State Barriers to Appropriating Public Health Emergency Response Funds During the 2009 H1N1 Response

Abstract

Objectives. We examined state-specific administrative barriers to allocating 2009 H1N1 influenza public health emergency response (PHER) funds.

Methods. We conducted a qualitative review of PHER grants management reports to identify and code barriers reported by states in allocating funds. Using linear regression, we examined the relationship between the percentage of funds allocated and each individual barrier and, separately, the cumulative effect of multiple barriers.

Results. States reported 6 barrier types, including regulatory issues (n = 14, or 28%), contracting issues (n = 14, or 28%), purchasing issues (n = 6, or 12%), legislative issues (n = 5, or 10%), staffing issues (n = 5, or 10%), and issues transferring funds between state and local health departments (n = 4, or 8%). In multivariate models, having experienced a purchasing barrier was associated with a significant decrease in PHER allocation (B = −26.4; P = .018). Separately, the cumulative effect of having 3 barriers was associated with a decrease in PHER allocation (B = −16.0; P = .079).

Conclusions. Purchasing barriers were associated with delayed use of PHER funds. Moreover, the cumulative effect of any 3 barriers hampered the allocation of funds. Understanding barriers to using funds can inform future funding guidance for improved efficiency of response efforts.

In response to the detection of novel influenza A (H1N1) in a number of people in the United States in spring 2009, Congress appropriated the Public Health and Social Services Emergency Fund.¹ This funding supported public health emergency response (PHER) grants, which were awarded to 62 state and local public health agencies. Funds such as these can be established by Congress with the purpose of assisting federal, state, and local public health entities to prepare for and respond to a public health disaster or emergency. In this case, the funds were established in preparation for a potential H1N1 influenza pandemic, including planning for and conducting mass vaccination campaigns. Funds could also be used to hire additional employees as necessary, to assist in purchasing all essential supplies (e.g., vaccine, vaccine administration supplies, storage facilities), or to cover additional administrative costs associated with the response.

As is common in disaster or emergency situations, Congress determines whether PHER funds are necessary for the specific public health disaster or emergency, and funds are distributed to states by the Centers for Disease Control and Prevention (CDC). Specifically, in this response, the CDC’s Division of State and Local Readiness distributed approximately $1.35 billion in PHER funds across the United States. (The CDC provides annual preparedness funding to state health departments through cooperative agreements. The terms of these agreements ensure that funds are used in specific ways and that consistent reports on the use of these funds are provided to the CDC by each state or entity at specified times. In the case of H1N1 response funding, a similar mechanism was used to disburse these PHER funds to the states.) Given the large amount of funds allocated to the states and the urgency with which those funds needed to be spent to protect the population from potential unnecessary illness and death, the capacity of states to efficiently allocate these types of emergency funds is an important issue.

To benefit from the availability of PHER funds, states need to be able to effectively and efficiently use these funds to prepare for or respond to the public health event. During the H1N1 response, some state public health agencies experienced challenges in being able to use the PHER funds in a timely manner or in the ways in which they would have preferred (e.g., hiring temporary employees).² Because of differences in organizational structure, scope of activities, laws, and policies (e.g., protocols for formal declarations of emergencies), different states may experience different types of barriers to spending PHER funds.^2–7 For example, states differ in their policies of when to declare an emergency.⁸ Such emergency declarations have the potential to determine exceptions to routine organizational processes such as protocols for spending approval, protocols for contracting with other agencies (i.e., external to public health or at varying levels within public health such as local health departments), the ability to fund or hire additional employees, or the ability to involve staff members who are not typically focused on emergency preparedness in the response activities.

