HHS - Fundamentals of Dispensing/Vaccination Clinic Design - Community-Based Mass Prophylaxis

[sharon sanders]

<table summary="This table gives the layout format of the bread crumb area and the center content area." border="0" cellpadding="0" cellspacing="0" width="100%"><tbody><tr><td height="30px">Community-Based Mass Prophylaxis </td> </tr> <tr> <td> Fundamentals of Dispensing/Vaccination Clinic Design

Whenever possible, planners should develop a single generic DVC design (including floor plan and patient flow plan) for use throughout their community. Having a single master DVC plan will simplify training and improve interoperability of staff, increase the ability to forecast patient flow and therefore resupply needs, and maintain flexibility to uniformly alter all DVCs in response to unfolding events.
DVC design may range from very simple to extremely complex, depending on the nature of the event, requirements of the response, and time frame for action. This Section covers general DVC functions, important factors that affect the efficiency and accuracy of mass dispensing operations, and methods for estimating patient flow through a single DVC and network of DVCs. This planning guide may be used in conjunction with a computer-based interactive spreadsheet model (the Bioterrorism and Epidemic Outbreak Response Model, or BERM) that allows users to calculate the number of personnel required to operate DVCs using either an antibiotic dispensing or a vaccination design.⁸⁴ The model allows the user to study the relationship between various population- and attack-related variables and DVC staffing requirements.
3. DVC Concept of Operation

DVC operations may be divided into core and support (or "non-core") functions. Core functions include all processes that directly facilitate the dispensing of drugs and vaccines and almost always involve one-on-one interaction between staff and patients. Exceptions include distribution of forms and patient briefings, in which one staff member may interact with a large number of people at once. Core stations are sites within the DVC where core functions take place. Support functions include all the processes that take place in the DVC that are critical in supporting the core stations. These tasks range from medication or vaccine resupply to security to command and control. Core and support staff are equally important to the overall success of the DVC plan.
A. Core Functions

Core functions are also called "Operations" when using Incident Management terminology.

1. Greeting Greeters have the dual role of directing people into the DVC and also screening the crowd (visually and/or via direct questions) for obviously ill patients who require immediate medical evaluation (i.e., skip steps (b) and (c), below) or individuals at higher risk for exposure (i.e., if time and location of exposure is known).
2. Form Distribution Most DVC plans will include some type of data collection using forms filled out by patients.^40,56,71,80 These forms serve multiple purposes, including guiding triage (e.g., by asking all those who checked off a certain box or set of boxes to proceed to medical or mental health evaluation) and facilitating followup (e.g., by asking for contact information).
3. Triage Triage involves using patient-completed forms (see (b), above) or protocol-based questions to identify people requiring medical evaluation and/or, depending on DVC design, mental health evaluation.^29,70,81,82 People who screen negative at triage can proceed directly to the dispensing station. Since it is protocol driven, triage does not necessarily need to be performed by health care professionals.
4. Medical Evaluation Acutely symptomatic individuals or those who have symptoms suggestive of illness due to the attack may require evaluation by health care professionals, preferably staff who are experienced in evaluation and stabilization of sick patients (e.g., paramedics and emergency department nurses as well as physicians).⁴³ Depending on time, resource availability, and linkages to health care facilities, medical treatment at DVCs may include initiation of antibiotics and other interventions prior to transport for seriously ill patients.
5. Transportation Assistance Patients deemed seriously or critically ill will require assistance getting into vehicles (e.g., ambulances, buses, vans) that can provide transportation to tertiary care or other higher level care facilities. The extent and complexity of potential DVC-based treatments for these patients (e.g., whether to start intravenous antibiotics for a suspected anthrax victim) will depend on the estimated time needed for transfer from the DVC to the definitive care facility.⁵⁰
6. Mental Health Evaluation If mental health activities are located in the DVC, they may vary from simple evaluation and treatment of acute panic and stress reactions to more extensive counseling for grief and depressive symptoms in the aftermath of an attack.⁷² The type, extent, and proper location of mental health evaluation in the DVC will vary based on details of the disease outbreak, time frame for response, space, and availability of trained mental health practitioners who can participate in DVC activities.⁸³ More elaborate DVC plans may call for separation of acute and non-acute mental health stations.
7. Briefing Briefings may improve compliance with medical regimens, decrease mental stress due to the event, and in some cases may be required by regulation (e.g., with Investigational New Drug [IND] protocols).⁵ Additionally, briefings may provide information about referrals to off-site counseling. Briefings should take advantage of the standardization and flexibility provided by pre-taped video/audio presentations, although these require additional resources and technical support (e.g., translation into multiple languages). The size and number of briefing rooms and the duration of briefings may limit the maximum rate of patient flow through a DVC, as described in the Summary of Patient Flow Diagrams below.
8. Drug Triage (Pharmacotherapeutic Evaluation) The purpose of drug triage is to rapidly identify people who require any drug regimen other than the standard drug type and dose (e.g., patients requiring a medication other than adult dose doxycycline for anthrax prophylaxis). Drug triage questions may be part of the written information form filled out at entry to the DVC (see (b), above) or asked of patients arriving in the dispensing area.^56,71 Families with children may be identified at drug triage for further assistance to determine pediatric dosages.
9. Dispensing or Vaccination (Express vs. Assisted) Patients may be directed to a single dispensing station that has staff available for pharmacotherapeutic consultation or, alternatively, to 1 of 2 dispensing areas designed to handle uncomplicated ("Express") or complicated ("Assisted") dispensing cases. Large DVCs with sufficient staff may achieve greater efficiency by establishing a separate dispensing line for people whose drug triage evaluation suggests the need for dose modification or an alternative drug. Assistance may include determining the correct type and dose of antibiotics for adults with reduced kidney function or medication allergies, or for children based on age, weight, and/or height as well as history of allergic reactions. Communities may opt to allow one person (e.g., the head of a household, the spouse or friend of someone who has a mobility impairment, etc.) to pick up medications for persons other than themselves; these cases may take additional time and should be directed to the "Assisted" dispensing area.
   All DVC plans should include some mechanism for quality assurance, such as designating a pharmacist or other health care professional to monitor the accuracy with which antibiotics or vaccines are being dispensed.
10. Form Collection and Exit Although patient information forms may have been used for triage purposes inside the DVC, exit staff may still be needed to check the accuracy of contact information as patients leave. Additionally, exit staff may be able to provide details of followup care, reinforce compliance messages, and even perform "spot check" for quality assurance (e.g., checking whether patients are receiving the correct medications).

