Enumerating the State and Local Public Health Workforce During the COVID-19 Response

Abstract

Objectives. To understand the landscape of the nonfederal governmental public health workforce and to identify replicable methods for future enumerations.

Methods. This enumeration of the state and local public health workforce was conducted from 2023 to 2024 and triangulated the National Association of County and City Health Officials (NACCHO) Profile 2022 and the Association of State and Territorial Health Officials (ASTHO) Profile 2022. We utilized Public Health Workforce Interests and Needs Survey (PH WINS) data from 2021 to assess demographic distributions across Department of Health and Human Services (HHS) regions in the United States.

Results. A total of 239 000 staff were employed in state and local health departments in 2022, a 2% increase since 2012. Sixteen states—including 6 in the Southeast—lost staff relative to population growth.

Conclusions. An uneven landscape of public health workforce density reflects chronic underinvestment in public health. The process of enumeration itself was also fraught with pitfalls and data limitations.

Public Health Implications. We recommend building on federal investments to develop dedicated funding streams for state and local public health. We also recommend amending federal efforts around enumeration to include governmental public health at all levels. (Am J Public Health. 2025;115(5):716–725. https://doi.org/10.2105/AJPH.2024.307964)

An accurate count of the public health workforce is essential for understanding the capacity of the governmental public health system, workforce trends, and policies needed to support public health infrastructure. However, governmental public health is not regularly enumerated by federal statistical agencies, with such cited reasons as lack of definition sets to capture all public health occupations and difficulty identifying public health agencies based on federal industry classifications.¹ Therefore, deriving total counts of employees is left to the field and occurs every 10 to 20 years. The infrequency of these cycles has proved challenging to a field facing transition and crises; in the decade after the Great Recession, state and local governmental public health is thought to have lost 15% to 20% of the workforce, which likely did not return by the onset of COVID-19.² During the COVID-19 pandemic, there was evidence of temporary increases in the workforce, though with substantial turnover.³ Yet, without official estimates from the federal government or formal enumeration, this growth was not well-characterized.

Since 1964, the public health workforce has been counted via synthetic enumeration, which utilizes survey data of hundreds of agencies from multiple data sources to produce informed estimates.⁴ Merrill et al. outlined the history of public health enumerations from 1923 through 2000, describing 19 unique enumerations during this period.⁴ They described an approximately 10% reduction in the total governmental public health workforce (federal, state, and local) between 1980 and 2000,⁴ from approximately 500 000⁵ (representing 220 staff per 100 000 population) to roughly 448 254⁶ (158 per 100 000), which, at the time, included many community health care workers in addition to other public health department employees. Merrill et al. discussed how the decrease might have been attributable to definitional changes in the public health worker taxonomy as well as actual decreases in the number of public health staff.⁴

In 2012, Beck et al. completed another synthetic enumeration of the public health workforce, which counted approximately 290 988 total public health workers (federal, state, and local), representing a further 35% reduction from 2000 levels.⁷ Although changes in the definition of a public health worker partly explain this decrease, a real reduction in workforce numbers also contributed.^4,6 Beck et al. estimated the 2012 state and local workforce at 233 882 employees, or about 80% of the total workforce. They also emphasized several recommendations for future enumerations, such as adopting a “standardized methodology for continuously monitoring the size and composition of the public health workforce.”^7(pS313)

Other public health workforce researchers have provided recommendations to improve future enumerations. Tilson and Gebbie noted that defining the scope of work in public health was essential in conducting public health workforce research.⁸ Gebbie et al. also cautioned against “quick fixes” to such research, which included relying on federal Bureau of Labor Statistics (BLS) data that are limited in their ability to properly characterize multidisciplinary fields like public health.¹ For example, historically, Standard Occupational Classification codes used by the BLS do not disaggregate the number of public health nurses from the group of registered nurses.¹ In 2023, Leider et al. called for an “overarching definition for workers across the collective public health workforce and a consensus-driven taxonomy of relevant occupations and disciplines.”^3(p333)

These recommendations provide a roadmap for navigating the challenges of public health enumeration. Building upon the foundations of prior enumerations, we updated the workforce enumeration (“Enumeration 2024”) with a more comprehensive analysis of the number and distribution of public health workers a decade after the last enumeration of Beck et al.,⁷ providing a methodology and baseline of comparison for future enumerations. This enumeration differs from previous workforce counts by including additional sources for triangulation, including Public Health Workforce Interests and Needs Survey (PH WINS) administrative fielding data and member reports from State Associations of County and City Health Officials (SACCHOs). We also used algorithmic missing data approaches to create more comprehensive estimates and credible intervals. In a parallel article in this issue (Leider et al., p. 707), we provide a full review of the enumeration of the public health workforce in 2022, including total counts, occupational profiles, and public health nursing workforce.⁹ The current article outlines methods for and findings from the local and state total counts of Enumeration 2024.