State public health agencies also vary by structure in terms of the centralization of decision-making as well as location and integration of the public health preparedness division within the organization.^3,7 In general, centralized states are more likely to make financial decisions at the state level than to contract with local public health departments to distribute funds to the local level, essentially allowing local decision-makers to determine how their funding is spent.⁹ One might posit that centralized states may be able to easily move PHER funds from the state health department to local health departments during a response event because they do not have to individually contract with local health departments, which decentralized states typically have to do. In terms of the location of public health preparedness divisions within state health departments, those organizations that place preparedness just below the state health officer may experience emergency response differently (e.g., greater support or involvement from agency leadership) than do states that have the preparedness division farther down the organizational chart. In addition, states that have a relatively large emergency preparedness staff might be able to navigate the internal administrative barriers more efficiently than do those that have to pull in nonemergency preparedness staff to effectively manage the response. Nonemergency staff will ultimately continue having other duties, and in a prolonged response event, this may result in various challenges.

Much of the efforts toward and research on improving public health preparedness, however, have focused on improving collaboration with and between external partners^10,11 and on preparing for specific types of events.^12–15 These efforts have been valuable and crucial to quickly bridging challenging gaps in public health preparedness and response planning, yet understanding the role of internal public health processes in emergency response situations is increasingly important and rarely examined.¹⁶ Experiences during the H1N1 response have the potential to provide insightful illustrations of how internal organizational response processes can affect public health and its response to an emergency or disaster event.^2,17–19

State grant management reports, provided to the CDC during the H1N1 response, indicated that variation existed in how efficiently states were able to allocate PHER funds or move funds to local health departments. Although no prescribed amount of funds was supposed to be spent at the midyear point, some states had allocated as little as 10% of their PHER funds, and other states had allocated 93%. This variation begged the question of whether these differences were related to barriers in the state systems. Thus, we conducted this study to identify barriers to allocating PHER funds for use in state and local health departments. Not only is it important to understand what specific barriers were reported, it is crucial to understand the relationship between those barriers and the ability to allocate funds. Because the timely allocation of emergency response funds is of critical importance to an effective emergency response, findings from this study may inform decisions, procedures, and requirements for accepting and using future influenza or other urgently needed emergency response funds.

METHODS

We used a mixed-methods approach to understanding barriers that states experienced and how those barriers related to the allocation of PHER funds. First, we qualitatively reviewed data provided by the 50 state public health departments to the CDC as part of required reporting on the PHER grant and the Public Health Emergency Preparedness (PHEP) cooperative agreement, the latter of which is the main annual emergency preparedness funding source. Grants management data include PHER midyear funding reports, H1N1 after-action reports, and PHEP 5-year prospective planning reports. These reports were submitted during the H1N1 funding and response period (October 2009–May 2010).

Second, using the information identified in the review of grant data, we used grounded theory to identify various barriers to allocating emergency funds.²⁰ In grounded theory, multiple sources of information are collected and coded on the basis of the information—usually text—and then those codes are grouped together conceptually and ultimately a theory may arise.²⁰

Data Extraction

Two researchers (V. Y. and J. U.) separately coded qualitative information provided in the grants management reports to identify barriers to allocating PHER funds. They initially coded the same 5 states and then met and walked through each barrier they identified in the grants management reports to determine whether both researchers found the same barriers for each state and coded them the same way. Once we agreed on the codes that would be used for the review of all 50 state grant management reports, the 2 researchers separately went through all the reports and indicated their codes in a spreadsheet. Depending on how an issue was discussed in the reports (i.e., more than once, described differently, or both), there were circumstances in which certain state experiences received 2 codes. The text was coded for whichever was most appropriate given the entire context in which it was presented. For each state, we recorded each of the identified barriers and the number of times a specific barrier was noted, developed a summary explanation for each barrier based on the issues identified in the data, and extracted sample quotes to give examples of the barriers identified. Once all states were coded, we met again to review the data extracted by each of the researchers and cross-referenced all data for consistency. We discussed a total of 5 discrepancies in the coded data, ultimately resulting in agreement in 100% of all cases (i.e., 100% interrater reliability).

