relationships and does not permit inferenced causal relationships among the outcomes and variables of interest. Data used for this study were self-reported by LHDs participating in the NACCHO survey and were not independently verified for accuracy. The information provided by LHDs may be incomplete, imperfect, or inconsistent. It should be noted that not all rural communities have LHDs, and the information contained in this study may not fully capture the entire scope of rural public health and clinical services. LHD response rates in centralized states—which

also tend to be concentrated in the south where clinical services are more commonly provided—may be lower than what is observed in decentralized states. Also, the approach taken to assign a county code for multicounty jurisdictions is a potential source of bias. However, use of the lead county LHD completing the survey may underestimate existing vulnerabilities within the smaller counties of a multicounty jurisdiction and is a more conservative approach. Furthermore, the survey inquired about specific program or service lines but did not probe the scale or scope of services. Although this study identified LHDs providing these services, it was not able to determine the extent to which these services were being provided within the overall context of LHD services. The authors also acknowledge the local political climate, and local governing board decisions on service mix cannot be accounted for in the study. This study does have several strengths. Although the body of research specific to local public health systems is growing, a very small proportion of the evidence focuses on rural health departments—which comprise approximately 60% of all LHDs. Although a cross-sectional study design was used, to the authors’ knowledge, this study is the first to examine factors associated with clinical service provision among LHDs and provides some context in which these services occurred. Only one www.ajpmonline.org

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

7

Table 3. Adjusted Multivariable Analysis of Select Maternal and Child Health Services by Covariates of Interest, 2013

Study variables No others providing service in community

EPSDT services

Prenatal care services

OR (LBL-UBL)

OR (LBL-UBL)

4.67 (3.74, 5.84)

6.59 (4.99, 8.69)

Rural Metropolitan

ref

Micropolitan

1.32 (0.99, 1.76)

1.36 (0.99, 1.86)

Rural

1.46 (1.07, 2.00)

2.43 (1.70, 3.47)

Area deprivation Not deprived

ref

Somewhat deprived

0.73 (0.53, 1.01)

0.91 (0.66, 1.26)

Deprived

0.86 (0.64, 1.20)

0.85 (0.62, 1.18)

Most deprived

1.31 (0.93, 1.85)

1.24 (0.86, 1.78)

Governance category State

ref

Local

0.47 (0.34, 0.65)

0.86 (0.62, 1.18)

Shared

1.96 (1.30, 2.95)

1.09 (0.70, 1.67)

2.08 (1.60, 2.70)

0.98 (0.76, 1.27)

r24,999

0.43 (0.28, 0.64)

0.16 (010, 0.25)

25,000–49,999

0.51 (0.33, 0.78)

0.40 (0.26, 0.62)

50,000–99,999

0.58 (0.38, 0.89)

0.71 (0.47, 1.07)

100,000–249,999

0.72 (0.47, 1.12)

0.96 (0.63, 1.44)

Local board of health Population served

Z250,000

ref

State not expanding Medicaid

1.84 (1.45, 2.32)

1.39 (1.08, 1.78)

High MMC enrollment (488.4%)

0.66 (0.49, 0.89)

0.99 (0.73, 1.33)

Physicians per 1,000 population Quartile 1 (0–0.76)

ref

Quartile 2 (0.77–1.22)

1.17 (0.82, 1.66)

0.66 (0.44, 0.98)

Quartile 3 (1.23–1.82)

0.81 (0.56, 1.16)

0.97 (0.65, 1.43)

Quartile 4 (41.83)

0.73 (0.50, 1.06)

0.87 (0.58, 1.32)

0.70 (0.54, 0.90)

0.56 (0.43, 0.75)

Federally qualified health center

Note: Boldface indicates statistical significance (po0.05). EPSDT, early periodic screening, diagnosis, and treatment; LBL, lower bound limit; MMC, Medicaid Managed Care; UBL, upper bound limit.

study7 examined changes in rural public health systems and receipt of clinical services over time, but this was limited to a single state. More qualitative studies specific to rural public health are available and provide important information; however, these are limited to small geopolitical jurisdictions.26,27

] 2016

Conclusions As ACA implementation and health system reform continues, the role of LHDs within the larger health system remains a salient issue. This study provides important insight into rural/urban differences in clinical service provision for two historic MCH services and

8

Hale et al / Am J Prev Med 2016;](]):]]]–]]]

underscores the importance of rural LHDs as a provider within the healthcare safety net. The unique position of rural LHDs should be considered in national policy discussions around the organization and delivery of public health services, specifically as it relates to the feasibility of the current NAM recommendations and continued national push away from clinical services provided by LHDs. This work was funded in part by the U.S. DHHS under contract to the Assistant Secretary for Planning and Evaluation (ASPE). The findings and conclusions of this report are those of the authors and do not necessarily represent the views of ASPE or DHHS. Dr. Hale conceptualized the study, conducted the data analysis, and participated in the drafting of the manuscript. Dr. Klaiman also conceptualized the study and participated in the drafting of the manuscript. Dr. Beatty conceptualized the study, provided support with data analysis, and participated in the drafting of the manuscript. Mr. Meit participated in the conceptualization of the study and drafting of the manuscript. No financial disclosures were reported by the authors of this paper.

