Comparison of Pharmacists and Primary Care Providers as Immunizers
Pharmacist-administered vaccination services were found to be more efficient, productive, and cost-effective than primary care visits or mass vaccination clinics.
A major success of modern public health has been the near-eradication in the United States of most vaccine-preventable diseases. Contained within this public health success are significant operational challenges to an already overtaxed primary healthcare infrastructure. Currently, vaccines are licensed for 27 diseases,1 with a number of new vaccines at or near licensure and new populations recommended to receive already-licensed vaccines.2 By way of example, the Advisory Committee on Immunization Practices (ACIP) recently recommended that all children age 6 months to 18 years—as well as their household contacts and out-of-home caregivers—be immunized against influenza.3 An estimated 83 million children are in this age cohort,4,5 and the 100 million visits required to immunize all these children would consume, at a minimum, 2.9 million person-hours each influenza season.6 That figure does not include the costs and person-hours required for indirect support and administrative services, which our research shows far exceeds the direct service costs.7,8
Primary care physicians are already reeling under present service demands and cost-containment initiatives, to the point that the American College of Physicians has issued a warning that “primary care, the backbone of the nation’s healthcare system, is at grave risk of collapse.”9 Given that vaccination recommendations will further expand over the coming years, it is worth identifying other service providers that have the capacity, competency, and business model to meet this nation’s expanding vaccination requirements.
Pharmacists would appear to be excellent candidates to meet this growing demand. There are approximately 243,000 pharmacists (or 79.5 pharmacists per 100,000 Americans), of whom 62% are employed in the 39,000 chain and 24,500 independent pharmacies throughout both urban and rural America.10,11 Most of them are already coordinating care with the 264,086 primary care providers (or 90 per 100,000 Americans),12 providing the potential to nearly double the nation’s vaccination capacity. Pharmacists are explicitly authorized to immunize in 49 states,13 have effectively boosted influenza vaccination rates during past influenza seasons,14,15 and are trusted by both patients16-19 and primary care physicians.20 Practical guidelines exist for establishing vaccination practices for pharmacists,21,22 and there are national certification programs approved by the Centers for Disease Control and Prevention (CDC) and the American Pharmacists Association.23
However, increasing capacity is not enough. If vaccinations provided by pharmacists are more costly, or if pharmacists are less efficient or less productive than primary care providers, then both the sustainability and total potential increase in capacity are constrained. We used traditional systems engineering techniques to assess the fiscal viability and potentially added capacity pharmacists represent relative to primary care.
METHODS
Design and Study Population
The data set is derived from a larger study examining operational issues surrounding expanded recommendations for influenza vaccinations. Following approval by the University of California, San Diego Institutional Review Board, a convenience sample of 15 ambulatory primary care settings and 11 pharmacies in San Diego was used to collect workflow observations of vaccination services. The primary care and pharmacy sites represented a broad range of organization types, ranging from small Federally Qualified Health Centers and independent pharmacies through multispecialty/multisite for-profit organizations, including national chains and military hospitals. All told, the sites provide more than 143,596 vaccinations to more than 20,000 individuals annually. Characteristics of the sites are shown in
Table 1
.
Primary care observations were conducted during the 2006-2007 influenza season, followed by pharmacy data collection the following influenza season. Observations were encoded using the Observational Checklist of Patient/Client Encounters, a standardized workflow data acquisition tool described in earlier publications.24
Our study observations were conducted with a convenience sample of subjects eligible for influenza and/or other vaccinations. Patients were observed at primary care sites during routinely scheduled primary care appointments (operationally defined as a visit at which any new or current patient arrives at the clinic for a scheduled appointment) and mass influenza vaccination clinics (operationally defined as a visit at which any new or current patient arrives at the clinic without a scheduled appointment with the sole purpose of receiving an influenza vaccination). Client observations at the pharmacies all occurred during nonscheduled, walk-in vaccination encounters.