B. Support Functions

Support functions are also called "Logistics" when using Incident Management terminology.

1. Drug/Vaccine Inventory, Preparation, and/or Re-supply CDC SNS supplies both unit-of-use antibiotic regimen bottles and bulk supply of antibiotics and vaccines that can be repackaged at central materiel distribution centers such as the Receipt, Store, and Stage (RSS) site. Inventory support staff at each DVC will be responsible for restocking dispensing stations with ready-to-use doses of antibiotics and/or reconstituted vaccines. Re-supply staff should receive training in cold-chain techniques and proper use of mobile cold storage devices (e.g., Vaxicools).
2. Patient Traffic Directors DVC sites and entry points must be externally identified using appropriate signage (e.g., using relevant languages in areas with non-English speaking populations). Inside the DVCs, personnel are needed to help direct patients from station to station and to assist in managing crowds when bottlenecks form.⁴⁰
3. Data Entry Data entry staff may be needed to transfer patient information from written forms to computerized databases to facilitate epidemiological investigation of the attack, assessment of the mass prophylaxis campaign, and followup care for treated patients.
4. Translation services Planning for translation services includes ethnographic evaluation of covered populations and identification of personnel who will be available to provide appropriate translation services under crisis conditions.
5. Communications/Information Technology Support Secure and reliable communication links inside each DVC, between different DVCs in a given community, and from DVCs to a central Command and Control center are critical to the successful implementation of any DVC plan. In addition, key DVC operations including inventory management and data entry may require computer support and secure Internet access for Web-based services.
6. Food Service Local factors will determine whether DVCs can support on-site food preparation and/or distribution for staff. If not, planners will need to find alternative means of providing meals, snacks, and beverages during DVC activations.
7. Facilities Maintenance The need for facilities services will depend in part on the extent of on-site food service and/or preparation, but extends to maintenance of toilet facilities for both staff and patients, as well as facilities for staff trained in proper disposal of contaminated medical waste.
8. Security DVC security includes maintaining crowd control both outside and inside the DVC as well as securing medication stocks, confidential patient information, and communication and computer equipment inside the DVC. Additionally, security staff are needed to ensure the personal safety of DVC staff. While the past several years have seen increasing attention by the public health community to these security needs for mass prophylaxis campaigns, there is as yet no consensus on the number of individuals required to achieve these goals. One approximation (derived from live exercises at several U.S. sites with the SNS training package [the "TED"] is that for every four to five core staff assigned to the DVC, there should be approximately one security staff.
9. Managers DVC managers are considered support staff because they do not have direct patient care responsibilities. However, the command staff may be thought of as a separate work group. Section Four of these Guidelines provides a description of a model command structure for DVC operations.

