A National Study Examining Closed Points of Dispensing (PODs): Existence,Preparedness,Exercise Participation,and Training Provided

Abstract

The Centers for Disease Control and Prevention recommends using open points of dispensing (PODs) and alternative modalities, such as closed PODs, for mass dispensing of medical countermeasures. However, closed POD existence has not been assessed. In 2013 we sent an online questionnaire to US Cities Readiness Initiative (CRI) and non-CRI public health disaster planners. Chi-square tests were used to determine differences between CRIs and non-CRIs when comparing having at least 1 closed POD, and to compare having a closed POD and perceived mass dispensing preparedness. A total of 301 disaster planners participated. Almost all (89.3%, n=218) jurisdictions have considered establishing a closed POD, and three-quarters (74.2%, n=181) currently have at least one. CRIs were more likely than non-CRIs to have a closed POD (85.0% vs 58.5%, X²=21.3, p<.001). Those with 1 or more closed PODs were more likely to believe their jurisdiction could distribute medical countermeasures within 48 hours compared to those without a closed POD (78.5% vs 21.5%; X²=10.8, p=.001). Half had a written plan and/or written standing orders (59.1% and 52.5%, respectively). Almost half (42%, n=72) have done no preevent training for POD staff in the past 2 years; almost 20% (18%, n=32) do not plan to offer any just-in-time training. Nearly 40% (n=70) have conducted no exercises in the past year. Closed PODs contribute to community preparedness; their establishment should be followed by development of written plans, worker training, and exercises.

The CDC recommends using open points of dispensing (PODs) and alternative modalities, such as closed PODs, for mass dispensing of medical countermeasures, but closed PODs have not been assessed. Of public health disaster planners who were surveyed, almost all jurisdictions they represented have considered establishing a closed POD, and three-quarters currently have at least one. Those with 1 or more closed PODs were more likely to believe their jurisdiction could distribute medical countermeasures within 48 hours compared to those without a closed POD. Closed PODs contribute to community preparedness, and their establishment should be followed by development of written plans, worker training, and exercises.

Biological events, whether natural or man-made, necessitate a quick response in order to distribute medical countermeasures that will minimize morbidity and mortality related to the event. During a large-scale public health emergency, medical countermeasures and supplies may need to be dispensed to many thousands of individuals in a very short time. In order to aid in public health preparedness for mass dispensing of medical countermeasures, the Centers for Disease Control and Prevention (CDC) created the Cities Readiness Initiative (CRI) in 2004.¹ The CRI provides funding to state health departments across the nation, including 72 of the major metropolitan statistical areas (MSAs), accounting for 57% of the US population.² This funding is intended to enhance CRI jurisdictions' ability to rapidly dispense medical countermeasures to their citizens.

The CDC indicates that communities should aim to be able to dispense medical countermeasures to their entire population within 48 hours.¹ Rapid mass dispensing is complex and requires a network of points of dispensing (PODs). PODs are referred to as either “open” or “closed.” Open PODs are sites that are available to everyone in the general public and thus need to be located in centralized areas that provide easy access to all community members.³ Closed PODs are hosted by private agencies or businesses that have a formal agreement with a health department to dispense medical countermeasures to their employees, students (if it is an academic institution), and/or volunteers and to these individuals' household members.¹ Closed PODs can provide great benefit to communities by increasing access to medical countermeasures for high-risk individuals for whom getting to an open POD would be challenging, such as inmates or nursing home residents. In addition, closed PODs can help relieve the burden of dispensing all medical countermeasures through open PODs by shifting the responsibility of dispensing from public health departments to the agency or organization hosting the closed POD. Closed PODs also provide a benefit to the hosting agency, because they allow for rapid distribution of medical countermeasures to employees and their family members, which may increase the staff's sense of security.¹

Open POD preparedness can be assessed in multiple ways. Currently, the CDC and state public health personnel conduct an annual assessment of health departments that are part of the CRI, using the technical assistance review (TAR) to determine how well prepared jurisdictions are to dispense medical countermeasures during a public health emergency.⁴ A 2011 report from the CDC indicates that preparedness has increased overall since earlier assessments conducted in 2006, although some jurisdictions still do not meet the CDC standards of at least 69% on the technical assistance review.^5,6 A 2008 report from the Institute of Medicine¹ indicates that few communities are prepared to dispense medical countermeasures to all community members within CDC's goal of 48 hours.

