Outline ONLY for a research paper. (will post for paper later)

Hospital Contributions to the Delivery of Public Health Activities in US Metropolitan Areas: National and Longitudinal Trends Rachel A. Hogg, DrPH, MA, Glen P. Mays, PhD, MPH, and Cezar B. Mamaril, PhD, MA

Public health services in the United States are delivered through local and state public health systems that are characterized by partnerships between governmental public health agencies and other public and private organizations. The importance of these partnerships was under- scored in 2003 when the Institute of Medicine called for the use of partnerships and coordi- nation with local health systems to help de- crease inefficiency and duplication of public health efforts across community partners implementing programs to improve population health.1 Although the primary responsibility for protecting the public’s health lies with the governmental public health agency, commu- nity partners can help local health departments deliver services more effectively.2 When used correctly, partnerships may allow organizations to provide services they are not able to do alone by facilitating the sharing of information, resources, and expertise across organizations.3---5

The recent economic downturn may have made partnerships an even more effective tool to ensure the provision of public health services, given the current and predicted cuts in staffing at many state and local health agencies.6,7

Hospitals play an important role in local public health systems and provide public health activities targeted at improving popula- tion health. Public health activities typically include monitoring community health status, investigating health problems and hazards in the community, developing policies that support individual and community health, educating and informing individuals about prevention, developing policies and laws to protect and promote health, and ensuring the quality and safety of air, water, and food.1

Reform measures in the Patient Protection and Affordable Care Act and recent reports such as the 2012 Institute of Medicine report on primary care and public health represent

a national movement that calls for greater integration of health care and public health systems, with hospitals taking focus as an important actor in potential collaboration.8---10

However, substantial gaps exist in what is known and understood about hospitals’ engagement in traditional public health activi- ties and how that may change over time in light of policy changes and priority shifts. These gaps may primarily result from the relatively long time it may take to see changes in system partnerships and health outcomes and the rather complicated nature of the analysis necessary to control for the many confounding factors in these relationships.3,5

A substantial amount of research exists regarding the provision of charity care services and Medicaid payment shortfall by hospitals, but very little work has examined how hospi- tals participate in public health activities.11---15

Previous research has suggested that whether a hospital contributes to public health activities

will be influenced by certain market and community characteristics. Economic and noneconomic factors including hospital type, market share, and community health needs are likely to determine the extent of public health activity participation.12,14,16 Hospital participa- tion in public health systems may affect the ability of the system to deliver public health activities by reducing the burden on the local health department. The local public health system may be able to operate more efficiently and provide a larger scope of activities by relying more heavily on other organizations, such as the hospital, to perform certain public health functions.17

Evidence has suggested that partnerships and collaboration in public health activities matter, making it likely that hospital participa- tion will play an important role in population health protections.3,4,18 Mays et al.18 identified hospitals as one of the leading contributors in the public health system, although the scope of

Objectives. We investigated changes in hospital participation in local public

health systems and the delivery of public health activities over time and

assessed the relationship between hospital participation and the scope of

activities available in local public health systems.

Methods. We used longitudinal observations from the National Longitudinal

Survey of Public Health Systems to examine how hospital contributions to the

delivery of core public health activities varied in 1998, 2006, and 2012. We then

used multivariate regression to assess the relationship between the level of

hospital contributions and the overall availability of public health activities in the

system.

Results. Hospital participation in public health activities increased from 37% in

1998 to 41% in 2006 and down to 39% in 2012. Regression results indicated

a positive association between hospital participation in public health activities

and the total availability of public health services in the systems.