Return to Contents
4. DVC Design and Patient Flow

This section reviews a number of possible clinic layouts (also called patient flow diagrams) to help local DVC planners develop a DVC-based mass prophylaxis plan that is appropriate for local needs, including population size, staff resources, and response time frame.
A. Patient Flow Plans

One of the first tasks for mass prophylaxis planners is to determine which core stations will be included in the basic DVC plan for their community and what the physical arrangement of those stations will be in each DVC. The choice of core stations for a given DVC design depends on a number of factors, including the target patient flow rate (larger, more complex station arrangements will inevitably lead to slower patient throughput), the availability of personnel to adequately staff those stations, and the physical space for DVC activities. The patient flow diagrams shown below illustrate DVC designs of increasing complexity. Note that while increasing complexity generally requires increased staff and time, it does allow for valuable (and perhaps necessary) additional processes, such as data collection.

1. Diagram 1: Basic High Flow Model
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">Very high patient flow rates would require streamlining of many DVC functions in order to decrease the total processing time of the average patient to the minimum needed to accurately dispense medications and/or vaccines. The most basic "high-flow" DVC design is pictured at right. It consists of only 4 core stations (triage, medical evaluation, transport assistance, and drug dispensing). Note that this DVC does not include stations that may be required (e.g., briefings) or considered useful given the nature of the event (e.g., distribution and collection of data collection forms).
   </td> <td align="center" width="70%"></td> </tr> </tbody></table>
2. Diagram 2: High-Flow with Entry Screening
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">This diagram shows the addition of a greeting and screening station to the basic 4-component plan. This would facilitate rapid identification and isolation of symptomatic patients in the setting of an outbreak of a contagious illness or in the case of a rapidly fatal disease such as anthrax. This floor plan was used to attain patient flow rates of over 1,000 per hour in the high-flow antibiotic dispensing exercise called Operation TriPOD in New York City (May 22, 2002).
   </td> <td align="center" width="70%"></td> </tr> </tbody></table>
3. Diagram 3: Form Distribution and Collection
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">This layout includes stations for form distribution prior to triage and form collection directly prior to exit. In addition to allowing data collection for epidemiological and medication followup purposes, patient forms may be designed to facilitate triage and medical evaluation. Forms with "check-off" boxes listing medical contraindications and potential drug interactions may eliminate the need for repetition at subsequent stations. Staff may annotate these forms as a convenient way of communicating with other (downstream) DVC staff regarding patient management (e.g., a patient who checks off a potentially conflicting medication may have that box highlighted at triage in order to let the specialists in the medical evaluation station know why the patient had been sent over for further management).
   </td> <td align="center" width="70%"></td> </tr> </tbody></table>
4. Diagram 4: Mental Health Evaluation
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">This layout includes a mental health evaluation station located after triage and before drug dispensing. Patients with acute and/or debilitating symptoms of panic, fear, etc. in response to a large-scale disease outbreak or bioterrorism event may not be in a position to comprehend and follow even straightforward medical regimens. For this subset of patients, mental health crisis counseling may improve their ability to comprehend medical instructions and successfully utilize prophylactic medications.
   </td> <td align="center" width="70%"></td> </tr> </tbody></table>
5. Diagrams 5A and 5B: Briefing
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">Here, patient briefing is included as a separate station within the DVC. Briefings may be mandated (e.g., in settings where informed consent is required for administration of an investigational drug) or deemed useful for improving patient adherence to medication regimens and/or for calming fears about a bioterrorist event.
   Briefing stations may be situated before or after triage, depending on a variety of factors such as the importance of rapidly identifying patients with symptoms suggestive of attack-related illness. If briefing is mandatory and the likelihood of symptomatic disease is low (e.g., in a pre-event mass vaccination campaign for smallpox), then the briefing station should be situated to capture a large proportion of patient flow (Diagram 5A).
   </td> <td align="center" width="70%"> Diagram 5A
   </td> </tr> <tr valign="top"> <td width="30%">Conversely, if early detection is a priority (e.g., as with inhalational anthrax, where on-site treatment of suspected cases may involve rapid administration of antibiotics), then briefing should be limited to patients who are asymptomatic or those whose symptoms have been fully evaluated and deemed not to require immediate intervention beyond routine prophylaxis (Diagram 5B).
   </td> <td align="center" width="70%">Diagram 5B
   </td> </tr> </tbody></table>
6. Diagrams 6A and 6B: Express Dispensing
   <table border="0" width="100%"> <tbody><tr valign="top"> <td width="30%">The 2001 Capital region response to the anthrax attacks validated the notion of creating a separate drug triage area and "express drug" line to facilitate rapid dispensing of antibiotics to asymptomatic persons with no drug contraindications.^56,71 The justification for adding this additional step to overall patient flow is that pre-identification of complicated cases (i.e., those requiring consultation with a pharmacist or physician prior to receiving medication) will speed overall processing time in the dispensing area by not clogging all dispensing points with potentially long delays. This will allow better utilization of specialist staff at designated "assisted drug dispensing" areas.
   </td> <td align="center" width="70%">Diagram 6A
   </td> </tr> <tr valign="top"> <td width="30%">Diagrams 6A and 6B show that these changes in drug dispensing operations are compatible with both DVC plans covered in Diagram 5. Drug triage may be facilitated by appropriately designed forms that allow easy identification of adult and pediatric patients whose age, medical conditions, or preexisting medications necessitate dose or drug adjustment. For example, check boxes may be aligned on the form to highlight positive responses upon quick (i.e., less than 10 second) visual scanning.
   </td> <td align="center" width="70%">Diagram 6B
   </td> </tr> </tbody></table>
7. Summary of Patient Flow Diagrams These plans show patient flow paths through a variety of DVC designs and illustrate the potential complexity and variability of DVC station layouts. While such plans may be adopted "as is," they are intended to serve more as a starting point for local planning than as final designs.