METHODS

We estimated employee counts, including vacant positions, at the state and local levels using a synthetic enumeration approach. The synthetic enumeration of the state and local public health workforce was conducted by triangulating (cross-verifying data across disparate data sets¹⁰) two 2022 data sets: the National Association of County and City Health Officials (NACCHO) Profile 2022,¹¹ providing workforce information at the local agency level, and the Association of State and Territorial Health Officials (ASTHO) Profile 2022,¹² providing workforce information at the state agency level. Although every state provided employee estimates for state agencies via the ASTHO Profile, employee counts were only available for 37% of local agencies from the NACCHO Profile. To partially address the issue of missing data for the local public health workforce, we retrieved additional employee data for 2022 from 18 SACCHOs (e-mail communications, February–April 2024) and a small fielding of PH WINS¹³ in 2022 that identified agency sizes. We interpolated these additional data with the NACCHO Profile 2022 responses, increasing the effective response rate for local agencies to 42%. For the remaining local agencies where employee and full-time equivalent (FTE) counts for 2022 were unavailable, we conducted multiple imputation by chained equations^14,15 using historical data as predictors, including the NACCHO Profile (employees and FTEs) in 2016 and 2019,^11,16,17 jurisdictional population within each local health agency in 2016, 2019, 2020, and 2022,^11,16,17 SACCHO reports (employees and FTEs) in 2021 (e-mail communications, February–April 2024), state-level governance information, and the PH WINS fielding data in 2017 and 2021.^13,18 We used predictive mean matching to impute all variables with missingness, and the algorithm conducted 20 imputations with 100 iterations for 2 imputation approaches representing 2 possible sequences for imputation.⁹ We determined the number of imputations and iterations through a sensitivity analysis (see Methods Appendix, available as a supplement to the online version of this article at http://www.ajph.org), increasing these values until no significant improvement in convergence was observed. The resulting 40 sets of local workforce estimates (employee and FTE counts) were bootstrapped 1000 times, aggregated to the state sum, and averaged to serve as the “local imputed” estimate for each state.¹⁹

We then combined local imputed estimates for each state from the NACCHO Profile with state estimates from the ASTHO Profile. The ASTHO Profile had 3 data points of interest: state health agency central office counts, reported regional office counts, and reported local counts. To triangulate these data points into 1 state and local workforce estimate for each state, we analyzed each state’s governance²⁰ and utilized 1 of 3 constructions to estimate total state and local staffing (see Methods Appendix for each state’s construction). We aggregated state-level estimates into Department of Health and Human Services (HHS) region estimates where relevant. After developing state- and HHS region–wide estimates of total employee counts, we modeled 2 sources of uncertainty—that because of missingness and potential reporting error—and used them to develop reasonable 95% credible intervals (see Methods Appendix for more detail). For the following analysis, we allowed for a reporting error of up to 20%—the most conservative of our a priori identified reasonable error margins—as there were no empirical estimates of reporting uncertainty available. The sensitivity analysis in the Methods Appendix displays how different assumed uncertainty levels affected the 95% credible intervals.

To calculate the change in the public health workforce between 2019 and 2022, we used the same constructions of total state and local staffing using ASTHO Profile 2019 data and imputed local estimates for 2019 from the imputation process described in this section. We utilized PH WINS data from 2021¹⁴ to estimate the distribution of demographic characteristics across HHS regions.

RESULTS

The total estimation of state and local governmental public health staff in 2022 was approximately 239 000 (95% credible interval = 227 963, 251 840), of which we estimated that 70% were embedded in local health departments. Staff were not evenly distributed across regions, as shown in Figure 1, which also shows that areas with fewer absolute staff tended to have more staff per capita. Regions 1, 7, 8, and 10 (which represent New England; IA, KS, MO, and NE; the Mountain West; and AK, ID, OR, and WA, respectively) had fewer than 15 000 total staff, but all had more than 84 staff per 100 000 population—that is, they had the smallest absolute workforces, but among the highest per capita workforces. This trend was observed in reverse as well: regions 4, 5, and 9, which comprise the Southeast (KY southeast to FL), the Midwest (MN east to OH), and the Southwest (CA, HI, NV, and AZ), all had large numbers of staff (especially region 4, which had almost 45 000 workers) but fewer than 72 workers per 100 000 population. However, region 6 (the South from LA to NM) had the fewest workers per 100 000: fewer than 60.