Researchers have evaluated various aspects of open POD preparedness, some of which may be applicable to closed PODs as well, given that many of the planning considerations are similar for both types of mass dispensing sites. Many past studies have consisted of modeling efforts to examine interventions aimed at improving POD throughput,^7-10 such as using a head-of-household dispensing method,⁸ and strategies to control traffic and parking at the POD.¹¹ A 2013 study by Everett et al¹² demonstrated how geographic information systems (GIS) can be used to identify the best locations in a community for open PODs. Two studies^13,14 described how POD exercises were used to measure throughput and identify ways that processes can be improved. POD preparedness, as defined by the CDC's technical assistance review, was measured twice, 1 year apart in 2004 and 2005, and was found to be increasing, though only CRI sites were assessed.² In 2006, Connecticut POD workers' competence and training needs were assessed;¹⁵ researchers found that a third of them needed additional training in at least 5 areas in order to perform the duties outlined on their job action sheets, regardless of whether they were paid or volunteer staff. As an alternative to measuring POD functioning or preparedness through modeling or exercises, some researchers have examined patient satisfaction with their experience at PODs. A 2010 study reported that almost 85% of POD clients rated their experience at a POD used for distributing influenza vaccine as “excellent.”¹⁶

Although disaster planning experts and the CDC have recommended that public health agencies implement alternative dispensing modalities, such as the use of closed PODs, to maximize POD throughput in a community, little research exists on this alternative dispensing modality. There are multiple assumed limitations to closed PODs—for example, they typically exist in nonmedical agencies, they may not be able to provide injections or vaccine, and they may lack written plans and training—though these limitations have not been quantified scientifically.

The only existing published literature on closed PODs includes case studies and lessons learned from past exercises. A 2006 article¹⁷ describes a business and public health collaborative effort to establish a mass dispensing site through a single large business in Georgia; the term “closed POD” was never used by the authors, but the arrangement described fits the closed POD definition. The authors used the success case method to evaluate the business-based closed POD project and found that the collaborative model was successful.¹⁷ Another article in 2006 described a newly developed software program aimed at easing prophylaxis distribution during biological events; the software was tested in a hospital-based mass dispensing exercise, with the assumption that the hospital would function as a closed POD during a future event.¹⁸ The exercise found that the new software improved throughput and decreased staffing needs, and the author reported that the software would be used in both open and closed PODs.¹⁸

Doyle-Moss¹⁹ reported on lessons learned from the University of Rhode Island's efforts to establish a closed POD. Veltri et al²⁰ described a case study involving a hospital-based exercise aimed at assessing the facility's ability to administer vaccine to staff, which mimics a closed POD mass vaccination process. To date, no research has been conducted nationwide on the extent to which closed PODs have been or are being established, nor has the preparedness of existing closed PODs been assessed in a standardized manner. The purpose of this study is to examine the percentage of communities that currently have established a closed POD(s), the training being provided or planned for closed POD staff, and exercises being conducted to assess closed POD preparedness.

Methods

This study consisted of an online questionnaire sent to public health disaster planners in the United States in summer and early fall 2013. Recruitment was attempted with all 456 CRIs and a random sample of 500 non-CRI jurisdictions over a 3-month period. Jurisdictions were contacted by phone, and a recruiter asked to speak with the person in charge of disaster planning for the community. The disaster planner was then informed of the study and asked if he or she would be willing to participate. If the disaster planner agreed to participate, a recruitment email was sent that included a link to the online questionnaire administered through Qualtrics software (version 2013). A modified Dillman's Total Design Method²¹ was used to maximize response rates; this consisted of sending a follow-up email 2 weeks after the initial recruitment email (to those who were reached by phone and who agreed to participate) to maximize response rates. The recruitment plan also aimed to leave at least 3 voice messages with each jurisdiction or disaster planner when trying to recruit by phone; however, this process proved infeasible because of time constraints. No names or other identifying information were included on the survey. The Saint Louis University Institutional Review Board approved this study.

Instrument

No existing research measuring closed POD preparedness could be found in the literature; we found only case studies and summaries of lessons learned. Therefore, a new instrument needed to be created for this study. The instrument created was based on CDC's technical assistance review criteria for open POD preparedness,⁵ a literature review summarizing infection prevention and training recommendations for open PODs,³ and findings from a mass dispensing workshop published in a report by the Institute of Medicine.¹

The instrument was pilot tested by a group of 10 public health professionals from across the US who are responsible for open PODs in their communities. Pilot testing consisted of assessing the length of time needed to complete the questionnaire, ease of use, and clarity and thoroughness of items and answer options. Feedback from pilot testing was used to refine the instrument. The final 54-item instrument measured: (1) whether a jurisdiction has considered having a closed POD(s); (2) criteria used to choose closed POD site(s) (3 items); (3) whether the jurisdiction has 1 or more closed PODs; (4) percentage of jurisdiction covered by closed POD(s); (5) groups or agencies currently covered by a closed POD (12 items); (6) groups or agencies being considered as a potential future closed POD (11 items); (7) whether the closed POD(s) have existing written standing orders; (8) whether the person who will sign the standing orders during an emergency has been identified and consented to signing the orders; (9) preevent training provided to closed PODs workers (8 items); (10) just-in-time training planned for closed PODs workers (8 items); and (11) exercises conducted by closed PODs in the past year (6 items). In addition, demographic information was collected.