B. Bottlenecks

In order to minimize bottlenecks in patient flow, a DVC should be designed with a floor plan that prioritizes the expected transit pattern of the following 3 patient groups:

• Uncomplicated cases: individuals who are asymptomatic and/or unexposed (e.g., in the case of smallpox) and have no pre-existing conditions requiring specialized dispensing/prophylaxis regimens.
• Specialized-care cases: individuals with mild, non-outbreak related symptoms and/or pre-existing conditions or possible contraindications requiring specialized dispensing.
• Seriously ill individuals: identified upon entry, at triage, or at medical evaluation with disabling and/or life-threatening symptoms needing immediate medical attention and transportation to a health care facility.

Since the majority of patients will most likely fall into the first group, even small errors in DVC layout for this group may adversely impact overall patient flow rate. In this case, a small delay may not necessarily affect outcomes for the individual patient, but it may affect the operating efficiency of the mass prophylaxis campaign as a whole. Specialized-care cases will constitute a smaller group but require increased time and/or resources. Small, infrequent delays will have less of an effect on DVC operating efficiency, but a constant delay eventually will lead to the formation of bottlenecks and may adversely affect clinic operations. Seriously ill individuals will constitute a small minority of patients processed by the DVC in all but the most dire of scenarios. Consequently, design flaws in medical evaluation and emergency transportation areas will not significantly affect overall DVC operating efficiency. However, these delays may adversely affect the care of individual patients. Ideally, the routing of patients from entry to exit should be as direct as possible for all 3 groups, with seriously ill individuals traveling the shortest distance from entry to transport.
C. Making Things Flow: The Highway Traffic Analogy