References 1. Wall S. Transformations in public health systems. Health Aff. 1998;17 (3):64–80. http://dx.doi.org/10.1377/hlthaff.17.3.64. 2. Berkowitz B, Ivory J, Morris T. Rural public health: policy and research opportunities. J Rural Health. 2002;18:186–196. http://dx.doi.org/ 10.1111/j.1748-0361.2002.tb00930.x. 3. Berkowitz B. Rural public health service delivery: promising new directions. Am J Public Health. 2004;94(10):1678–1681. http://dx.doi. org/10.2105/AJPH.94.10.1678. 4. IOM. Committee on the Study to Improve Public Health. The Future of Public Health. Washington, DC: National Academies Press; 1998. 5. IOM. Committee on Assuring the Health of the Public in the 21st Century. The Future of the Public’s Health in the 21st Century. Washington, DC: National Academies Press; 2003. 6. IOM. Committee on Public Health Strategies to Improve. For the Public’s Health: Investing in a Healthier Future. Washington, DC: National Academies Press; 2012. 7. Hale NL, Smith M, Hardin J, Brock-Martin A. Rural populations and early periodic screening, diagnosis, and treatment services: challenges and opportunities for local public health departments. Am J Public Health. 2015;105(suppl 2):S330–S336. http://dx.doi.org/10.2105/ AJPH.2014.302449. 8. Beatty KE, Meit M, Hale N, Khoury A, Masters P. Clinical service delivery disparities along the urban/rural continuum. Frontiers in Public Health Services and Systems Research. February 21, 2015. 9. Basco WT, Rimsza ME. Pediatrician workforce policy statement. Pediatrics. 2013;132(2):390–397. http://dx.doi.org/10.1542/peds.20131517. 10. American College of Obstetricians and Gynecologists. ACOG Committee Opinion No. 586: Health disparities in rural women. Obstet Gynecol. 2014;123(2, pt 1):384–388.

11. Cohen D, Coco A. Declining trends in the provision of prenatal care visits by family physicians. Ann Fam Med. 2009;7(2):128–133. http: //dx.doi.org/10.1370/afm.916. 12. Johnson R, Morris T. Rural Public Health: Issues and Considerations. A Report to the Secretary, U.S. Dept. of Health and Human Services. The National Advisory Committee on Rural Health; February 2000. 13. National Association of County and City Health Officials. 2013 National Profile of Local Health Departments. Washington, DC: NACCHO; 2014. 14. U.S. DHHS, Health Resources and Services Administration Health Workforce. Area Health Resources Files (AHRF): National, State and County Health Resources Information Database. Rockville, MD: U.S. DHHS. 15. Baer LD, Johnson-Webb KD, Gesler WM. What is rural? A focus on urban influence codes. J Rural Health. 1997;13(4):329–333. http://dx. doi.org/10.1111/j.1748-0361.1997.tb00975.x. 16. Hart LG, Larson EH, Lishner DM. Rural definitions for health policy and research. Am J Public Health. 2005;95(7):1149–1155. http://dx.doi. org/10.2105/AJPH.2004.042432. 17. Hale N, Probst J, Robertson A. Rural area deprivation and hospitalizations among children for ambulatory care sensitive conditions. J Community Health. 2016;41(3):451–460. 18. Coburn AF, McBride TD, Ziller EC. Patterns of health insurance coverage among rural and urban children. Med Care Res Rev. 2002;59 (3):272–292. http://dx.doi.org/10.1177/1077558702059003003. 19. Crosby R, Monica ML, Vanderpool RC, Casey B. Rural Populations and Health: Determinants, Disparities, and Solutions. San Francisco, CA: Jossey-Bass; 2012. 20. Beatty KE, Wilson KD, Ciecior A, Stringer L. Collaboration among Missouri nonprofit hospitals and local health departments: content analysis of community health needs assessments. Am J Public Health. 2015;105:S337–S344. http://dx.doi.org/10.2105/AJPH.2014.302488. 21. Patient Protection and Affordable Care Act. 42, U.S.C. § 18001 (2010). 22. Wilson KD, Mohr LB, Beatty KE, Ciecior A. Describing the continuum of collaboration among local health departments with hospitals around the community health assessments. J Public Health Manag Pract. 2014;20 (6):617–625. http://dx.doi.org/10.1097/PHH.0000000000000030. 23. Sampson G, Miner Gearin KJ, Boe M. A rural local health department– hospital collaborative for a countywide community health assessment. J Public Health Manag Pract. 2015;21(1):23–30. http://dx.doi.org/ 10.1097/PHH.0000000000000088. 24. Harris JK, Mueller NL. Policy activity and policy adoption in rural, suburban, and urban local health departments. J Public Health Manag Pract. 2013;19(2):E1–E8. http://dx.doi.org/10.1097/PHH.0b013e318252ee8c. 25. Wetta RE, Pezzino G, LaClair B, Orr S, Brown MB. Voices across Kansas. J Public Health Manag Pract. 2014;20(1):39–42. http://dx.doi. org/10.1097/PHH.0b013e3182a9572a. 26. Beatty KE, Mayer J, Elliott M, Brownson RC, Abdulloeva S, Wojciehowski K. Barriers and incentives to rural health department accreditation. J Public Health Manag Pract. 2016;22(2):138–148. http://dx. doi.org/10.1097/PHH.0000000000000264. 27. Beatty K, Harris JK, Barnes PA. The role of interorganizational partnerships in health services provision among rural, suburban, and urban local health departments. J Rural Health. 2010;26(3):248–258. http://dx.doi.org/10.1111/j.1748-0361.2010.00285.x. 28. Mays GP, McHugh MC, Shim K, et al. Getting what you pay for: public health spending and the performance of essential public health services. J Public Health Manag Pract. 2004;10(5):435–443. http://dx. doi.org/10.1097/00124784-200409000-00011. 29. Mays GP, Smith SA. Evidence links increases in public health spending to declines in preventable deaths. Health Aff. 2011;30(8):1585–1593. http://dx.doi.org/10.1377/hlthaff.2011.0196.

www.ajpmonline.org