Conclusions. Hospital collaboration does play an important role in the overall

availability of public health services in local public health systems. Efforts to

increase hospital participation in public health may have a positive impact on the

scope of services provided and population health in US communities. (Am J

Public Health. 2015;105:1646–1652. doi:10.2105/AJPH.2015.302563)

RESEARCH AND PRACTICE

1646 | Research and Practice | Peer Reviewed | Hogg et al. American Journal of Public Health | August 2015, Vol 105, No. 8

services in which they participated was still relatively narrow at less than 40%. The range of activities in which hospitals participate can vary, and previous studies have found that hospitals are most likely to collaborate with the local health agency in the delivery of services, administration of population-based programs, and assessments of community health needs.2,3,18

Common goals, mission, and vision can also help align public health and hospital collabo- rative efforts, which can be seen in the influx of partnerships resulting from preparedness funding and a drive toward ensuring commu- nity readiness and protection.19 Duplication and activity redundancy can also occur in public health systems in which poor commu- nication and congestion problems prevent positive collaboration between hospitals and other organizations.1,20

Heightened interest in the bridging of hos- pitals and public health systems coupled with the existence of very little empirical evidence about the type of public health activities in which hospitals participate, raises important questions about the role hospitals play in the provision of public health protections. We used longitudinal observations to examine 3 central questions: (1) to what extent do hospitals participate in delivering public health activities within communities, (2) how has this partici- pation changed over time, and (3) what effect do the contributions made by hospitals have on the availability of public health activities in communities?

METHODS

We used a longitudinal cohort design of data from the National Longitudinal Survey of Public Health Systems to analyze how hospital participation in local public health systems has changed over time and affects the scope of public health activities available in the system. A stratified random sample of local health officials from nationally representative local public health systems (n = 397) were surveyed in 1998, 2006, and 2012 (70% response rate) to measure the availability of 20 core public health activities within their jurisdictions, the range of organizations that delivered each activity, and the perceived effectiveness of each activity. The sample included all US agencies meeting the national definition of a local health

department that served a population of 100 000 or more residents.21 Although this sample is limited to agencies that serve large jurisdictions, it does cover about 65% of the US population.2

The National Longitudinal Survey of Public Health Systems captures how many of the 20 core public health activities are being provided in the community and to what extent other public and private organizations are contributing in providing those activities.2

The activities included in the survey can be categorized into the 3 core functions of public health: assessment, policy development, and assurance. These 20 activities were identified on the basis of public health professionals’ expert opinion and their representation of common public health activities.22,23 We generated descriptive statistics to examine the variation in what activities hospitals typically contribute to and how those contributions have changed over the 3 time points included in the data. To maintain consistent observations across all years, we examined 19 of the 20 activities surveyed. We ran equality of pro- portion tests for statistically significant differ- ences across each year for each activity using the 2012 data as the comparison year.

We used a multivariate regression model for panel data to estimate the association between hospital participation in the local public health system and the scope of public health activities provided. To estimate a relationship between hospital participation and the supply of public health activities in the system, we also needed to control for other factors associated with the provision of public health, including relevant socioeconomic and demographic measures. For the purpose of this portion of the analysis, we included only those participants who responded to all 3 waves of the National Longitudinal Survey of Public Health Systems and had matching control variable data (n = 631). In addition, because we used longitudinal data the model needed to control for autocorrelation that exists between observations of the same public health systems over time. We used a fixed-effects specification to control for any time-invariant confounding factors that may have had an impact on the dependent variable.

The main dependent variable of interest is the overall proportion of public health activities available in the local public health system.

The model controlled for organizational and community characteristics along with variables that account for the participation of other organizations in the system in public health activities (see Table 1 for variable descriptive statistics). We ran specification tests to ensure that the most appropriate variables were in- cluded and to rule out omitted variable bias in the model. We ran a correlation matrix to determine how related the various organiza- tions’ levels of participation were to each other and to assess for multicollinearity in the model. Some of the variables were moderately to highly correlated (Table 2). However, using the Kennedy threshold of 10, the variance inflation factor test did not indicate high enough levels for model correction.24 We ran a second model with interaction terms included to account for dual participation between hospitals in the public health system and the other organizations contributing to public health activities. By na- ture, the activities included in the survey data likely represent a collaborative enterprise linked to many organizations in the system. Including these interaction terms between the hospital and the other organizations helps identify any sig- nificant impact the organizations made together on the total availability of public health activities available while also testing for duplication.