Highway automobile traffic provides a useful analogy for understanding the factors that can interfere with efficient patient flow in the DVC. There are 4 basic causes of traffic jams: merges (e.g., of 2 lanes into one), surges (e.g., rush hour), accidents (which can block both the affected lane and adjacent ones due to spectators), and tolls or other features of the road that slow the speed of all cars (e.g., bridges).
Merges are easily managed at the DVC design stage: DVC stations and staffing should be arranged to avoid line merges, especially in areas seeing the majority of patient flow, since these will lead to backups and queues that may interfere with DVC support functions (e.g., re-supply). Backups due to surges in arrivals are more difficult to solve at the DVC design stage, since these may require shifting or adding staff or shortening processing times (e.g., shortening patient information briefings or forms) to handle increased patient flow. DVCs that are designed to process a certain patient volume without bottlenecks should therefore have a staff member assigned to monitor and report incoming patient flow in order to expedite these types of adjustments before DVC operations are affected by any sustained surge.
The DVC equivalent of a traffic accident could be a patient who is seriously ill on arrival and requires immediate assistance at the entrance to the DVC (thus blocking traffic) or any person or group who requires special assistance (e.g., a large family, a disruptive individual, etc.). Contingency plans to manage these and other unexpected events should be developed with the goal of identifying DVC locations and staff to isolate the affected individual(s) from the main flow of patients in order to minimize bottlenecks that compromise overall DVC efficiency.
The DVC analogy to the toll booth is the core station (e.g., triage, drug dispensing) that requires a certain processing time for each patient. The length of this processing "delay" will be determined by the length of the protocols and forms and by the waiting time at each station; for patient information stations, this delay is determined by the duration of briefings plus any question and answer period. Designers of briefing scripts and station protocols need to balance thoroughness with efficiency and economy in order to maintain smooth patient flow. Ideally, planners should also pre-designate ways in which these processes can be shortened if bottlenecks do occur and lines begin to form (e.g., pre-identifying portions of a briefing that can be eliminated or provided by alternate communications modalities like radio or television).
D. Planning For Breakdowns: What to Do About Lines (Queues)?

Lines (technically called "queues") may form inside the DVC whenever the arrival rate exceeds the processing rate. In general, this may result from a "surge" in new arrivals (i.e., above the baseline for which staff has been allocated), from a decrease in staff below what is needed to process the baseline arrival rate, or from an increase in the time needed to process individuals in the DVC. This is so important and seemingly obvious that it bears repeating: bottlenecks and their resultant queues arise, in general, from only 3 causes:

• Too many patients.
• Too few staff.
• Too much to do.

DVC floor plans should take into account the possibility that queues will form at each station as part of the natural variation in arrivals, staffing, and processing times. The amount of space allocated for these queues cannot be accurately predicted prior to running of the DVC. One general rule of thumb is that for 2 stations with identical arrival rates but different processing times (and therefore different numbers of staff assigned for baseline operation), the removal of a staff member at the quicker station will lead to more rapid development of a queue than removal of a staff member at the slower station.
The reason for this is that each staff member at the quick station processes more individual patients per unit time than staff at the slow station. Thus the relative loss of each additional staff member from the quick station will have a greater impact on maximum queue size (though not necessarily on queue duration, if the problem is rectified by addition of a staff member to the station) than from a slow station. In short, staff reductions at quick stations have the potential to produce large (though potentially short-lived) queues, while staff reductions at slow stations may produce smaller queues of longer duration.
E. Additional Factors Affecting the Efficiency Of DVC Operations

Several factors in addition to the design of the DVC floor plan and station placement may also affect the efficiency of DVC operations. These include:

1. Accessibility Families are an important and often overlooked group with unique accessibility issues from the standpoint of DVC design, since they will likely travel as a single large group. Therefore, DVC layouts should be flexibly configured to accommodate both individuals and large family groups as the "patient." Plans that incorporate inaccessible locations or transit routes (e.g., DVCs placed at sites with stairs or other features that restrict movement of people with assistive mobility devices) may require additional personnel to provide assistance to affected individuals.
2. Translation services Lack of adequate translators may seriously impede processing of non-English speaking patients and therefore cause bottlenecks that may slow overall DVC operations. Communities with limited personnel for translation may want to pre-print or pre-record material for specific response scenarios in appropriate languages.
3. Dispute resolution DVC policies regarding triage and dispensing may lead to disagreements between staff and patients. In order to prevent these disagreements from causing bottlenecks at the station where they occur, DVC planners should designate a specific location within the DVC and an operational protocol for mitigating anticipated disputes (e.g., establishing guidelines governing whether a single individual can be given medications for multiple other individuals).
4. Geography Population density and geographic location may influence the development and implementation of prophylaxis plans. For example, finding appropriate locations for DVC operations in rural areas requires consideration of unique transportation, resupply, and communication issues compared to those encountered in urban settings.