We identified a total of 6 types of barriers (Table 1). Then, using the identified barriers, we empirically examined the relationship between specific barriers and states’ ability to allocate their funds (measured as the percentage of PHER funds allocated). We also examined whether the cumulative number of barriers experienced by each state was associated with the percentage of funds allocated. Last, we examined whether state characteristics such as size, geographic location, organizational structure (centralization), and number of PHEP staff (the number of staff supported in part or entirely by the PHEP cooperative agreement, referred to as emergency preparedness staff or employees for brevity) were associated with any of the identified barriers.

TABLE 1—

Type and Frequency of Barriers to Allocating State Public Health Emergency Response Funding Identified in Grant Reports of State Health Departments: United States, October 2009–May 2010

Type of Barrier	Definition	Sample Quote	No. (%)
Regulatory issues	PHER funds allocation affected by state-specific rules or guidelines	“The state budget change process is lengthy due to the requirement of notification and approval of several control agencies.”	14 (28)
Contracting issues	Accounting or administrative process barriers or difficulties getting signatures or approvals in a timely manner	“Once the Notice of Grant Award is received Reimbursable Service Agreements must be created (accounting structures). This process can take as little as a week or up to 2 months if ledger accounts are not already created. Further, administrative signatures can be the most significant barrier in this process.”	14 (28)
Purchasing issues	Process barrier related to purchasing	“For any capital asset purchased (e.g., cold chain storage, improvements to vaccine tracking systems, computers, ventilators), capital authority must be granted by the Legislative Budget Board. Emergency requests can take weeks or months.”	6 (12)
Legislative issues	Related to a state law or requiring a legislative vote	“State law mandates critical or unanticipated budget changes, such as the award of PHER funds, be approved by the legislature, which requires at least a 30-day notice requirement.”	5 (10)
Staffing issues	Issues in hiring new staff or retaining specialized staff because of questions of continued funding	“Due to current financial situation, the governor and legislature have enacted/instituted extraordinary review processes for new hires.”	5 (10)
Funds transfer issues	Issues transferring funds from the state to the local level	“Getting the money out to the local health departments was a problem. Under the established rules/procedures the money was sent out as an advance, but this created problems as the local health departments were not familiar with doing it that way.”	4 (8)

Note. PHER = public health emergency response.

We analyzed data indicating the cumulative percentage of PHER funding allocated by each state as of March 2010, the midyear point in the H1N1 funding year and a mandatory reporting period, and merged them with the data on barriers. We used these data because they were available for each state and represented a point in time at which response planning had long been under way. Moreover, PHER funding terms stipulated that these dollars needed to be allocated within 1 year of disbursement; thus, there was little to no variation in these data at the end of the year. In addition to financial data, the CDC also provided data indicating the number of state PHEP employees per 100 000 capita. We obtained data (with permission) on public health organizational structure (i.e., centralized, decentralized, or mixed or shared) from the authors of a previous study that examined state public health characteristics (Beitsch et al.²¹). Data from the US Census were merged with these data and included US region (i.e., Northeast, South, Midwest, West) and state population size measured as a categorical variable (i.e., small, population < 4 000 000; medium, population = 4 000 000–8 999 999; large, population ≥ 9 000 000).³

Analysis

We first examined the frequency of the barriers identified. Next, we used the χ² test and analysis of variance to examine the relationship between each barrier and state-level variables (e.g., US census region, population size, and centralization). In addition, using an independent-sample t test, we explored whether each barrier, or a cumulative effect of more than 1 barrier, was associated with the percentage of PHER funds allocated at midyear reporting. Last, using linear regression modeling, we examined the dependent variable of percentage of PHER funds allocated in relation to the independent variable of barriers experienced (models included the individual barriers or the cumulative number of barriers) and state characteristics, including the number of PHEP staff per 100 000 capita, population size, US census region, and centralization. We conducted these analyses with SPSS version 16 (SPSS Inc., Chicago, IL) and considered P levels of less than or equal to .10, less than or equal to .05, and less than or equal to .01 for statistical significance. When working with a relatively small sample size (such as 50 states) and limited power, it is useful to examine statistical significance at a P level of less than .1.