We performed logarithmic transformation on the independent organization participation measures and dependent variable to reduce skewness and create a more normal distribu- tion to improve model fit. A portion of the local health departments in our sample did report no participation of certain organizations captured by the National Longitudinal Survey of Public Health Systems. Because the natural log of zero is undefined, we added a small positive constant of 0.001 to the participation measures to ensure that the log function would work. Sensitivity tests were run to ensure that the addition of a positive constant would not have a significant impact on the model estimates. Given that multiple time points were included in the sample, we included a time trend variable in the model to control for systematic differences across time.

RESULTS

Hospital participation in public health activ- ities has fluctuated overtime. The level of joint

RESEARCH AND PRACTICE

August 2015, Vol 105, No. 8 | American Journal of Public Health Hogg et al. | Peer Reviewed | Research and Practice | 1647

production in all activities increased from about 37% in 1998 to 41% in 2006 and down to 39% in 2012 (Table 3). Among the 3 functions of public health, hospitals are most likely to participate in assessment and policy activities. In 1998 and 2012, hospitals’ highest area of contribution was in policy activities; in 2006, assessment activities were the highest. Some variation has occurred in the top 5 activities in which hospitals are most likely to participate (Table 3). Hospitals have consis- tently had a high level of contribution to the community needs assessment process, investi- gation of adverse health events, and developing a network of support and communication re- lationships that include health organizations and the media. Development of a community health action plan and activities that link people

to services providing age-specific health needs were the next highest in 1998, with develop- ment of the health action plan remaining consistent in 2006 and linking people to services dropping in favor of laboratory ser- vices related to investigation of adverse health events and meeting routine diagnostic needs. A similar pattern emerged in 2012, with labora- tory services remaining in the top 5 and pro- cess to prioritize community health needs replacing development of a community health action plan. As previously observed, all of the activities in which hospitals were most likely to contribute fell into the assessment and policy categories with the exception of activities that link people to age-specific services, where levels dropped from 58% in 1998 to 39% in 2012, a statistically significant difference.

Those activities in which hospitals were con- sistently least likely to participate across the 3 years fell into the assurance category and included organizational assessment of the local public health agency, regular evaluations of the effects of public health services on community health status, and use of recognized process and outcome measures to monitor public health programs.

Estimates from the fixed-effects multivariate regression model with and without interaction terms are shown in Table 4. Both models found a positive association between hospital partic- ipation in public health activities and the total availability of public health services in the systems (P < .001), and the association remained consistent even after controlling for the contri- butions of other public and private organizations

TABLE 1—Changes in Hospital Participation: National Longitudinal Survey of Public Health Systems, United States, 1998, 2006, and 2012

Hospital Participation in 19 Core Public Health Activities 1998 (n = 351), % 2006 (n = 236), % 2012 (n = 241), %

All activities 37.17 41.15 39.28

Assessment activities 40.65 46.30 43.02

Policy activities 42.00 45.68 45.29

Assurance activities 28.73 31.90 29.53

By activity

Assessment activities

1. Community needs assessment process that describes the prevailing health status in the jurisdiction 58.12 62.29 61.41

2. Survey of the population for behavioral risk factors 22.90 14.04 22.41

3. Investigation of adverse health events, including communicable disease outbreaks and environmental health hazards 56.13 75.32a 59.75

4. Laboratory services to support investigations of adverse health events and meet routine diagnostic and surveillance needs 48.86 63.40 55.60

5. Analysis of the determinants of and contributing factors to priority health needs, and the population groups most affected 46.55 50.64 43.15

6. Analysis of age-specific participation in preventive and screening services 13.22 11.97 15.77

Policy development activities

7. Network of support and communication relationships that includes health-related organizations and the media 66.38 73.08 75.52

8. Activities to inform elected officials about the potential public health impact of decisions under their consideration 26.57a 40.34 39.83