Data Analysis

All data analyses were performed using the R statistical program. A closed POD preparedness score was developed by assigning up to 1 point for each of 3 components:

1. Written plan for the closed POD(s) (yes for all PODs: 1 point; yes for some PODs: 0.5 points; no for all PODs: 0 points);

2. Standing orders in place (yes: 1 point; no: 0 points); and

3. Designated signer for standing orders (yes: 1 point; identified but not yet consented: 0.5 points; no: 0 points), for a total 3 possible points.

Six distinct exercises were identified, and the number each jurisdiction had conducted in the past year was counted (ie, exercise score). Preevent training provided in the past 2 years and just-in-time training planned were analyzed on a per-jurisdiction basis as well as overall to determine which training sessions were conducted and/or planned most frequently.

All 3 outcomes were compared to the demographic variables. Comparisons were made using independent samples t-tests for binary demographic variables and Kruskal-Wallis tests for categorical variables with more than 2 categories. Chi-square tests were used to determine differences between CRIs and non-CRIs when comparing whether they had at least 1 closed POD and which of the training exercises were done most frequently. Chi-square was also used to compare those with and without at least 1 closed POD and perceived preparedness (ie, perceived ability of the jurisdiction to distribute mass antibiotic prophylaxis to all community members in 48 hours). Fisher's Exact test was used to compare percentage of community covered by closed POD(s) and perceived preparedness. Regression models were investigated, but none yielded an R₂ above 0.25, so results were omitted.

Results

By extending the recruitment window to 4 months, recruiters were able to speak with 632 individuals who were asked to participate in the study; 20 refused to participate. Links to the online survey were provided to the 612 consenting individuals; 301 surveys were returned. Forty-four questionnaires were excluded because of excessive missing data, yielding data on 257 subjects, for a 41% response rate. Of the 257 completed surveys, nearly all of the respondents were full-time employees (85%, n=218), and most had a bachelor's degree or higher (76%, n=196). Most (68%, n=173) were female. More than half (57%, n=147) worked in a CRI jurisdiction. Only 4.3% (n=11) reported working for a tribal jurisdiction. Over half of the respondents (50%, n=129) covered jurisdictions of fewer than 100,000 people, with another 26% (n=68) covering jurisdictions with between 100,000 and 500,000 people. Approximately half (48%, n=123) reported having no formal medical training; 23% (n=60) were nurses, and another 14% (n=35) were emergency medical technicians (EMTs). Few (9%, n=23) were physicians or had some other type of medical training. Most respondents had been in their current role for at least 2 years and reported at least 6 years' work experience in disaster planning. About a third (37%, n=94) had been in their current role for 2 to 5 years, and another 27% (n=70) had been in their current role for 6 to 10 years. About a third (35%, n=90) have 6 to 10 years' experience, and another third (33%, n=84) had 11 or more years' experience in the field.

Interest Expressed in Closed PODs

Most respondents (89.3%, n=218) indicated that their jurisdiction has considered establishing a closed POD(s) in their community. Those who reported that closed PODs are being considered (n=218) were asked which criteria were being used to decide which businesses or agencies would be approached for possible recruitment as a closed POD site; respondents could choose more than 1 criterion. Almost three-quarters (72.9%, n=159) indicated that potential closed POD sites were chosen based on whether they have medical personnel on site. Many (70.6%, n=154) reported that they considered business or agency size when choosing closed POD sites; half (49.5%, n=108) indicated that business type influenced whether the site would be approached for recruitment.

Closed POD Network

Three-quarters of respondents (74.2%, n=181) reported having at least 1 closed POD in their jurisdiction. Overall, 82.0% (n=200) of respondents reported that they believed their jurisdiction could meet the CDC goal of distributing antibiotic prophylaxis to their entire community within 48 hours (ie, perceived preparedness). Those who reported having at least 1 closed POD (n=181) were significantly more likely than disaster planners from jurisdictions who did not have at least 1 closed POD to report perceived preparedness (78.5% with ≥1 closed POD believed they could distribute antibiotics in 48 hours vs 21.5% of those without a closed POD who believed their jurisdiction had this capacity; X²=10.8, p=.001).