<table border="1" cellpadding="8" cellspacing="0" width="90%"> <tbody><tr> <td>News from the field: Lessons learned about clinic dynamics at the San Francisco smallpox vaccination clinic exercise, June 17, 2003 (processing approximately 200 people per hour):

1. Make a single flow control point outside the clinic to regulate patient arrivals.
2. Patient questions (about symptoms and clinic operations) can bottleneck the greeting station, causing backups out the front door. Refer these to briefings.
3. Make sure signage is clear, visible, multilingual, and consistent in message.
4. Make sure all clinic staff give consistent directions.
5. Long queues can easily co-mingle, causing confusion and "missed" stations.

Source: Vaccination Ventures: Explanation and Outcomes of a Mass Smallpox Vaccination Clinic Exercise held June 17, 2003 by the San Francisco Department of Public Health. Prepared by Amy E. Pine, M.P.H., Immunization Program Manager Communicable Disease Prevention Unit, San Francisco Department of Public Health. Online at https://www.dph.sf.ca.us/Reports/Jun...lJune17Rpt.pdf
</td> </tr> </tbody></table> Return to Contents
5. DVC Calculations: Estimating Clinic and Staff Numbers

The following section presents an overview of one method for calculating the number of DVCs necessary for a mass prophylaxis campaign and the number of staff needed to operate a given DVC. Planners interested in the mathematical details behind this approach and in performing more detailed calculations are referred to the accompanying computer model, the Bioterrorism and Epidemic Outbreak Response Model (BERM) and its technical appendix (which is reproduced as an appendix to this report).
A. Number of DVCs

Planners can determine how many DVCs they will use for a mass prophylaxis campaign in 1 of 3 ways:

1. By determining the total number of sites available in their community. This would require fitting all necessary DVC activities into those pre-selected sites. An example of this was the 1995 Minnesota meningococcal meningitis vaccination campaign that took place at a single site with approximately 300 staff.³⁶
   Advantage: Fits the mass prophylaxis plan to existing community structures.
   Disadvantage: May result in mismatch between sites and population/available staff size.
2. By determining the total number of staff needed to operate a clinic (e.g., deciding that, for security or other reasons, each DVC should have no more than 100 core staff operating at a given time). Dividing the total staff required for a prophylaxis campaign (determined using the BERM model) by the per-DVC staff size gives the number of DVCs that should be established.
   Advantage: Ease of planning (one size fits all).
   Disadvantage: Requires estimate of total staff needed for campaign.
3. By estimating the maximum number of patients that could be processed at a standard DVC (that is, the patient flow rate). These estimates can be derived from a number of sources: nationally publicized exercises (e.g., ranging from roughly 200 patients per hour for a single smallpox vaccination clinic [San Francisco, 2003] to up to 1,200 patients per hour for a single high-flow anthrax antibiotic dispensing clinic [New York City, 2002]); from previous local experience with influenza vaccination and other public health clinics; or from calculations based on the duration of patient briefings at the proposed DVCs, as described below. The number of DVCs needed is then calculated by dividing the community-wide patient flow rate (which is just the number of patients needing prophylaxis or vaccination divided by the number of days or hours allotted for the campaign) by the per-clinic patient flow rate.
   Advantage: Fits the mass prophylaxis plan to population size and response time.
   Disadvantage: May be difficult to estimate per-DVC patient processing rate.

The following scenario illustrates this last method using the length of briefings to determine DVC flow rate, and therefore overall number of DVCs:

1. Calculating per-DVC patient processing rate:
   City A has a population of one million residents, all of whom require prophylaxis in 6 days or less using DVCs that have mandatory briefings for all patients. Planners have decided that each DVC in City A will have three briefing areas capable of handling 50 patients apiece and that each briefing (with a question and answer period) will take 20 minutes. Since these briefings are mandatory, they represent the critical rate-determining step for patient flow: assuming that there are sufficient patients to fill every briefing, each DVC will have a maximum patient flow rate of
   
   3 briefing rooms x
   50 patients per briefing x
   20 min per briefing = a 3 briefings per room per hour
   = 450 patients per hour, or 8 patients per minute.
   
