Skip to main content
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2023 Feb 1.
Published in final edited form as: Health Aff (Millwood). 2021 Mar 25;40(4):664–671. doi: 10.1377/hlthaff.2020.01084

US Public Health Neglected: Flat Or Declining Spending Left States Ill Equipped To Respond To COVID-19

Y Natalia Alfonso 1, Jonathon P Leider 2, Beth Resnick 3, J Mac McCullough 4, David Bishai 5
PMCID: PMC9890672  NIHMSID: NIHMS1863668  PMID: 33764801

Abstract

The COVID-19 pandemic has prompted concern about the integrity of the US public health infrastructure. Federal, state, and local governments spend $93 billion annually on public health in the US, but most of this spending is at the state level. Thus, shoring up gaps in public health preparedness and response requires an understanding of state spending. We present state spending trends in eight categories of public health activity from 2008 through 2018. We obtained data from the Census Bureau for all states except California and coded the data by public health category. Although overall national health expenditures grew by 4.3 percent in this period, state governmental public health spending saw no statistically significant growth between 2008 and 2018 except in injury prevention. Moreover, state spending levels on public health were not restored after cuts experienced during the Great Recession, leaving states ill equipped to respond to COVID-19 and other emerging health needs.


In the United States, the COVID-19 pandemic has increased public scrutiny of the public health system’s ability to protect health and respond to emergencies. Differences in the speed and effectiveness at which disease transmission was slowed or halted shed light on differences in the system’s capability to adapt to emerging needs quickly and effectively. Nations with strong public health systems, such as South Korea, Singapore, and New Zealand, were able to rapidly contain and control the spread of the epidemic.13

In the US, in contrast to well-functioning, co-ordinated national public health systems, gaps in federal, state, and local public health capabilities have contributed to dramatic cross-state differences in the timing of COVID-19 control measures, testing rates, contact tracing capabilities, and health system readiness. These gaps can be partially explained by limited spending on public health in the US, which is 2.5 cents (in 2018 US dollars) per every dollar spent on health, according to official estimates.4

The COVID-19 pandemic’s large-scale impact across the nation revealed the vulnerability of many state and local public health systems. However, even in the decade before 2020, the US saw rising rates of preventable deaths. Recent estimates suggest that alcohol and opioid use have slowed gains in life expectancy during the past decade,5,6 and behavior-mediated health epidemics, similar to COVID-19, has demonstrated dramatic cross-state differentials.6 For example, 2018 drug overdose death rates ranged from 6.9 per 100,000 in South Dakota to 51.5 per 100,000 in West Virginia.7

Disparities in public health systems and funding across states and localities contribute to some of the variation in health outcomes observed from state to state.8,9 It is well documented that effective public health interventions can significantly improve population health and are cost-effective.811 The Institute of Medicine’s 2012 public health finance report also asserted the critical role of governmental public health services and financing in ensuring population health.12 A poorly funded US public health system limits the ability of public health agencies to coordinate with each other and to mount the multisector responses necessary to stay ahead of complex public health threats.11 To help state and local public health agencies be adequately prepared and ready to protect and promote the health of their residents, several national initiatives have been undertaken. The Public Health National Center for Innovations, in concert with private philanthropic organizations and the Centers for Disease Control and Prevention, has delineated a minimum package of public health services called the Foundational Public Health Services model. More recently, it refined the list of essential public health services, offering a blueprint of public health services for state and local health departments to protect and promote the health of residents.13 Furthermore, state and local health departments can choose to seek voluntary accreditation by the Public Health Accreditation Board to identify and address gaps in their provision of essential public health services.14 However, accreditation remains voluntary. At this time, US state and local health departments need not carry out all of the essential public health services—or carry them out well—to continue to receive funding to stay in operation.

As stipulated in the US Constitution, state and local health departments have the primary responsibility for ensuring the public’s health and have historically been the primary source of funding for public health activities. However, after state and local funding cuts in the Great Recession and with the creation of the national Prevention and Public Health Fund (PPHF) in the Affordable Care Act (ACA), the federal government’s share of public health funding for state and local jurisdictions has grown.15,16 This study focused on trends in state-level spending on public health as a potential driver of changes to funding the state and local governmental public health enterprise. We assessed total spending and spending for specific public health categories at the state level from 2008 to 2018, capturing state spending trends both during and after the Great Recession.