The remainder of the data analysis results focus on the subset of the sample who reported having at least 1 closed POD. Of those who reported having at least 1 closed POD (n=181), almost half (47%, n=85) reported that 0% to 10% of their jurisdiction's population is covered by the closed POD, and another 38.7% (n=70) indicated that 11% to 30% of their population is covered by a closed POD(s). Only 7.7% (n=14) indicated that their closed POD network covers 51% or more of their jurisdiction's population.

CRIs were significantly more likely than non-CRI jurisdictions to report having at least 1 closed POD (85.0% vs 58.5%, X²=21.3, p<.001). The impact of having more closed POD coverage was assessed in relation to perceived preparedness. When examining only those jurisdictions that had at least 1 closed POD (n=181), those jurisdictions that reported that ≥31% of their community members are covered by a closed POD(s) were significantly more likely to report perceived preparedness compared to those whose closed POD network covered ≤30% of their community members (100% of those with ≥31% coverage reported perceived preparedness vs 84.5% of those with ≤30% coverage, p<.05).

Respondents who indicated that they had at least 1 closed POD (n=181) were asked to identify the group(s) covered by an existing closed POD. Hospitals and first response agencies were the most frequently reported organizations being covered by a closed POD (79.6%, n=144, and 77.9%, n=141, respectively) (Table 1). Academic institutions and corrections agencies were the least likely organizations to have an established closed POD (14.9%, n=27, and 6.1%, n=11, respectively). CRIs were significantly more likely than non-CRI jurisdictions to have an established closed POD covering the following groups: first response agency, city/county employees, long-term care, faith-based organization, nonhospital healthcare agency, private business, federal or state employees, utility company, and academic institution (Table 1). There was no difference between CRI and non-CRI jurisdictions and whether they had a closed POD covering a hospital, postal service, or corrections agency (Table 1). Many jurisdictions reported having multiple established closed PODs. Respondents could identify up to 11 groups covered by an established closed POD, giving a possible range of 1 to 11 established closed PODs. Responses ranged from 1 to 11, with an average of 5 groups covered by each jurisdiction.

Table 1.

Groups Currently Covered by a Closed POD for CRI Versus Non-CRI Jurisdictions

			CRI vs Non-CRI^a N=180
	All Respondents N=181		CRI N=125		Non-CRI N=55
Group	%	n	%	n	%	n	CRI vs Non-CRI P value^b
Hospital	79.6	144	81.6	102	74.5	41	NS
First responders	77.9	141	85.6	107	60.0	33	<.001
City/county employees	60.8	110	68.8	86	43.6	24	=.01
Long-term care agency	59.1	107	64.8	81	47.3	26	<.05
Faith-based agency	46.4	84	52.8	66	32.7	18	=.01
Nonhospital healthcare agency	43.6	79	51.2	64	27.3	15	<.01
Private business	40.9	74	49.6	62	20.0	11	<.001
Federal or state employees	33.7	61	39.2	49	21.8	12	<.05
Utilities or public transit	21.0	38	25.6	32	10.9	6	<.05
Postal service	19.9	36	22.4	28	12.7	7	NS
Academic institution	14.9	27	18.4	23	7.3	4	<.05
Corrections	6.1	11	4.0	5	10.9	6	NS

Denominator does not match the total sample for “all respondents” because of missing data.

Determined by the X² test.

NS=not significant.

In addition to identifying existing closed POD coverage, respondents who indicated that they had at least 1 closed POD (n=181) were asked to identify the group(s) being considered for establishment of a future closed POD. Hospitals and first response agencies were the organizations most frequently reported as being considered for potential future coverage through a closed POD (27.1%, n=49, and 25.4%, n=46, respectively) (Table 2). Far fewer jurisdictions reported considering establishing a closed POD for utility companies (12.2%, n=22), the postal service (10.5%, n=19), or academic institutions (6.1%, n=11) (Table 2). There was no difference between CRI and non-CRI jurisdictions on whether they were considering a closed POD covering any organization or site, except for first response agencies; CRIs were significantly more likely to report considering establishing a closed POD at a first response agency compared to non-CRIs (30.4% vs 14.5%, X²=5.0, p<.05) (Table 2).

Table 2.