   This means that for the proposed DVC to process patients without bottlenecks, it must be designed to handle eight patients per minute from entry to exit. If not, then either briefings will go unfilled (because patients cannot get to them) or stations downstream from the briefings will be overwhelmed and will back up (when patients emerge from the briefing rooms).
   The reader can create a basic spreadsheet model to calculate the maximum patient flow rate for any DVC that has mandatory briefings. This model lets you see the effects of varying the number of briefing rooms, the number of patients per briefing, and the time needed for each briefing.
   Using any standard spreadsheet program, create the following fields:
   <table border="1" cellpadding="5" cellspacing="0" width="90%"> <tbody><tr> <th scope="col">Name</th> <th scope="col">Formula</th> <th scope="col">Example</th> </tr> <tr> <td scope="row">Number of briefing rooms</td> <td>=A</td> <td>e.g., 3</td> </tr> <tr> <td scope="row">Number of patients/briefing</td> <td>=B</td> <td>e.g., 50</td> </tr> <tr> <td scope="row">Duration of each briefing (minutes)</td> <td>=C</td> <td>e.g., 20 minutes</td> </tr> <tr> <td scope="row">Maximum patient flow per clinic given these parameters (per minute)</td> <td>=(AxB)?C=D</td> <td>e.g., 7.5 patients per minute</td> </tr> <tr> <td scope="row">Maximum patient flow per clinic given these parameters (per hour)</td> <td>=(AxBx60)?C=E</td> <td>e.g., 450 patients per hour</td> </tr> </tbody></table> For comparison, recent vaccination campaigns against meningococcal meningitis in Alberta, Canada and Minnesota achieved processing times of between 2.7 and 13 patients per minute per DVC.^85,86

1. Calculating community-wide flow rate:
   Next, the rate of prophylaxis for the entire population must be calculated. For City A, the overall rate of prophylaxis for the entire population is one million divided by the six days available for treatment, or 116 patients per minute. If each DVC can process 8 patients per minute (calculated above), then 14 DVCs are needed to provide prophylaxis to the entire community in the allotted time (assuming 24-hour operation of each DVC). In contrast, if DVC processing is limited to only 2.7 patients per minute (as was seen in the Alberta, Canada meningitis vaccination campaign), then 43 DVCs are needed to carry out community-wide prophylaxis in the specified time frame of six days. To model these calculations, the reader may continue to build the spreadsheet as follows:
   <table border="1" cellpadding="5" cellspacing="0" width="90%"> <tbody><tr> <th scope="col">Name</th> <th scope="col">Formula</th> <th scope="col">Example</th> </tr> <tr> <td scope="row">Population size</td> <td>=F</td> <td>e.g., 1,000,000</td> </tr> <tr> <td scope="row">Number of days for prophylaxis</td> <td>=G</td> <td>e.g., 3</td> </tr> <tr> <td scope="row">Community-wide patient flow rate (patients per minute)</td> <td>=F?G?24?60=H</td> <td>e.g., 231 patients per minute</td> </tr> <tr> <td scope="row">Community-wide patient flow rate (per hour)</td> <td>=F?G?24=I</td> <td>e.g., 13,889 patients per hour</td> </tr> <tr> <td scope="row">Number of clinics required to achieve prophylaxis goal in allotted time</td> <td>=H?D
   or = I?E</td> <td>e.g., 30.8, which rounds up to 31 clinics</td> </tr> </tbody></table> The estimated number of DVCs required to complete a prophylaxis campaign is one of the most critical calculations in all of mass prophylaxis planning. There are considerable logistical differences in setting up, staffing, and running a handful compared to several dozen DVCs. Every local DVC planning team should attempt to determine the number of DVCs required for community-wide prophylaxis under different response scenarios.