Study Data And Methods

DATA

We obtained state government public health expenditures from the Census Bureau’s State Finance Division files.17 These data are reported by state budget offices to the Census Bureau. Census officials then code the state data using government organization function codes and the associated financial expenditure codes.18 These activity-coded census data offer significant advantages compared with other sources of public health financial data, such as data from the Association of State and Territorial Health Officials or from the Centers for Medicare and Medicaid Services (CMS).16,19 Census records tabulate all spending from all state government agencies, not only state health departments. Furthermore, the census data delineate the activities and objectives for which the funds were used. The primary challenge of using census data is the need to harmonize definitions of which activities are expenditures for medical care services for individuals versus public health activities aimed at populations.

We addressed these challenges by categorizing public health spending at a granular level of detail. Because the census codes do not differentiate public health spending from other non-hospital health spending, were coded the 2.7 million expenditure records originally categorized as “Health-Other” (census financial expenditure code 32 for “non-hospital health spending”). Data reduction was achieved by collapsing the 2.7 million individual object-level expenditures into 83,000 program-level records, which we manually recoded. We applied a Uniform Chart of Accounts crosswalk—a set of expenditure coding guidelines20—to each program-level expenditure to differentiate state spending for core public health functions from payments for certain personal health (medical care) services or Medicaid. Disagreements were resolved through iteration and consensus by the full project team.17,21 For details on the categorization of spending, see online appendix exhibit A1.22

Records excluded spending allocations made by federal agencies that were deploying their public health employees at the state level, as well as public health spending by county governments. However, state public health expenditures that were supported by federal grant dollars, as well as state-to-county intragovernmental financial transfers, were captured. Data included coding on the type of program and service category using the Foundational Public Health Services conceptual model, which describes the capacities and programs that state and local health departments should be able to provide to all communities and for which costs can be estimated.17,23,24 Methods and approaches for data categorization are detailed at length elsewhere.17,21

We evaluated complete data for forty-nine of the fifty states, excluding California (which does not report these data to the Census Bureau), spanning 2008–18. We examined public health spending in aggregate and for eight key public health activity categories.20 Expenditure data were inflation adjusted to 2018 US dollars using the Bureau of Economic Analysis’s annual state and local implicit price deflator. We produced annual per capita estimates of public health expenditures dividing each state’s total expenditure each year by its population for that year. Annual population estimates were extracted from the Census Bureau’s Population Division data sets.25,26

TREND ANALYSIS

We ran bivariate regressions of spending per capita on both linear and quadratic time trends to evaluate whether spending systematically increased or decreased over time. This analysis was repeated separately for 2008–09 and 2010–18, after the Great Recession.

We also evaluated spending trends by state for each category and subcategory of public health spending. We defined spending as having a downward (or upward) trend if the three-year smoothed value in 2016–18 was lower (or higher) than the 2008–10 smoothed value and the coefficient on the linear trend was statistically significant. Expenditures with no significant spending change across eleven years or no significant coefficient on the time trend variable were identified as a “flat trend.” We used Stata/SE, version 14, to run regressions. Methodological details are in the appendix.22 We counted the number of states with each spending pattern—up, down, or flat—by spending category and subcategory for each state.

LIMITATIONS

The Census Bureau’s state expenditure data provide detailed and comprehensive insight into how states spent public health funds over time, but limitations are present. The heterogeneity in submission formats from states and interstate differences in the granularity of reporting required the team to use judgment to harmonize coding. More precise estimates occur for states that report more detail, and less precise estimates occur for states that report less detail. The potential for data measurement errors introduced through inappropriate coding was moderated by having teams of two to three coders examine records. Our comparison of intrastate spending trends also attenuated risk for potential bias due to interstate reporting variations. Our period of analysis ran through 2018 (latest year with available data), so public health funding allocations for the COVID-19 pandemic response in 2020 were not yet available for inclusion in this analysis. Finally, data about what was spent on public health activities say nothing about the quality of execution of the essential public health functions. This article is only presenting data on public health spending, not examining the population health outcomes more broadly. However, it is worth noting that the Period we studied coincides with three years during which US life expectancy fell (2014–17).27