Groups Being Considered for Coverage by a Closed POD

			CRI vs Non-CRI^a N=180
	All Respondents N=181		CRI N=125		Non-CRI N=55
Group	%	n	%	n	%	n	CRI vs Non-CRI P value^b
Hospital	27.1	49	26.4	33	29.1	16	NS
First responders	25.4	46	30.4	38	14.5	8	<.05
City/county employees	22.7	41	19.2	24	29.1	16	NS
Long-term care agency	22.7	41	24.0	30	20.0	11	NS
Faith-based agency	22.1	40	20.8	26	23.6	13	NS
Nonhospital healthcare agency	16.0	29	14.4	18	18.2	10	NS
Private business	16.0	29	16.0	20	14.5	8	NS
Federal or state employees	14.4	26	12.8	16	16.4	9	NS
Utilities or public transit	12.2	22	12.8	16	10.9	6	NS
Postal service	10.5	19	8.0	10	16.4	9	NS
Academic institution	6.1	11	4.8	6	9.1	5	NS

Denominator does not match the total sample for “all respondents” because of missing data.

Determined by the X² test.

NS=not significant.

Closed POD Preparedness

Those who reported having at least 1 closed POD (n=181) were asked 3 questions related to aspects of closed POD preparedness: (1) whether the closed POD(s) had a written plan, (2) if the closed POD(s) had existing written standing orders, and (3) if a designated person had been identified to sign the standing orders and if he or she had consented to sign them. From these 3 measures, a closed POD preparedness score was calculated (see Methods for details). More than half of those who reported having at least 1 closed POD (59.1%, n=107) indicated that they had a written plan for all closed PODs; however, 13.8% (n=25) reported having at least 1 closed POD, but did not have a written plan for the site. Although only half of those who have at least 1 closed POD (52.5%, n=95) indicated that the closed POD(s) has existing written standing orders, almost all of those jurisdictions with written standing orders for their closed POD(s) (90.5%, n=86) indicated that a qualified signer had been identified and had agreed to sign them. The closed POD preparedness scores spanned the entire range from 0 to 3, with an average preparedness score of 1.74. The distribution was highly bimodal, with 48% of respondents scoring a 1 or lower, and 35% of respondents scoring 3, the highest score. In investigating possible explanations for the variability of preparedness score, being in a CRI jurisdiction showed a clear trend, with non-CRI jurisdictions more likely to have scores of 0, 0.5, and 1, and CRI jurisdictions more likely to have scores of 1.5 and above; however, the differences in rates did not reach statistical significance (p=0.21).

Training for Closed PODs

Respondents who reported that their jurisdiction had at least 1 closed POD (n=181) were asked to identify the topics on which POD staff had been trained during the past 2 years; 8 training topics were assessed, and respondents could choose more than 1. The average number of preevent training sessions conducted in the past 2 years was 2.9 (SD 3.0). However, almost half (42%, n=72) reported having done no preevent training in the past 2 years. The remaining respondents were evenly distributed between 1 and the maximum 8 training sessions. Jurisdictions in CRIs averaged 1 more preevent training activity than those not in a CRI (p<.05). The most frequently reported preevent training topics were: (1) collecting personal identifying information, (2) assessing POD clients for medication and/or vaccine contraindications, and (3) screening procedures for identifying potentially contagious individuals (Table 3). Labeling and repackaging of medications were the 2 least frequently reported preevent training topics provided (Table 3). CRI jurisdictions were significantly more likely than non-CRIs to have provided 4 types of preevent training topics (p<.05 for all), though there was no difference between CRIs and non-CRIs for the other 4 training topics (Table 4).

Table 3.

Preevent Training Provided in the Past 2 Years to Closed POD Staff

			CRI vs Non-CRI^a N=180
	All Respondents N=181		CRI N=125		Non-CRI N=55
Training Topic	%	n	%	n	%	n	CRI vs. Non-CRIP value^b
Collecting personal identification	48.1	87	53.6	67	36.4	20	<.05
Assessing POD clients for medication contraindications	43.1	78	48.8	61	30.9	17	<.05
Screening to identify symptomatic individuals	40.9	74	46.4	58	29.1	16	<.05
Interpreting antibiotic dosing per protocol	39.8	72	45.6	57	27.3	15	<.05
Vaccine administration procedures	34.8	63	35.2	44	34.5	19	NS
Adverse event reporting procedures	34.3	62	36.8	46	29.1	16	NS
Labeling of medication	30.9	56	34.4	43	23.6	13	NS
Repackaging of medication	17.1	31	19.2	24	12.7	7	NS

Denominator does not match the total sample for “all respondents” because of missing data.

Determined by the X² test.

NS=not significant.

Table 4.