FIGURE 1—

Total State and Local Staff as Well as Staff per 100 000 Population by HHS Region: United States, 2022

Note. Only data from states and the District of Columbia are represented in the figure. Regions are defined by the Department of Health and Human Services (HHS).

Sixty-two percent of states had a total workforce per 100 000 people served of between 50 and 100 (Figure 2). Although most states improved their staff-to-population ratio from 2019 to 2022, 16 states, mostly clustered in the South, lost staff relative to their population growth. Indiana and New Hampshire were unique in having lost staff relative to population between 2019 and 2022 and, in 2022, having fewer than 50 public health workers per 100 000, whereas Utah, Washington, and the District of Columbia all gained more than 25 staff per 100 000 in the prior 3 years and by 2022 had a density of more than 100 public health workers per 100 000.

FIGURE 2—

Public Health Workforce per 100 000 in 2022 and Change From 2019 Relative to Population Growth: United States

Note. No information regarding change in workforce since 2019 was available for New Mexico. Hawaii lost staff relative to population growth. The District of Columbia gained more than 25 per 100 000 population between 2019 and 2022 and had more than 100 staff per 100 000 population in 2022.

The racial distribution of the workforce—broken out by HHS region in Table 1—was mostly consistent with each region’s respective geography, with all regions except for regions 6 and 9 having majority-White workforces. Region 9, which comprised California, Hawaii, Nevada, and Arizona, had the largest proportions of Native Hawaiian/Other Pacific Islander, Asian, and Hispanic or Latino staff. Region 6 (the South from LA to NM) also had a large proportion of Hispanic or Latino staff. The largest proportion of Black or African American public health staff was in region 3 (the mid-Atlantic from WV to DE) and region 4 (the Southeast from KY to FL), consistent with the larger proportional population of Black or African American populations in the Mid-Atlantic and Southeast.²¹ Regions 5, 7, and 8 (the Midwest, Plains States, and Mountain West) had the highest proportion of White staff. Region 5 also had the lowest proportion of Hispanic or Latino staff and region 8 (the Mountain West) had the fewest Black or African American staff.

TABLE 1—

Distributions of Race/Ethnicity and Age of Public Health Workers by HHS Region: United States, 2021

	Region, %
	1	2	3	4	5	6	7	8	9	10	National, %
Race/ethnicity
American Indian or Alaska Native	0.7	0.9	0.9	0.7	0.6	1.9	1.4	1.1	0.8	1.2	0.9
Native Hawaiian or other Pacific Islander	0.2	0.1	0.4	0.2	0.1	0.2	0.2	0.4	1.2	0.4	7.4
Asian	4.8	11.3	5.6	2.4	3.4	4.0	2.5	2.9	20.2	6.8	15.3
Black or African American	17.3	20.0	25.8	23.3	11.1	18.2	6.2	2.0	7.5	4.6	18.0
Hispanic or Latino	14.4	12.9	10.2	17.1	7.2	30.0	11.0	12.0	33.7	13.6	0.4
White	59.5	50.2	53.1	51.5	74.9	41.2	75.2	78.1	31.0	69.6	53.7
≥ 2 races/ethnicities	3.1	4.7	4.0	4.8	2.8	4.4	3.5	3.6	5.5	3.9	4.3
Age, y
< 31	16.2	10.7	13.6	12.8	15.3	12.0	14.8	14.9	12.3	14.3	13.3
31–40	23.1	24.8	22.4	19.5	25.3	24.3	24.4	29.3	26.2	27.6	24.0
41–50	25.6	24.9	22.5	24.4	24.9	25.6	23.7	27.6	26.9	25.7	25.1
51–60	23.1	27.2	25.6	28.5	22.6	25.5	23.9	19.9	24.1	21.5	25.0
≥ 61	12.0	12.4	15.8	14.8	11.9	12.6	13.2	8.4	10.5	10.9	12.6

Note. HHS = Department of Health and Human Services.

Source. Public Health Workforce Interests and Needs Survey (PH WINS) 2021.¹³

Table 1 shows a minimal variation between HHS regions in the age distribution of the public health workforce. Of note is the fact that between 28.3% (region 8) and 41.4% (region 3) of employees were older than 50 years.

The majority of the public health workforce across all 10 regions attained at least undergraduate degrees, as seen in Table 2. Undergraduate educational attainment (i.e., associate or bachelor’s) ranged from 39.4% in region 1 to 53.2% in region 5. Between 26.4% (region 4) and 50.4% (region 1) held graduate degrees (i.e., master’s or doctoral).