Study Results

In 2008 mean and median per capita population-weighted state government spending for public health was $80.40 and $62.37, respectively. By 2018 those figures had decreased to $75.83 and $54.28. In the eleven years of state governmental spending reviewed across all forty-nine states in the analysis, there was no significant growth in states’ average per capita spending on public health after accounting for inflation (exhibit 1). Flat or downward trends were observed for overall (total) state spending and for spending in each of the categories of public health activities, except for a statistically significant increase in spending for injury prevention from $0.24 to $1.89 per capita (p < 0:05). Moreover, two categories saw statistically significant (p < 0:05) spending decreases during this period, including maternal, child, and family health and environmental public health (exhibit 2). Exhibit 2 lists states’ average spending per capita in each category. The highest spending category was maternal, child, and family health, at $30.01 in 2008 and $23.64 in 2018. The maternal, child, and family health category also showed the largest drop in spending (more than $6 per capita, on average) over the course of the eleven-year period. Other categories, such as organizational capabilities, communicable disease control, and other public health, had a slight but not statistically significant increase in average per capita spending between 2008 and 2018 (all p > 0:05). Thus, the overall trend in spending in these three categories can be considered flat.

EXHIBIT 1. State governmental spending trends in public health activities, 2008–18.

EXHIBIT 1

SOURCE Authors’ analysis of data from the Census Bureau, Census of Governments, 2018. NOTE Per capita expenditure values are real (inflation-adjusted) 2018 US dollars.

Exhibit 2.

State public health expenditures per capita and spending trends, by category, 2008 and 2018

Mean expenditures ($)
Public health categories 2008 2018 p value
Maternal, child, and family health 30.01 23.64 0.033
Capabilitiesa 15.66 18.39 0.623
Environmental public healthb 12.05 7.71 0.032
Communicable disease control 8.73 9.32 0.076
Chronic disease prevention 5.03 4.42 0.416
Access and linkage to clinical care 2.32 1.97 0.985
Injury prevention 0.24 1.89 0.001
Other public health 6.36 8.49 0.246
Other health (not public health)c 112.62 105.88 0.190
Other social (not public health) 1.89 2.17 0.038
Other (not public health) 19.12 11.76 0.003

SOURCE Authors’ analysis of data from the Census Bureau, Census of Governments, 2018. NOTES Per capita expenditure values are real (inflation-adjusted) 2018 US dollars. Negative values may be observed. These values likely reflect reconciliation efforts or intra-agency transfers. These negative values are included in the census data set (and thus also are included in the final estimates put out by the Census Bureau). Each trend analysis included 539 observations (eleven years times forty-nine states).

a

Includes policy development, assessment, community partnership, organizational competencies, all hazards, and communications.

b

Includes prevention activities such as permitting, education, and regulation.

c

Categories designated as “not public health” are other health services and include clinical care, environmental protection (remediation and environmental quality), behavioral health, disability-related services, and other services.

The final three categories listed in exhibit 2 are “other health services” categories not considered public health activities. These categories include services such as clinical care, behavioral health, and environmental protection. On average, the total spending in these non–public health categories was higher than in the public health categories, but spending trends in these categories were also flat or downward during the eleven-year period, except for a statistically significant increase in spending for other social (not public health) from $1.89 to $2.17 per capita (p < 0:05).

More detailed analysis stratified by recession period showed that during two years of the recession (2008–09), spending was flat (p > 0:05) for all categories, and the subsequent trend over the nine years after the recession showed no signs of an uptrend or recovery except for a statistically significant increase in spending for injury prevention (p > 0:05) (appendix exhibit A2).22

Analysis of spending trends for individual states across categories and subcategories showed that spending was generally flat (exhibit 3). Overall, thirty-eight states had flat trends for chronic disease prevention. Other public health activity categories showed substantial cross-state variation in spending (see appendix exhibit A3).22 Downward trends were observed less commonly, depending on the category. Maternal, child, and family health and communicable diseases had the most states showing downward trends. Upward trends were observed for a few states for a few categories (for example, twelve states increased spending on communicable disease control, and twelve did so for injury prevention).