Just-In-Time Training Planned for Closed POD Staff

	All Respondents N=181
Training Topic	%	n
Collecting personal identification	69.1	125
Assessing POD clients for medication contraindications	34.1	116
Screening to identify symptomatic individuals	34.1	116
Interpreting antibiotic dosing per protocol	63.0	114
Vaccine administration procedures	62.4	113
Adverse event reporting procedures	61.9	112
Labeling of medication	61.3	111
Repackaging of medication	57.5	104

When asked about just-in-time training, about a third of respondents (37%, n=67) reported that they planned to offer all 8 topics, though almost 20% (18%, n=32) said they planned to offer none. The remaining respondents were evenly distributed between offering 1 and 7 just-in-time training activities. There was a negative correlation between the number of preevent and just-in-time training activities planned in a jurisdiction (r=–0.41, p<.001). The most common responses were no preevent training but all 8 just-in-time training activities (27%, n=49), no preevent or just-in-time training (7%, n=13), and all 8 preevent training activities but no just-in-time training (6.6%, n=12). Chi-square tests assessed the assumption that each type of training was equally likely to be done preevent only, just-in-time only, both preevent and just-in-time, and never planned. Repackaging of medications was most likely to never be done (p<.001) and least likely to be preevent only (p=.01) or both preevent and just-in-time (p<.05). Labeling of medications was also more likely to never be planned or planned only for just-in-time training, but this difference did not meet statistical significance. The frequencies of planned just-in-time training are outlined in Table 4.

Closed POD Exercises

Respondents who reported that their jurisdiction had at least 1 closed POD (n=181) were asked to identify the types of closed POD exercises that had been conducted during the past year; 6 exercise types were assessed, and respondents could choose more than 1. The average number of closed POD exercises performed in the past year was 1.8 (SD 2.1). Nearly 40% (n=70) reported that they had conducted no closed POD exercises; among non-CRIs, the rate was 51% (95% CI: [37%, 64%]). Of the demographic variables investigated, none reached statistical significance in explaining differences in the average number of exercises performed. The frequencies of closed POD exercises are outlined in Table 5.

Table 5.

Exercises Conducted by Closed POD in the Past Year

	All RespondentsN=181
Type of Closed POD Exercise	%	n
Staff or volunteer notification	47.5	86
Facility set-up	35.9	65
Dispensing throughput	30.9	56
Full-scale or site activation	28.7	52
Actual operation of POD during an event	22.7	41
Pick list generation	18.2	33

Discussion

This study found that almost all jurisdictions are considering establishing a closed POD in their community, and almost three-quarters already have 1 or more in place. However, the percentage of the community covered by jurisdictions' closed POD network varies widely, with half covering only a very small portion of their population while others plan to dispense medical countermeasures to almost half of their citizens through closed PODs. CRI jurisdictions were significantly more likely than non-CRIs to report having at least 1 closed POD, but the reasons for this are unknown. It seems likely that funding contributes, at least to some extent, to this difference. CRIs are provided federal and state funding to enhance community preparedness for public health emergencies. This funding allows investment of staff resources and supplies into the development of closed POD networks. Non-CRI jurisdictions could benefit greatly from the establishment of closed PODs, because these alternative dispensing modalities can decrease the burden at open POD sites by 40% to 50%.¹ In addition, establishment of closed PODs provides benefits of security and protection to the hosting agency's staff and their household members by ensuring rapid and convenient access to medical countermeasures.

Findings from this study indicate that jurisdictions that have established a single closed POD were highly likely to have developed multiple closed PODs, effectively creating a closed POD network in their community. The groups most frequently covered by closed PODs were hospitals, first response agencies, and city or county employees. This finding is in line with respondents' indication that closed POD sites have been chosen based on the existence of medical oversight at the agency and large business size, both of which would be common among hospitals and emergency medical services organizations. In addition, choosing hospitals, first response organizations, and city or county agencies as closed POD sites enhances community resilience by ensuring rapid dispensing of medical countermeasures to workers who are vital in responding to the public health emergency. Other businesses that provide essential services during disasters, such as utilities and public transit, and other organizations that house high-risk individuals or those who would have difficulty getting to and from an open POD, such as long-term care and corrections facilities, would also benefit greatly from establishment of a closed POD. Public health disaster planners should consider approaching these agencies about establishing a closed POD site.

Respondents in this study indicate that, although closed POD networks are beginning to be established, there are many gaps in planning in these existing closed PODs. Only about half of the current closed PODs have a written plan in place and/or have existing standing orders. Almost half have not provided any type of preevent training for closed POD staff, and 40% have not conducted even a single exercise of the closed POD in the past year. It should be noted that, although many jurisdictions reported that they have not provided preevent training to closed POD staff, many are planning extensive just-in-time education.