RESULTS

Data represented in these analyses include all 50 US states (the CDC funds New York City, Los Angeles County, and the city of Chicago directly, and we did not include data for these jurisdictions in the analysis). We identified a total of 6 different barriers to allocating funds during the review of the PHER grant reports. Table 1 presents the frequency, a summary explanation, and sample quotes from the grant reports data representing each of the barriers identified. The most frequently cited barriers were regulatory issues (14 states, or 28%) and contracting issues (14, or 28%). Regulatory issues were indicated when states reported a state-specific rule or guideline that affected PHER funds allocation. Contracting issues denoted any accounting or administrative process barriers such as experiencing difficulties getting signatures or approvals. Purchasing issues (6, or 12%), legislative issues (5, or 10%), and staffing issues (5, or 10%) were the next most common barriers. Purchasing issues were any difficulties related to the procurement process. Legislative issues were indicated when a state law or legislative vote was necessary for a specific PHER-related process. Staffing issues were noted when there were difficulties hiring new staff, retaining specialized staff because of changes in funding, or reassigning staff to response tasks. Last, 4 states (8%) identified funds transfer issues as a barrier to efficiently allocating PHER funds in circumstances in which there were problems transferring funds to the local level.

The mean state allocation was 42.7% (range = 10%–93%) of their PHER funds by midyear (March 2010). The mean number of PHEP staff per 100 000 capita was 2.36 employees, the median was 1.32, and the values ranged from 0.25 to 13 PHEP staff.

Table 2 presents the results of analyses examining the relationship between the percentage of PHER funding allocated by March 2010 and whether a given state had each specific barrier. Overall, only purchasing barriers were significantly associated (P = .05) with the percentage of PHER funds allocated, with states that reported purchasing barriers allocating just more than a quarter of PHER funds (28.5%), compared with all other states, which allocated almost half of their PHER funds (44.7%). The other 5 barriers were not related to the percentage of PHER funds allocated by states. None of the 6 barriers was significantly related to the number of PHEP staff per 100 000 capita (data not shown).

TABLE 2—

Relationship Between Percentage of Public Health Emergency Response Funds Allocated and Barriers Identified in Grant Reports of State Health Departments: United States, October 2009–May 2010

Type of Barrier	% of PHER Funds Allocated,a Mean (SD)	P
Regulatory		.59
Yes	45.1 (22.0)
No	41.7 (18.1)
Contracting		.19
Yes	48.6 (20.8)
No	40.5 (18.2)
Purchasing		.05
Yes	28.5 (14.7)
No	44.7 (18.9)
Legislative		.92
Yes	41.7 (17.6)
No	42.8 (19.4)
Staffing		.54
Yes	37.7 (11.4)
No	43.2 (19.8)
Funds transfer		.68
Yes	46.4 (24.3)
No	42.3 (18.8)

Note. PHER = public health emergency response.

^aAs of March 2010.

Table 3 presents results examining the relationship between any or more than 1 barrier identified and the percentage of PHER funds allocated. Whereas having at least 1 barrier identified was not associated with the proportion of PHER funds allocated (40.1% vs 45.3%; P = .36), the cumulative effect of having 3 barriers (no state had > 3 barriers) was significantly associated with a reduction in the amount of PHER dollars allocated (27.8% vs 40.1%; P = .03; Table 3). In sum, having 1 or 2 barriers is not a problem when it comes to allocating PHER funds, but once a state experiences 3 problems, a significant challenge arises. Additionally, the number of PHEP staff per 100 000 capita was negatively related to the number of barriers reported (1.5 employees per 100 000 capita in states with ≥ 1 barrier vs 3.2 employees in states with no barriers; P = .03; data not shown).