TABLE 2—

Distributions of Degree Attainment and Job Classification of Public Health Workers by HHS Region: United States, 2021

	Region, %
	1	2	3	4	5	6	7	8	9	10	National, %
Highest degree
No college degree	10.2	9.5	14.3	21.6	11.2	19.1	16.9	9.8	13.0	12.6	14.8
Associate	7.5	8.8	11.7	16.7	11.5	12.4	9.4	8.5	9.3	8.7	11.5
Bachelor’s	32.0	34.3	36.7	35.3	41.7	35.6	41.4	41.2	38.2	37.2	37.2
Master’s	43.3	38.1	30.4	22.6	32.0	27.8	28.5	35.7	31.4	34.8	30.6
Doctoral	7.0	9.3	6.9	3.8	3.7	5.1	3.8	4.8	8.1	6.7	5.9
Job classification
Administrative	25.8	28.5	26.1	31.6	22.7	32.1	29.7	25.2	29.0	28.1	28.3
Clinical and lab	18.2	24.1	28.5	28.1	30.3	24.1	19.5	21.5	28.6	19.6	26.2
Public health sciences	54.5	45.2	44.3	36.7	44.4	40.3	44.6	52.6	40.2	51.3	42.9
Social services and all others	1.4	2.1	1.2	3.6	2.6	3.4	6.2	0.7	2.1	1.0	2.6

Note. HHS = Department of Health and Human Services.

Source. Public Health Workforce Interests and Needs Survey (PH WINS) 2021.¹³

Public health job classifications are observed in Table 2 as well. The plurality of staff in all regions worked in public health sciences, although only a small proportion of staff were engaged in social services or other classifications. Compared with other measures, the distribution of job classifications had relatively little variation between regions.

Tables 1 and 2 are also displayed graphically in Figure A of the Appendix (available as a supplement to the online version of this article at http://www.ajph.org).

DISCUSSION

In conducting Enumeration 2024, the first look into the composition of a workforce responding to COVID-19 was made possible. Our estimates show a total of 239 000 public health employees at state and local agencies, a 15.7% increase from 2019 (206 500).³ Our estimates are comparable to those reported in the 2012 enumeration (roughly 234 000), although our approach and that undertaken in 2012 differed substantially, and the 2012 enumeration had not yet fully cataloged cuts associated with the Great Recession that saw the workforce fall to around 200 000.⁷ However, although the growth over the past few years (from 2019 to 2022) is encouraging, overall this decade, the public health workforce increased by only about 2% despite the nation’s population growing from about 309 million in 2010²² to roughly 336 million in 2024 (an increase of almost 9%).²³ Thus, the public health workforce has failed to keep pace with the growth of the US population. This is particularly visible in the Southeast and Mountain West, where many states lost staff relative to population growth between 2019 and 2022.

Ratios of workers to population served varied across HHS regions, from 59 per 100 000 in region 6 to 97 per 100 000 in region 1, and 8 states had fewer than 50 workers per 100 000. The geography of changes in the public health workforce merits further investigation, particularly as it relates to workforce and health policy differences between states. Public health worker demographics show that between 28.3% and 41.4% of employees were older than 50 years. This is notable because of the looming retirements of many in the Baby Boomer generation²⁴ as well as the low rates of recent public health graduates choosing to work in governmental public health agencies.²⁵ Individuals who identified as White constituted the majority of the workforce in all HHS regions except for regions 6 and 9, underscoring the necessity of efforts and funding dedicated to diversifying the workforce to reflect the changing racial makeup of the general population.^26,27 And regional educational attainment disparities may have impacts on the types of services that the public health workforce can provide.

This study affirms a major finding of past enumerations: incomplete data collected at different times by different organizations make it challenging and expensive to develop a clear picture of the workforce. Unlike staff in other fields, industries, and sectors, those in governmental public health are simply not counted by federal statistical agencies. Despite some work to modernize and include more public health occupation codes in federal lists, current federal occupation indices do not adequately capture the workforce²⁸; that task falls to membership organizations, foundations, and academia, and thus is an imperfect solution to answering a critical, basic question: How many people work in public health? Only once we have answered this first question can the next question follow: How many more do we need?