9. Process to prioritize community health needs 49.43 53.65 56.02

10. Development of community health action plan with community participation 60.63 65.24 52.70

11. Process to allocate resources in a manner consistent with community health priorities 34.48 39.48 40.25

12. Implementation of community health initiatives consistent with established priorities 16.00 17.17 16.60

Assurance activities

13. Deployment of resources to address priority health needs 35.16 30.47 36.10

14. Organizational assessment of the local public health agency 3.17 a

7.73 9.13

15. Activities that link people to services that address age-specific priority health needs 57.64a 50.64 39.42

16. Regular evaluations of the effects of public health services on community health status 13.51 11.64 12.03

17. Use of recognized process and outcome measures to monitor public health programs 11.27 6.90 10.79

18. Process to provide public information about community health status, needs, behaviors, and policy issues 48.71 61.64 53.94

19. Reports on community health issues regularly provided to the media 32.86 a

53.22 51.87

aP < .003 used, based on Bonferroni adjustment to account for multiple comparisons.

RESEARCH AND PRACTICE

1648 | Research and Practice | Peer Reviewed | Hogg et al. American Journal of Public Health | August 2015, Vol 105, No. 8

in local public health systems. After adjusting the main effect coefficient to account for interaction terms in the model, a 10% increase in hospital participation was associated with a 0.9% in- crease in the scope of public health services available in the local system.

Local health department and physician or- ganization participation in public health activ- ities was also positively associated with the overall availability of public health activities. Local health departments had the highest parameter estimate, with a 10% increase in

participation being associated with a 4% in- crease in activities. The physician organization parameter estimates were substantially less, both falling at a less than 1% increase in the scope of available activities. Of note, although not statistically significant, federal agencies,

TABLE 2—Variables Used in Model: National Longitudinal Survey of Public Health Systems, United States, 1998, 2006, and 2012

Variable 1998, Mean (SD) 2006, Mean (SD) 2012, Mean (SD)

Dependent variables

Total availability of all public health activities, % 65.24 (17.67) 70.23 (17.43) 67.68 (18.53)

Independent variables

Hospital participation, % 38.42 (20.27) 40.86 (19.22) 40.05 (20.96)

Local health department participation, % 38.49 (14.62) 40.13 (14.39) 36.98 (13.14)

Physician organization participation, % 21.04 (17.32) 24.00 (17.36) 19.84 (17.17)

Local agency participation, % 32.76 (21.51) 51.59 (24.22) 27.12 (21.87)

Health insurer participation, % 8.86 (14.70) 9.98 (14.40) 10.35 (16.44)

Community health center participation, % 12.54 (17.29) 28.11 (21.73) 27.20 (21.49)

Federal agency participation, % 6.87 (12.46) 11.69 (14.79) 8.64 (14.33)

Any state agency participation, % 36.76 (20.61) 46.53 (20.77) 38.77 (22.46)

School participation, % 27.71 (20.20) 27.71 (20.31) 25.78 (18.57)

University participation, % 16.29 (18.11) 21.56 (20.48) 22.36 (20.10)

Any nonprofit participation, % 32.58 (19.05) 34.20 (19.56) 32.34 (20.78)

Population below the poverty level, % 12.14 (5.26) 10.73 (3.93) 14.86 (4.93)

Population size of jurisdiction, no. 421 769.40 (855 482.20) 496 375.50 (967 630.70) 468 703.50 (843 759.60)

Centralized public health governancea 0.13 (0.34) 0.13 (0.33) 0.06 (0.23)

Mixed or shared public health governance 0.15 (0.36) 0.14 (0.35) 0.15 (0.35)

Hospital beds per 100 000 335.15 (209.20) 292.78 (183.07) 222.73 (112.96)

MDs available per 100 000 243.33 (229.95) 267.05 (200.25) 275.88 (196.17)

No. of hospitals in jurisdiction 5.94 (9.96) 5.55 (8.73) 4.31 (6.68)

aExpressed as a binary term, with 0 meaning no and 1 meaning yes.