Examining the closed POD preparedness information from this study in aggregate (general lack of written plans or standing orders, little to no preevent training, and few to no exercises), it appears that the feasibility of successful closed POD functioning in many communities is questionable. Just as with open PODs, closed PODs need to be planned carefully; staff require training on the plan, and the POD needs to be exercised during drills or an actual event so that gaps in planning can be identified and addressed. This general lack of preparedness found in this study is likely associated with the early stage of closed POD development, as the concept of closed PODs is relatively new. Research indicates that jurisdictions' open POD preparedness has increased over time;² it is likely that closed POD preparedness will continue to improve over time as well, though steps should be taken to ensure this progress. Jurisdictions that have a closed POD network in place should incorporate elements of planning and evaluation into their systems to ensure that the closed POD(s) will be successful when implemented during a public health emergency. This should include developing written plans, training staff, and conducting regular exercises to evaluate closed POD preparedness.

Developing a closed POD network has associated costs in terms of supplies and equipment as well as staff time needed to create the partnerships between public health and community agencies necessary to establish and sustain closed PODs. Estimating costs associated with development and maintenance of a closed POD network is beyond the scope of this study. However, disaster planners can minimize these anticipated costs by including closed POD volunteers in available training and exercises currently being offered to open POD staff and volunteers. This will not only decrease costs but will also increase preparedness across the jurisdiction, creating better community resilience. Furthermore, money invested in development of a closed POD network will be offset by the financial and health benefits of rapid dispensing of medical countermeasures. For example, a financial modeling study²² found that speed of medical countermeasure dispensing following an anthrax bioterrorism attack was the single most important determinant of lowering event-related mortality. If closed POD networks provide communities an effective method of reaching the CDC goal of being able to dispense medical countermeasures in 48 hours, the costs of establishing and sustaining closed PODs would be offset by the cost savings associated with lowering event-related mortality. Bravata et al²² reported that increasing medical countermeasure dispensing capacity up to 280,000 individuals per day is highly cost-effective to communities.

One of the most important findings from this study is that having at least 1 closed POD was significantly associated with higher perceived ability to meet the CDC standard of being able to distribute antibiotic prophylaxis to an entire community within 48 hours. This study also found an association between more extensive closed POD coverage (in terms of the percentage of the community covered by a closed POD) and perceived preparedness. In other words, a better closed POD network equated to higher perceived preparedness. This provides evidence that closed POD networks benefit community resilience with regard to mass dispensing of medical countermeasures and illustrates the importance of all communities, both CRI and non-CRI, beginning to better develop their closed POD network.

However, this finding should be interpreted with a bit of caution given that only communities' perceived ability to dispense antibiotic prophylaxis was measured; actual dispensing capability through open or closed PODs was not assessed in this study. It is also important to reiterate that disaster planners in this study reported that, although closed POD networks are beginning to be developed, much more extensive planning and exercises are needed to ensure that closed PODs are actually able to deploy successfully when needed. A poorly prepared closed POD may not provide maximum benefit to a community and may simply give false confidence to a jurisdiction without actually improving resilience.

Strengths of this study are that it is the first to examine the existence of, and extent of preparedness, for closed PODs. It is also a nationwide study involving not only CRIs (as are assessed by the annual CDC technical assistance review), but also a random sample of non-CRIs.

Some limitations should also be noted. The primary limitation is the moderate response rate. Recruitment for this study was quite challenging because of the difficulty in identifying and reaching the appropriate person to complete the questionnaire from each sampled public health department. Many jurisdictions were left 1 or more voice mails about the study, but it is unclear if the messages were left for the appropriate person or if the person actually received the message. In addition, when a disaster planner's email address was available online, a recruitment email was sometimes sent without actually speaking to the individual on the phone, which may mean that the recruitment email was either never received or was sent to the wrong person. Therefore, it is highly likely that our response rate was actually higher than what is reported. In addition, the nationwide sample and use of randomization for recruiting non-CRIs helps increase generalizability. There were also significantly more respondents from CRI jurisdictions compared with non-CRI jurisdictions, although approximately equal numbers were approached for recruitment, which may overestimate the existence of closed PODs throughout the US. Despite these limitations, this study is the first to examine the existence of closed PODs and the extent to which they are prepared to activate—information that is important for public health disaster planners.

US communities need to become better prepared for public health emergencies, including being ready to rapidly dispense medical countermeasures to large numbers of citizens. The CDC has recommended that a variety of modalities be used to achieve this, and the use of closed PODs is one such dispensing method. However, establishment of closed PODs is just the first step in better preparedness. Closed PODs need to be ready to activate, including have adequate written plans in place, training workers in advance and immediately before deployment, and exercising the closed POD prior to an actual event. Communities should continue working toward development and improvement of closed POD networks and other alternative dispensing modalities, to minimize morbidity and mortality during biological events.