B. Number of Staff

One of the most difficult features of DVC planning is determining how many staff would be needed to work at a given station within a DVC. The accompanying Bioterrorism and Epidemic Outbreak Response Model (BERM) allows calculations of the number of staff needed to carry out a prophylaxis campaign using 2 different pre-specified DVC designs, one for antibiotic dispensing and another for vaccination.⁸⁴ The calculations underlying these estimates are described in detail in the model's Technical Appendix, but deserve comment here as well. The main concept underlying these calculations is the notion that every DVC should be capable of what is called "steady-state operation." This means that every clinic should be capable of operating at full capacity without developing progressively larger bottlenecks, which would show up as queues.
In other words, the operational goal of any DVC should be that, at minimum, it does not continuously back up to the point of complete shutdown. A DVC operating at this "steady-state" has achieved a balance between the number of staff, the number of patients, and the time needed for those staff to process those patients such that there is no increase in bottlenecks or queues. While this may never actually occur during real-life operations (due to a variety of factors such as unpredictable surge arrivals, etc.), all DVCs should, at a minimum, be designed to achieve steady-state operation.
Fortunately, it is possible to calculate the number of staff needed to run a system that is operating in a steady-state manner. These calculations can provide planners with 2 important sets of data: either estimates of the minimum number of staff needed to process patients at a given rate of arrival and for a given processing time, or estimates of the maximum processing time permitted for a given number of staff to process patients at a given rate of arrival. The next section is a summary of the Technical Appendix of the BERM model for readers interested in how these calculations are carried out.
C. The Bioterrorism and Epidemic Outbreak Response Model (BERM)

The BERM model was created by researchers at Weill Medical College of Cornell University in 2003 under contract to the Department of Health and Human Services, Agency for Healthcare Research and Quality (AHRQ).^84,87 In contrast to previous spreadsheet models of bioterrorism response that have focused on resources needed for a military medical response, this model is designed for civilian response to bioterrorism and epidemic outbreaks requiring mass prophylaxis.⁸⁸ Its goal is to assist public health and emergency management planners to create customized community mass prophylaxis plans using a DVC-based dispensing approach.
Model inputs include community population size, time frame for response, characteristics of DVC operations (e.g., hours of operation, number of shifts, rest-time for workers), rate of patient processing at each DVC (calculated using 1 of the 3 methods described above) type of operation (pill dispensing or vaccination), operational setting (e.g., pre-event in which only a small proportion of patients will need medical attention vs. large-scale event in which up to 20 percent of patients will be symptomatic on arrival to the DVC), and DVC processing speed (baseline, slow, or fast). Depending on the type of operation chosen, the model runs off 1 of 2 generic DVC layouts of either an antibiotic dispensing clinic (similar to those set up in Washington, D.C. after the 2001 anthrax attacks) or a vaccination clinic (similar to the CDC smallpox vaccination model, but modified by the Weill/Cornell researchers). These clinic layouts represent a composite of several published plans, including those of the CDC, U.S. Public Health Service, Central Florida Regional Domestic Security Task Force, and the California Emergency Medical Services Authority.^80,89-91 Each layout defines the number and type of stations where core staff are needed. Additionally, users can customize the model by inputting the number and type of support staff needed for operation of a single DVC.
Once this information is entered into the model, it calculates what is needed for operating a multi-DVC-based, community-wide mass prophylaxis in the specified time frame. Specifically, it gives estimates of the number of DVCs needed to treat the entire community as well as the number and type of core and support staff at each DVC, for each shift, and for the entire prophylaxis operation. If the model results indicate that more staff are needed than are available in the community, then users can easily re-calculate how long it would take to cover the entire population using the actual number of staff on hand.
The ultimate purpose of the BERM is to allow planners to "think with numbers" as they go about formulating realistic mass antibiotic dispensing and vaccination contingency plans for their target populations. Using a model that provides numerical estimates forces critical examination of assumptions about prophylaxis clinic design and about the availability of human and materiel resources. Estimates derived from this model should be viewed as one type of data among many that may be useful for planning (other data might include previous local experience with immunization campaigns, or results of training exercises for bioterrorism response).
As with any model, the accuracy of the numerical estimates provided by this program depends on the quality of the underlying data on which they are based. For example, the station-specific processing time estimates used in this model have a large impact on outcomes (to demonstrate this, observe the change in overall staffing estimates for a given scenario under slow, baseline, and fast processing times). In order to improve ease-of-use, the model provides these three pre-set choices for processing times and three pre-set choices for disease prevalence (pre-event, small-scale event, or large-scale event). The trade-off here is with "realism" of the outputs, since real-world events rarely conform to such neat categories. However, examining how community-wide prophylaxis plans would need to adapt to these nine (3 x 3) scenarios may go a long way to exposing heretofore unidentified stresses on prophylaxis plans already in place or under development. (Additionally, for those who are interested in finer-grained customization of BERM outputs, the model includes a separate page for altering and customizing every element of these baseline scenarios.)




http://www.ahrq.gov/research/cbmprophyl/cbmpgde2.htm



</td></tr></tbody></table>