TABLE 3—Pearson Correlation Coefficients for Participating Organizations: National Longitudinal Survey of Public Health Systems, United

States, 1998, 2006, and 2012

Organization Hospital LHD Physician Local Agency Health Insurer CHC Federal Agency Any State Agency School University Any Nonprofit

Hospital 1.00 .47 .60 .57 .37 .47 .31 .47 .48 .49 .72

Local health department .47 1.00 .37 .47 .24 .25 .25 .39 .37 .28 .48

Physician organization .60 .37 1.00 .47 .49 .49 .39 .42 .44 .42 .49

Local agency .57 .47 .47 1.00 .33 .42 .41 .51 .41 .42 .61

Health insurer .36 .24 .49 .33 1.00 .36 .40 .32 .27 .42 .39

Community health center .47 .25 .49 .42 .36 1.00 .33 .39 .38 .44 .49

Federal agency .31 .25 .39 .41 .40 .33 1.00 .47 .29 .44 .35

Any state agency .47 .39 .42 .51 .32 .39 .47 1.00 .30 .39 .45

School .48 .37 .44 .41 .27 .38 .29 .30 1.00 .39 .42

University .49 .28 .42 .42 .42 .44 .44 .39 .39 1.00 .52

Any nonprofit .72 .48 .49 .61 .39 .49 .35 .45 .42 .52 1.00

Note. LHD = local health department; CHC = community health center.

RESEARCH AND PRACTICE

August 2015, Vol 105, No. 8 | American Journal of Public Health Hogg et al. | Peer Reviewed | Research and Practice | 1649

schools, and universities all had negative co- efficient estimates.

A positive and statistically significant inter- action between hospital and local health de- partment participation (P < .01) suggests an increased scope of services when the 2 orga- nizations are participating in the local public health system together. Hospitals’ interaction with physician organizations and universities were also statistically significant (P < .01), and although the interaction with physician orga- nizations had a positive coefficient, that with universities was negative.

DISCUSSION

Hospitals do participate in local public health systems and do contribute to the availability of local public health activities. The proportion of activities hospitals participate in producing and what activities they participate in have both changed over time. Our results suggest that hospitals are most likely to participate in public health activities that focus on assessment and policy. Consistent hospital participation in community needs assessment and the investi- gation of adverse health events highlights an

important area in which hospitals have the potential to affect population health. In recent years, hospitals have also increased their par- ticipation in the provision of laboratory ser- vices that may include the support of adverse event investigation and routine diagnostic and surveillance needs. This agrees with previous research and theory that highlighted popula- tion programs and community health assess- ments as the public health activities hospitals are most likely to participate in producing.2,3,18

Participation in these assessment public health activities suggests that hospitals may already be in tune with community and population needs. Targeting increased collaboration between hospital and local health department efforts to assess community needs may provide a useful opportunity for resource sharing, especially in communities in which both entities are dupli- cating assessments.

Consistent participation in the development of support and communication relationships that include health-related organizations sug- gests that hospitals play an important role in bridging relationships that may have a positive impact on population health. Hospitals might provide an untapped source of community engagement that local health departments could leverage as a tool to help build and support the local public health system. As of 2012, hospitals have also increased their participation in the process of prioritizing community health needs. This is likely a policy extension of their required commu- nity health assessments and may also provide initial insight into hospitals’ consideration of the community-benefit provision portion of the Affordable Care Act. Initial findings that in- dicate hospital joint participation in community health needs assessment may provide a useful starting point for examining the portions of health reform targeted at increasing hospital participation in public health.