TABLE 3—

Relationship Between the Percentage of Public Health Emergency Response Funds Allocated and Number of Barriers Identified in Grant Reports of State Health Departments: United States, October 2009–May 2010

Barrier	% of PHER Funds Allocated,a Mean (SD)	P
Presence of barrier		.36
Any (≥1)	45.3 (18.8)
No barrier	40.1 (19.4)
Total no. of barriers		.03
0	40.1 (19.4)
1	45.4 (14.1)
2	55.6 (17.8)
3	27.8 (19.1)

Note. PHER = public health emergency response.

^aAs of March 2010.

Last, results of the regression analyses indicated that having experienced a purchasing barrier was associated with a 26.4% point decrease in PHER allocation at midyear (B = −26.4; P = .018; Table 4). However, having experienced a contracting barrier was associated with a 20.4% point increase in PHER allocation at midyear (B = 20.4; P = .05). Separately, the cumulative effect of experiencing at least 3 barriers was associated with a 16% point decrease in PHER allocation at midyear (B = −16.0; P = .079).

TABLE 4—

Relationship Between the Percentage of Public Health Emergency Response Funds Allocated and State-Reported Barriers to Allocation Identified in Grant Reports of State Health Departments: United States, October 2009–May 2010

State Variable	% of PHER Funds Allocated, B (SE)	% of PHER Funds Allocated,a B (SE)
Individual barriers		Omitted
Regulatory	−5.6 (10.5)
Contracting	−26.4* (10.6)
Purchasing	20.4* (10.0)
Legislative	2.3 (12.5)
Staffing	−8.2 (9.6)
Funds transfer	−0.3 (12.7)
Experienced 3 barriers	Omitted	−16.0 (8.85)
No. of public health emergency preparedness staff per 100 000 persons	0.9 (1.5)	−0.3 (1.43)
Regionb
South	14.1 (10.7)	1.3 (9.3)
Midwest	7.1 (9.25)	4.8 (9.5)
West	7.7 (9.23)	5.1 (9.0)
State population sizec
Medium	−4.9 (7.26)	−2.1 (7.1)
Large	5.1 (12.0)	−2.6 (10.1)
Centralizationd
Decentralized	13.1 (10.1)	3.7 (9.1)
Mixed or shared	−5.0 (11.3)	−7.6 (11.7)

Note. PHER = public health emergency response.

^aAs of March 2010.

^bNortheast is the reference region.

^cSmall (population < 4 000 000) is the reference size. Medium is defined as population = 4 000 000–8 999 999. Large is defined as population ≥ 9 000 000.

^dCentralized is the reference level of centralization.

*P < .05.

DISCUSSION

Given the large amount of PHER funds allocated to the states and the urgency with which those funds needed to be spent, the capacity of states to allocate PHER funds for planning and response to the H1N1 outbreak is an important issue that can inform future response activities. (PHER funds were initially provided for a 12-month period; it was particularly important to be able to use these funds quickly because they were meant to prepare to respond or begin the response to what could have been a widespread outbreak.) Because some states had allocated as little as 10% of their PHER funds 6 months into the response, we examined what barriers might have slowed or prevented the allocation of these funds and whether certain barriers were associated with having allocated significantly less funds by the midyear reporting period. Identifying barriers to the timely allocation of PHER funds is important because future public health emergencies may require quicker allocation of these types of funds to effectively protect the public’s health.

Overall, we identified 6 potential barriers in our analysis of the PHER grant reports. In both bivariate and multivariate analyses, only purchasing issues and the cumulative effect of 3 barriers were significantly related to reductions in the percentage of PHER funds allocated, despite states’ after-action reports indicating that these individual issues influenced their response efforts.² Thus, a state health department that experienced numerous types of barriers or a purchasing barrier, regardless of the number of PHEP staff, was more likely to experience delays in using its emergency funds. Purchasing barriers may be reflective of other inefficiencies in a state’s system in general or the grants administration system specifically, although we are unable to provide more insight into this barrier. Moreover, why experiencing a contracting barrier is associated with a significant increase in the percentage of funds allocated by midyear is unclear. We can speculate that this barrier may have been easily overcome or that an unmeasured confounder is associated with having this barrier and the increased allocation of PHER funds.