Questions about how to grow the workforce sustainably are manifold in the field. A recent study from the Office of the Assistant Secretary for Planning and Evaluation (ASPE) and Mathematica finds similar challenges in scoping and scaling the workforce.²⁹ They offer a host of recommendations, including building pathways, leveraging nongovernmental partnerships, and explicitly planning surging strategies. Such surging strategies necessitate a larger initial base, or foundation, of the public health workforce in order to be able to surge when necessary.³ One of the major challenges observed during COVID-19 was that response was performed sometimes to the exclusion of standard public health activities, even mandated ones.³⁰ This reality was born of an undersized workforce, though how undersized is difficult to know.³

We echo previous enumeration researchers who called for more funding and training on reporting public health agency workforce numbers. We also advocate for reaching a consensus as a field as to how to define a public health worker, including a potential tiered taxonomy of what is classified as such. This taxonomy must be adopted and utilized across the discipline. Furthermore, we recommend amending federal efforts around occupation- and industry-based enumeration to more explicitly include governmental public health at all levels. Mandating that enumerations be conducted by a federal agency should improve the consistency, coherence, and replicability of estimates over time, as well as provide more regular snapshots of the workforce.

Limitations

Although we were able to take many lessons from scholars who conducted prior enumerations, we were not able to incorporate all recommendations. For example, there remains debate on how to define public health workers, and lack of a universally accepted definition of a “public health worker” may have influenced who was reported, and thus counted, as working in public health. Furthermore, much of the data used in this work were collected during COVID-19. Because health agencies were triaging services at this time, it is possible that data are not as comprehensive or accurate as in other, nonpandemic years.

Our findings describe the public health workforce at the time the data were collected in 2022. This was at the beginning of the rollout for Centers for Disease Control and Prevention (CDC) funding for pandemic response. There is no guarantee that funding will be renewed by the government after it expires in some years. It is also unknown how the expiration of funding will affect workforce numbers into the future, and research endeavors over the next few years should explore this question.

There were a number of assumptions that had to be made to enable the development of the total counts estimates and the credible intervals, including decisions around the imputation algorithm, the triangulation of multiple data sources into statewide estimates, and the estimation of reporting error. Regarding multiple imputation, we must assume that data are missing at random—that is, that any patterns or underlying reasons for missing data are explained by other data elements present in the data set—which is an inherently unverifiable assumption.³¹ To use predictive mean matching in our multiple imputation process, we must assume that the distribution of the missing data are the same as that of the observed data. Additionally, predictive mean matching cannot create estimates beyond the range of the observed data, which may cause bias.³²

Lastly, public health staff serving in Tribes, Territories, and Freely Associated States were not included in this enumeration because of a lack of current and available data accessible to the Enumeration 2024 team. Future enumeration work should seek to incorporate data from these communities to more accurately reflect the state and local public health environment in every part of the United States. This article also did not include analysis of the federal public health workforce (e.g., the Commissioned Corps of the US Public Health Service, the CDC, and other related agencies), although they were counted as part of Enumeration 2024.⁹

Policy Implications

The state and local public health workforce, which provides the majority of public health services in the United States, increased only by about 2% in the last decade (from approximately 234 000⁷ to approximately 239 000), despite the nation growing in population by nearly 9%. Sixteen states, including 6 in the Southeast, lost staff relative to their population growth between 2019 and 2022, and 8 states around the nation have fewer than 50 public health workers per 100 000 population.

This reality reflects the persistent environment of underinvestment and understaffing of state and local governmental public health. Although there have been substantive one-time investments in governmental public health by the federal government since the data collection for this enumeration,³³ we do not anticipate such efforts alone to be enough. Adding to the concern is the fact that the majority of public health workers in all regions are older than 40 years and that the workforce pipeline for governmental public health is inconsistent at best.³⁴ At this critical moment, when the COVID-19 pandemic is fresh in the minds of the citizenry and lawmakers, the nation, states, and localities should pursue dedicated funding streams to reinvest in governmental public health. Additionally, the federal government should invest in long-term efforts to develop a coherent and consistent method for enumerating the public health workforce at regular intervals. This has benefited other fields, and is of substantial importance to understanding the current state of governmental public health needs.¹

This will likely prove to be challenging, however, as the current enumeration was funded by COVID-19 response allocations in the form of the Public Health Infrastructure Grant (PHIG). PHIG is unlikely to be renewed with the ending of the public health emergency.³³ Furthermore, even if funding were consistently allocated to public health enumeration efforts, systemic training in workforce data reporting would be needed across local health departments and state health agencies to ensure accurate counts and comparability of successive enumerations. Ideally, enumeration would be tasked to 1 federal agency—or potentially to the Public Health Accreditation Board as part of the accreditation process or as a component of evaluating agency performance—so that transfer of knowledge could be facilitated. Although challenging, these barriers must be addressed if future attempts at disinvestment of public health are to be anticipated and averted.