The low level of hospital participation in public health assurance activities highlights an important area in which stakeholders and local health departments may want to specifically target efforts toward increasing collaboration. Overall percentages of hospital participation in assurance activities rank significantly lower than assessment and policy; however, some of the assurance activities may be critical to ensuring population health within a

TABLE 4—Multivariate Regression Model of the Influence of Participation Measures on the

Scope of Public Health Activities Available: National Longitudinal Survey of Public Health

Systems, United States, 1998, 2006, and 2012

Variable

Fixed Effects, No Interaction Terms,

Coefficient (SE)

Fixed Effects, Interaction Terms,

Coefficient (SE)

Hospital participation (log) .056** (.011) .161** (.032)

Local health department participation (log) .266** (.02) .404** (.037)

Physician organization participation (log) –.010 (.006) .020* (.01)

Local agency participation (log) .020** (.006) .012 (.011)

Health insurer participation (log) –.002 (.004) .002 (.006)

Community health center participation (log) –.001 (.005) .008 (.007)

Federal agency participation (log) .001 (.004) –.001 (.006)

Any state agency participation (log) .049** (.01) .030 (.017)

School participation (log) .001 (.008) –.004 (.011)

University participation (log) .010* (.005) –.009 (.007)

Any nonprofit participation (log) .043** (.009) .021 (.014)

Hospital · local health department .067** (.017) Hospital · physician organization .022** (.007) Hospital · local agency –.007 (.006) Hospital · health insurer .005 (.005) Hospital · community health center .009 (.005) Hospital · federal agency –.001 (.005) Hospital · any state agency –.009 (.012) Hospital · school –.003 (.006) Hospital · university –.014** (.005) Hospital · any nonprofit –.011 (.007) % of population below poverty level –.001 (.005) .000 (.005)

Population size .000 (.000) .000 (.000)

Population squared .000 (.000) .000 (.000)

Centralized public health governance .047 (.062) .000 (.060)

Mixed or shared public health governance .043 (.040) .033 (.038)

Hospital beds per capita .000 (.000) .034 (.000)

MDs available per 100 000 .000 (.000) .000 (.000)

No. of hospitals in jurisdiction .000 (.006) .000 (.006)

*P < .05; **P < .01.

RESEARCH AND PRACTICE

1650 | Research and Practice | Peer Reviewed | Hogg et al. American Journal of Public Health | August 2015, Vol 105, No. 8

community. Targeting resources toward prior- ity community health needs, linking people to the services they need, and evaluating the effectiveness of public health services are ac- tivities that already coincide with many of the core public health services in which hospitals actively participate. Participation in assurance activities may represent a significant gap in the system in which hospitals could be playing a more active role.

This analysis does suggest that hospital par- ticipation in the local public health system is associated with an increasing scope of public health services available in the community, especially when the hospital and local health department are contributing together. Efforts to increase hospital participation in public health may have a positive impact on the scope of services provided and population health in our communities. Stakeholders may want to target hospitals as potential service- and resource- sharing mechanisms for the local health depart- ment. Our findings confirm that health reform efforts targeted toward increasing the role of hospitals in public health may have a positive impact on the local public health system.

Our findings provide evidence of a synergis- tic relationship between hospitals and local health departments working together in local public health systems. The positive and signif- icant interaction terms indicate that more is gained when the 2 organizations work together in services, as opposed to one substituting for the other. Similar relationships were found to exist between hospitals and physician organi- zations. Open communication and coordinated collaboration between organizations may be an important factor to synergistic relationships by helping to prevent waste and unnecessary service duplication. The existence of a syner- gistic relationship also indicates that organiza- tions go beyond substitution in how they provide public health services, and other forms of collaboration would have a positive impact on the scope of services as well.

Although the hospital was the primary or- ganization of interest in this analysis, our findings also suggest some important informa- tion about the other organizations contributing to the local public health system. Local health department and physician organization partic- ipation in public health activities were also positively associated with the overall

availability of public health activities. These findings are not surprising given that local health departments tend to be the frontline providers of public health services. Our initial findings suggest that further investigation into the relationship between the physician organi- zations in a community and local public health may be a worthwhile investigation.