Since we examined the use of PHER funds during the 2009 to 2010 H1N1 response, our results provide insight into the PHER fund allocation experiences during this response event, information that is valuable for plans to reduce or remove barriers to allocating funds in future public health emergencies. We do not necessarily provide information about barriers to using other past, present, or future federal PHER funds, but our results do draw attention to the importance of examining internal organizational processes as 1 factor that may improve PHEP. Given our findings, and previous research that indicated that after-action reports are not consistently being used to actively address weaknesses experienced during events,¹⁷ table-top exercises may be 1 strategy for examining and revising the internal response processes of state public health agencies. Such exercises could rely on after-action reports to facilitate the development of a catalog of process barriers that can inform performance improvement processes to overcome the identified barriers and guide the development of intraorganizational emergency response plans.^2,17 Because most table-top exercises have historically focused on improving interorganizational coordination and collaboration,^10,13 intraorganizational exercises may provide valuable resolutions to the organizational issues experienced during the H1N1 response. Additionally, developing intraorganizational response plans may require a unique set of team members from within the public health agency, as well as members of other offices, such as procurement and contracts.¹⁹ Consideration for essential participants should be given in accordance with the various actors and offices involved in the administration of a response and the processing and use of PHER funds.

As an alternative to developing intraorganizational response plans, a performance management process could be used to address internal issues that arose during the H1N1 response. There is growing experience with performance management techniques within public health departments, especially given the recent focus on public health accreditation and the release of revised resources such as the Turning Point Performance Framework and Toolkit.^22–24

This study is the first to empirically examine state use of PHER funds and associated barriers to allocating these funds during a response event. In addition, this study included all 50 states and is therefore inclusive and generalizable when discussing state experiences during the H1N1 event. Nevertheless, we note several limitations. First, the data were based on self-report and may be subject to desirability bias. Such a bias may stem from the nature of grants reporting data whereby states may prefer to present successes instead of failures or difficulties. In addition, grant management reports and the data they provide may be subject to recall bias because these reports are typically based on an individual’s or team’s recollection of experiences. Thus, it is possible that the data we analyzed had inaccuracies. Last, given the nature of qualitative analyses, barriers identified in the qualitative review were subject to interpretation. In some circumstances, certain issues could potentially have been coded differently and, as management and preparedness researchers, we used collective judgment to code for that which was most appropriate given the entire context of the text. We believe these methods are scientifically justified and serve in analyzing performance data for the purpose of identifying issues and potential strategies to mitigate these issues.²⁵ As is common with qualitative research, this process is the first step toward understanding organizational experiences during a public health response emergency, and additional research is needed.

We did not examine the relationship between organizational response barriers and vaccination rates. Although the H1N1 outbreak was not as widespread as initially anticipated, almost half of the 162 million doses of H1N1 vaccine produced for the response went unused, and less than half of the US population was vaccinated.²⁶ A number of factors may have contributed to the mass vaccination campaign and overall vaccination rates, although it is difficult to attribute causality to any 1 indicator, including state response barriers.

In conclusion, this study provides valuable new insight into state experiences allocating PHER funds during a response event. When state health departments experience numerous types of barriers or, specifically, a purchasing barrier, they experience delays in using emergency funds. More research is needed to examine these barriers and the impact of delays in using funds—for the purpose of both addressing these barriers and protecting the public’s health in future disaster and emergency response events. Experiences described in this study should be considered when planning for future events that require the expedited use of federal funds in future influenza or other emergency and disaster response events. Our findings may be of interest to officials and policymakers who have the ability to determine funding protocols and potentially aid states in overcoming common barriers experienced during the H1N1 response.