Data Collection
Data collection began with research staff approaching potential subjects and seeking permission to observe them throughout the course of their encounter. Patients/clients were followed from the moment they entered the clinic module or pharmacy until they exited all services, using the Observational Checklist of Patient/Client Encounters—a 52-item checklist that captures 3 major domains of information: (1) operational conditions (eg, staffing ratios) present at the time of the observation, (2) whether specific activities that are part of the published standards for vaccinations were present (eg, patients/clients being asked about potential allergies to eggs), and (3) the time patients/clients spent during specific stages of the encounter (eg, time spent in registration, time waiting for the provider). Observations included details such as whether a provider/pharmacist simply asked a patient/client about vaccination status or actually discussed the purpose of immunizing against influenza. The combination of data types permitted triangulation of data analysis in which unit costs, productivity, and the maximum patient/client throughput for a given unit of time and labor could be estimated.
Data Analysis
Analysis began with entering the individual workflow observations into an Excel worksheet and examining data for anomalous findings (eg, data entry error where recorded observation times all occurred in the morning with 1 data point in the middle indicating a time in the afternoon). Each anomalous finding was evaluated to determine whether it could be corrected with certainty; if this was not possible, the observation was discarded. Where a specific data field was incomplete for a specific subject observation (eg, whether the receptionist updated patient/client contact information), the specific field from that observation was eliminated from analysis (thus, the numerator/denominator may be different for specific observed activities). However, if several observation points were missing from a particular patient/client observation, the observation was discarded. As a result, 3 of 258 routine primary care, 8 of 349 mass vaccination, and 0 of the 100 pharmacy observations were removed from analysis. Decline-to-participate rates were less than 1% (3 declinations at the primary care sites and 0 at the pharmacies). Observational data for each of the 3 delivery strategies from the primary care and pharmacy sites were analyzed in a nested fashion to estimate unit costs, efficiency, and productivity.
Direct Unit Costs
Direct unit costs were established by adding labor, material, and overhead costs associated with immunizing a single patient/client. Included in labor costs were all personnel who had direct contact with a patient/client. As routinely scheduled exams in primary care offices include, by definition, a wide array of health issues, only the time directly related to vaccination activity was used in the direct unit costs analysis. This compares with mass vaccination and pharmacist visits (observed pharmacists’ vaccinations were all nonscheduled, walk-in visits, in which the only service provided was the vaccination), in which the sole purpose was to immunize; therefore, all the costs were accrued to vaccinations. Time factors were derived either by direct observation (eg, the time required to administer a vaccine) or by calculating the additional time spent in a particular activity by comparing those activities in which a vaccination occurred with those in which they didn’t occur (eg, duration of pre-exam when vaccinations occurred vs duration of pre-exam when vaccinations did not occur). Labor costs were established using the hourly pay for each type of healthcare personnel listed by the US Bureau of Labor Statistics for 2007.25 Similarly, capital or “overhead” costs were established using the Medical Group Management Association’s published data for 2005 (for providers).26,27 Material costs included the average cost of vaccine, syringe, and supplies listed in the CDC Pink Book.28
Productivity
Productivity is operationally defined as the ratio of services rendered to resources used—in our case, the number of eligible patients/clients immunized per unit of labor, overhead, and materials. Productivity inherently identifies optimal ratio of resources—in our case, receptionists, medical assistants, and providers to desired outputs—in this case, appropriately vaccinated patients.
Productivity was calculated using the following formula:
Productivity = (L + C + M) / N,
where L = labor costs, C = capital costs, M = material costs,and N = number of individuals who could be immunizedper hour.