We also found a negative, although not significant, relationship between increased federal agency, school, and university partici- pation in public health and the overall scope of activities. A number of possible reasons exist that may provide insight into a negative asso- ciation. Federal organizations typically partici- pate in local public health systems in times of need or disaster, and a restriction of the activities available would not be surprising as a result of the concentrated focus on a few critical services that would arise. Similar cir- cumstances may arise regarding university participation. In addition, universities may act as an advisor or consultant in local public health rather than as an active partner in public health services. School participation in public health likely has a very narrow focus on promotion of activities such as healthy eating and physical activity among children and ado- lescents.25,26 Increased participation of schools may limit the scope of overall services because of that narrow focus and specialization. Future research that teases out these negative rela- tionships may help provide important insights into some constraints and downfalls to collab- oration in local public health systems.

Although we have provided some important findings regarding hospital collaboration in local public health systems and its effects, there are several limitations to this study and future analyses to be explored. First, the sample in- cludes local health systems that serve a popu- lation of 100 000 or more and cannot be generalized to those jurisdictions with smaller populations. Second, our analysis quantifies the relationship between the hospital and the local public health system but does not provide detailed information on the individual com- munity characteristics that may predict suc- cessful collaboration or act as barriers. Further analyses that examine the factors that indicate successful hospital participation in public health could provide valuable information to those communities and local health departments

seeking to create or enhance a partnership. In addition, because the data are being reported from the local health department perspective, it is possible that the full spectrum of public health activities in which hospitals participate may be over- or underreported. Validity tests performed during the first wave of data col- lection found that the local health official accurately reported those activities hospitals were engaged in providing.

Third, our measures of organization partic- ipation only examined whether the hospital participated in the public health activity, not how it contributed. Identifying whether hospi- tals provide services or resources to public health activities may help increasing under- standing of the relationship. Last, the avail- ability of local public health activities is likely in part the result of several unmeasured factors in the system, making the hospital participation variable endogenous with the dependent vari- able in this analysis.

We have identified what public health ac- tivities hospitals participate in and provided empirical evidence of what hospital participa- tion in production can do to the total supply of activities and health protections in the local public health system. The range of activities in which hospitals are most likely to participate does suggest that hospitals support population health activities, but the fluctuation in levels of engagement indicates that other factors may influence increases and decreases in hospital participation. Decision-makers in public health practice and policy may want to use the findings from this analysis as a way to bench- mark areas that require additional attention to foster hospital engagement while also high- lighting those that already appear to be suc- cesses. Future trends under the Affordable Care Act will likely only call for increased integration of traditional health care and public health systems; this study lays the groundwork to track how those policy interventions affect the joint production of public health activities and hospital participation. j

About the Authors Rachel A. Hogg is with the School of Public Affairs, University of Colorado, Denver. Glen P. Mays and Cezar B. Mamaril are with the Department of Health Management and Policy, University of Kentucky College of Public Health, Lexington.

RESEARCH AND PRACTICE

August 2015, Vol 105, No. 8 | American Journal of Public Health Hogg et al. | Peer Reviewed | Research and Practice | 1651

Correspondence should be sent to Rachel A. Hogg, University of Colorado, SPA, Campus Box 142, PO Box 173364, Denver, CO 80217 (e-mail: rachel.hogg@ ucdenver.edu). Reprints can be ordered at http://www.ajph. org by clicking the “Reprints” link. This article was accepted January 7, 2015.

Contributors R. A. Hogg designed the study, oversaw data collection, conducted the data analysis, and drafted the article. G. P. Mays assisted with data analysis and article revisions. C. B. Mamaril assisted with data cleaning and the analysis.

Acknowledgments This study was supported by grants from the Robert Wood Johnson Foundation (71147 and 70363).

Note. The content is solely the responsibility of the authors and does not necessarily represent the official views of the Robert Wood Johnson Foundation.

Human Participant Protection This study was approved by the institutional review board at the University of Kentucky.

References 1. Institute of Medicine 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.

2. Mays GP, Scutchfield FD, Bhandari MW, Smith SA. Understanding the organization of public health delivery systems: an empirical typology. Milbank Q. 2010;88 (1):81---111.

3. Mays GP, Halverson PK, Stevens R. The contribu- tions of managed care plans to public health practice: evidence from the nation’s largest local health depart- ments. Public Health Rep. 2001;116(suppl 1):50---67.