References

Davis

, Kammersall

, Altevogt

; Institute of Medicine. Dispensing Medical Countermeasures for Public Health Emergencies: Workshop Summary. Washington, DC: National Academies Press; 2008.

Nelson

, Willis

, Chan

, Shelton

, Parker

. Federal initiative increases community preparedness for public health emergencies. Health Aff (Millwood), 2010; 29(12):2286-2293.

Rebmann

, Coll

; 2009. APIC Emergency Preparedness Committee. Infection prevention in points of dispensing. Am J Infect Control, 2009; 37(9):695-702.

Cities Readiness Initiative. Centers for Disease Control and Prevention website. Updated September 11, 2013. http://www.cdc.gov/phpr/stockpile/cri/. Accessed June 20, 2014.

Appendix 2: Cities Readiness Initiative Technical Assistance Review Scores by Metropolitan Statistical Area and Individual Planning Jurisdictions. Centers for Disease Control and Prevention website. http://www.Sept_Update_Report_9-13-2011-Final-appendix2.pdf.

Appendix 6: Cities Readiness Initiative Technical Assistance Review Scores for 2007-2008. Centers for Disease Control and Prevention website. Updated September 21, 2010. http://www.bt.cdc.gov/publications/2010phprep/appendix/appendix6.asp. Accessed June 20, 2014.

Lee

, Chen

, Pietz

, Benecke

. Modeling and optimizing the public-health infrastructure for emergency response. Interfaces, 2009; 39(5):476-490.

Agocs

, Fitzgerald

, Alles

, et al. Field testing a head-of-household method to dispense antibiotics. Biosecur Bioterror, 2007; 5(3):255-267.

Hupert

, Xiong

, King

, et al. Uncertainty and operational considerations in mass prophylaxis workforce planning. Disaster Med Public Health Prep, 2009; 3(Suppl 2):S121-S131.

10.

Lee

, Chen

, Pietz

, Benecke

. Disease propagation analysis and mitigation strategies for effective mass dispensing. AMIA Annual Symposium proceedings; AMIA Symposium; 2010:427-431.

11.

Baccam

, Willauer

, Krometis

, Ma

, Sen

, Boechler

. Mass prophylaxis dispensing concerns: traffic and public access to PODs. Biosecur Bioterror, 2011; 9(2):139-151.

12.

Everett

, Potter

, Wheaton

, Gleason

, Brown

, Lee

. Geospatial analytics to evaluate point-of-dispensing sites for mass immunizations in Allegheny County, Pennsylvania. J Public Health Manag Pract, 2013; 19(Suppl 2):S31-S36.

13.

Spitzer

, Hupert

, Duckart

, Xiong

. Operational evaluation of high-throughput community-based mass prophylaxis using just-in-time training. Public Health Rep, 2007; 122(5):584-591.

14.

Stergachis

, Wetmore

, Pennylegion

, et al. Evaluation of a mass dispensing exercise in a Cities Readiness Initiative setting. Am J Health Syst Pharm, 2007; 64(3):285-293.

15.

Durante

, Melchreit

, Sullivan

, Degutis

. Connecticut competency-based point of dispensing worker training needs assessment. Disaster Med Public Health Prep, 2010; 4(4):306-311.

16.

Balbi

, Calcagni

. Effects of the design and practice of points-of-dispensing on patient-reported satisfaction: municipal H1N1 clinics in Rhode Island. Disaster Med Public Health Prep, 2011; 5(2):106-111.

17.

Buehler

, Whitney

, Berkelman

. Business and public health collaboration for emergency preparedness in Georgia: a case study. BMC Public Health, 2006; 6:285.

18.

Young

. Pharmacist's software design aids mass dispensing clinics. Am J Health Syst Pharm, 2006; 63(5):400-402.

19.

Doyle-Moss

. Establishing a prophylactic drug dispensing clinic: the University of Rhode Island model. Disaster Manage Response, 2007; 5(3):87-92.

20.

Veltri

, Yaghdjian

, Morgan-Joseph

, Prlesi

, Rudnick

. Hospital emergency preparedness: Push-POD operation and pharmacists as immunizers. J Am Pharm Assoc, 2012; 52(1):81-85.

21.

Dillman

, Smyth

, Christian

. Internet, Mail, and Mixed-Mode Surveys: The Tailored Design Method. 3d ed. Hoboken, NJ: Wiley; 2008.

22.

Bravata

, Zaric

, Holty

, et al. Reducing mortality from anthrax bioterrorism: strategies for stockpiling and dispensing medical and pharmaceutical supplies. Biosecur Bioterror, 2006; 4(3):244-262.