Exhibit 3. State spending trends, by public health category, 2008–18.

Exhibit 3

SOURCE Authors’ analysis of data from the Census Bureau, Census of Governments, 2018. NOTES Negative values may be observed. These values likely reflect reconciliation efforts or intra-agency transfers. These negative values are included in the census data set (and thus also are included in the final estimates put out by the Census Bureau). Each category includes 539 observations (eleven years times forty-nine states).

Appendix exhibits A4a and A4b provide additional details on spending patterns by state to highlight which states performed better in terms of spending levels and trends.22 Only five states had upward trends in total public health spending: Florida, Ohio, Rhode Island, South Carolina, and Vermont. However, only Rhode Island and Vermont had per capita spending in the top twenty-fifth percentile. Delaware had the highest spending per capita in 2018, at $271.72, and it was the only state with at least five upward trends out of the eight categories of public health spending, but its total public health spending trend was flat.

Last, of forty-eight subcategories of public health spending (see appendix exhibit A5),22 spending was flat in more than half of the states for all subcategories, except one (organizational capabilities). Spending levels were low and trends were flat for emergency preparedness and areas associated with deaths of despair and stagnated mortality rates (for example, substance abuse, tobacco, obesity, and cardiovascular disease prevention; appendix exhibits A4a, A6, and A7).22 The sum of all state hazards preparedness and response spending for forty-nine states (excluding California) was $388 million in 2018, or $1.81 per capita (data not shown).

Discussion

States play a critical role in ensuring the public’s health at all times, including before, during, and after pandemics and other emergency response efforts. Public health extends beyond emergency response to encompass a wide array of ongoing preventive or protective activities. In addition to improving health outcomes, evidence shows that effective public health departments may lower Medicare and Medicaid spending.28,29 Our study leveraged a novel data source to analyze state-level trends in spending for public health activities since the Great Recession. We found that most states saw spending for public health remain flat (statistically unchanged) or decline in the past decade, after inflation and population changes were accounted for. This stagnation in public health spending occurred despite a 4.3 percent annual average rise in national health care expenditures for disease care and overall annual average economic growth of 3.3 percent during the same period.16

The public health spending trends observed in this study (that is, a decrease from $80.40 per capita in 2008 to $75.83 in 2018) are consistent with estimates produced by the Association of State and Territorial Health Officials. This association estimated that median annual per capita spending across states remained largely unchanged at $80.5 per capita in 2010 and $78.8 in 2018.30 In contrast, the CMS National Health Expenditure Accounts estimates for state and local public health activity increased from $61,680 ($203 per capita) in 2008 to $81,515 ($249 per capita) in 2018.16 Spending levels between these studies differ for reasons detailed elsewhere, but primarily because our estimates separate out non–public health activity categories currently captured in the Government Public Health Activity estimate in the National Health Expenditure Accounts.16 Our estimates focus on public health spending by nonhealth government agencies (for example, departments of environment and agriculture).21 We have argued elsewhere that CMS’s approach and official estimates substantially overinflate actual spending on governmental public health.4

Flat public health spending coincided with observed declines in life expectancy, pervasive health disparities, and rising mortality rates, especially for White Americans ages 45–65.6,3133 Public health spending remained flat despite the Trump administration’s declaration in October 2017 that the opioid crisis was a public health emergency.34 Similarly, total public health spending showed no response to the decade’s major public health events such as the emergence or reemergence of Ebola, Zika virus, West Nile virus, Middle East Respiratory Syndrome, measles, and other communicable diseases. We observed increases of categorical money directed at specific response activities during these events but no systematic increase in total public health system spending. Although these communicable disease events may have foreshadowed a larger outbreak to come, we found no evidence that states significantly increased their spending to bolster preparedness or response capacities. Our analyses found that since 2011 states have spent roughly $1.90 per capita per year on hazard preparedness, whereas the per capita cost of the COVID-19 pandemic is estimated at $50,000 per capita.35