4. Halverson PK, Mays GP, Kaluzny AD, Richards TB. Not-so-strange bedfellows: models of interaction between managed care plans and public health agencies. Milbank Q. 1997;75(1):113---138.

5. Roussos ST, Fawcett SB. A review of collaborative partnerships as a strategy for improving community health. Annu Rev Public Health. 2000;21:369---402.

6. Association of State and Territorial Health Officials. Profile of State Public Health. Vol 2. Washington, DC: Association of State and Territorial Health Officials; 2011.

7. National Association of County and City Health Officials. 2010 National Profile of Local Health Depart- ments. Washington, DC: National Association of County and City Health Officials; 2011.

8. Stoto MA. Population Health in the Affordable Care Act Era. Washington, DC: AcademyHealth; 2013.

9. Institute of Medicine Committee on Integrating Primary Care and Public Health. Primary Care and Public Health: Exploring Integration to Improve Population Health. Washington, DC: National Academies Press; 2012.

10. Patient Protection and Affordable Care Act Pub. Law No. 111---148, 42 U.S.C. §§; 18001---18121.

11. Duggan M. Hospital market structure and the behavior of not-for-profit hospitals. Rand J Econ. 2002;33(3):433---446.

12. Singh SR. Not-for-profit hospitals’ provision of community benefit: is there a trade-off between charity

care and other benefits provided to the community? J Health Care Finance. 2013;39(3):42---52.

13. Norton EC, Staiger DO. How hospital ownership affects access to care for the uninsured. Rand J Econ. 1994;25(1):171---185.

14. Horwitz JR. Does nonprofit ownership matter? Yale J Regul. 2007;24(1):139---204.

15. Horwitz J, Nichols A. What Do Nonprofits Maximize? Nonprofit Hospital Service Provision and Market Owner- ship Mix. Cambridge, MA: NBER; 2007.

16. Lee SY, Chen WL, Weiner BJ. Communities and hospitals: social capital, community accountability, and service provision in US community hospitals. Health Serv Res. 2004;39(5):1487---1508.

17. Mays GP, Scutchfield FD. Improving public health system performance through multiorganizational part- nerships. Prev Chronic Dis. 2010;7(6):A116.

18. Mays GP, Halverson PK, Baker EL, Stevens R, Vann JJ. Availability and perceived effectiveness of public health activities in the nation’s most populous commu- nities. Am J Public Health. 2004;94(6):1019---1026.

19. Ingram RC, Scutchfield FD, Mays GP, Bhandari MW. The economic, institutional, and political determinants of public health delivery system structures. Public Health Rep. 2012;127(2):208---215.

20. Valente TW. Social Networks and Health: Models, Methods, and Applications. Oxford, UK: Oxford University Press; 2010.

21. National Association of County and City Health Officials. 2005 National Profile of Local Health Depart- ments. Washington, DC: National Association of County and City Health Officials; 2006.

22. Miller CA, Richards TB, Davis SM, et al. Validation of a screening survey to assess local public health performance. J Public Health Manag Pract. 1995;1(1): 63---71.

23. Turnock BJ, Handler AS. From measuring to improving public health practice. Annu Rev Public Health. 1997;18:261---282.

24. Kennedy P. A Guide to Econometrics. 5th ed. Cam- bridge, MA: MIT Press; 2003.

25. Story M, Nanney MS, Schwartz MB. Schools and obesity prevention: creating school environments and policies to promote healthy eating and physical activity. Milbank Q. 2009;87(1):71---100.

26. Coleman KJ, Geller KS, Rosenkranz RR, Dzewaltowski DA. Physical activity and healthy eating in the after-school environment. J Sch Health. 2008; 78(12):633---640.

RESEARCH AND PRACTICE

1652 | Research and Practice | Peer Reviewed | Hogg et al. American Journal of Public Health | August 2015, Vol 105, No. 8

Reproduced with permission of the copyright owner. Further reproduction prohibited without permission.