This analysis has focused on state-level public health spending because states accounted for the vast majority of the nation’s $93 billion in public health spending in 2018.16 The radically different funding levels for various public health activities by each state do not appear to be attuned to epidemiological developments occurring during the period observed. Public health is tasked with assessing disease threats and organizing responses to those threats. For these responses to succeed, a funding stream should reasonably be expected to align with the assessed threat. Our observation of flat spending in the face of rising mortality (and rising economic resources) shows that funding is not being calibrated to need. Furthermore, since stagnant public health spending was seen in almost all states during the period studied, this indicates widespread neglect across the spectrum rather than neglect associated with a particular political ideology or regional geography. Public health departments seldom develop politically supportive constituencies analogous to those who lobby for other governmental services, such as police, roads, schools, or housing.36

In a federal system, the government has the ability to offer technical support and assistance to states in detecting and responding to emerging problems. In this regard, the federal government’s Prevention and Public Health Fund, created as part of the ACA, was conceived as a vehicle to shore up gaps in the states’ responses to public health threats. However, PPHF funds were repeatedly reallocated away from their intended purpose from 2012 to 2016.37 For example, in fiscal year 2013 the Obama administration reallocated almost half of PPHF funds to support outreach to boost enrollment in ACA insurance plans.30 Subsequently, the Trump administration also reallocated PPHF spending away from public health emergency response to pay for health care delivery, and Congress launched more than fifty attempts to repeal the PPHF (with no replacement) as part of efforts to dismantle the ACA.37

There are likely to be immediate short-term increases in public health spending to support COVID-19 responses by state and federal government. The pattern of unresponsive public health spending in the face of public health threats that we have documented in states over the past decade serves as a benchmark for the “old normal.” Without institutional reform, states are susceptible to continued neglect of public health with only short-term emergency infusions of federal funding in the midst of a crisis such as COVID-19. Without substantial and sustained investment by states and ongoing robust federal support, the US may well continue its “default” approach to public health funding: “neglect, panic, repeat.”38

Conclusion

To achieve and maintain adequate public health protections and improve the nation’s health over the long term, sustainable funding and policy mechanisms to safeguard both state and federal investments in public health are needed.39 The US needs to ensure that the spending priorities and infrastructure of public health agencies are aligned with current health problems and emerging threats. The National Academy of Medicine (formerly the Institute of Medicine), among others, has offered key recommendations related to administrative changes and financial reforms needed to facilitate rational and efficient use of governmental public health funds.12,40 To cope with dwindling or stagnant core support from state budgets, public health departments have become compartmentalized and fragmented in their pursuit of project- and grant-based funding. As a result, staff are fragmented and allocated to line items for a particular disease (for example, HIV) or special population. Federally sourced, ring-fenced funding for public health infrastructure at the local level that cannot not be reallocated away from stipulated public health activities, as the UK did when reforming its public health system,41 would not only strengthen the US public health system’s ability to respond to new threats but also enable better tracking of funding and leverage more health departments toward accreditation. Ultimately, the financial foundation for a better public health system rests on understanding how badly state and local health departments have been neglected and how important they are for the nation’s collective health and safety now and well into the future.

Supplementary Material

appendix

Acknowledgments

This study was funded by the Robert Wood Johnson Foundation through its Systems for Action Research (Grant No. 78116). The authors thank the Census Bureau for provision of data and Rebecca Reif for assistance with data cleaning.

Contributor Information

Y. Natalia Alfonso, Department of International Health (Health Systems Program), Johns Hopkins Bloomberg School of Public Health, in Baltimore, Maryland.

Jonathon P. Leider, Division of Health Policy and Management, University of Minnesota School of Public Health, in Minneapolis, Minnesota.

Beth Resnick, Department of Health Policy and Management, Johns Hopkins Bloomberg School of Public Health.

J. Mac McCullough, College of Health Solutions, Arizona State University, in Phoenix, Arizona.

David Bishai, Department of Population, Family, and Reproductive Health, Johns Hopkins Bloomberg School of Public Health.

NOTES

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

appendix

RESOURCES