Efficiency
Encounters involve more than simply immunizing patients/clients as quickly and cheaply as possible. Educating patients/clients, assessing for valid contraindications, determining true eligibility, and documenting the encounter all are activities recommended by the CDC/National Vaccination Program, ACIP,29 and by the Joint Commission standards for documenting clinical encounters in ambulatory care settings.30
In general, these activities can be considered quality indicators. For purposes of this study, quality indicators were operationally defined as review and documentation of health history (to assess for eligibility, contraindications, and egg allergies); assessment of basic biometrics (eg, height, weight); observation of the patient/client for allergic reaction; and documentation of either vaccine administered or vaccination delay/refusal when recommended.
When productivity includes quality indicators, the resulting analysis is said to quantify the relative efficiency of the service. Efficiency is essentially how much of a service can be provided correctly for a given set of costs. Efficiency was calculated using the following formula:
(N * PV * PHHR * PD) / (L + C + M)
where N = number of individuals who could be immunized per hour, PV = probability of being immunized, PHHR = probability of health history reviewed, PD = probability of vaccination documented, L = labor costs, C = capital costs, and M = material costs.
RESULTS
A total of 698 subjects were observed in this study (
Table 2
). Compared with patients at primary care or mass vaccination sites, pharmacy subjects were more likely to have received an influenza vaccination the previous year, to have a primary care provider, to perceive influenza as a potentially severe illness, and to seek vaccinations without a physician recommendation.
Unit Costs
Pharmacy vaccination delivery strategies resulted in significantly lower unit costs than the other 2 delivery strategies. In part, this is because of the lower overhead costs, but it also is the result of faster throughput or cycle time. The time devoted to vaccination activities and subsequent unit costs for all 3 strategies are detailed in
Table 3
.
Pharmacists’ vaccinations also proved to be relatively more productive and efficient than the other 2 vaccination strategies. Overall production value—where the lower the production costs, the better—were best for pharmacies at $2.03 compared with $7.67 for vaccinations provided at routinely scheduled primary care office visits. This difference reflects both the lower unit costs and the relatively higher number of clients that can be seen for a given unit of time and labor. The production value analysis is shown in
Table 4
.
Compared with primary care or mass vaccination clinics, pharmacists also proved to be more consistent in following ACIP national guidelines when vaccinating patients/clients. Patients/clients vaccinated at pharmacies were far more likely to be asked about allergies, have their health history reviewed for contraindications, receive a vaccination information sheet, receive a copy of their vaccination record, and have their contact information collected and verified (
Table 5
). The combination of the high rate of protocol compliance and low production costs results in the best (ie, lowest) efficiency value of the 3 strategies analyzed (
Table 6
).
DISCUSSION
In general, expanding the number of individuals eligible for vaccinations will require new strategies and health delivery partnerships. Based on this limited study, pharmacies appear to be well positioned to reach a large number of individuals, within a fiscally sound business model that also adheres to ACIP and CDC vaccination standards. Indeed, pharmacies seem to combine the strengths of both mass vaccination clinics and scheduled appointments at primary care sites. Strikingly, the pharmacies in our study found it in their interest to provide vaccinations even though pharmacists are not paid a professional fee for administering the vaccine, as primary care providers are. Pharmacists are reimbursed for the vaccine product and receive a dispensing fee under the prescription benefit portion of the patient/client’s health plan.
It could be argued that the relative productivity and efficiency advantages seen in pharmacy vaccination services are more a reflection of compliant clientele than internal operational conditions. Demographic data suggested that there were some differences in individuals receiving vaccinations at primary care and pharmacy sites, but that these characteristics would not be expected to account for the differences found in this analysis. Although pharmacy clients were more likely to have received an influenza vaccination the previous year, to have a primary care provider, and to perceive influenza as a potentially severe illness, these are reasons why patients/clients would seek vaccinations, not why the provider/pharmacists would adhere to vaccination standards or have a lower cost-basis for providing the service.
CONCLUSION
Our study suggests that within the context of a changing national healthcare system, pharmacies are well positioned to provide key preventive, non—medication-dispensing services such as vaccinations. We already have seen a transition in which vaccinations provided by public health departments are now provided within a medical setting. Given the already heavily burdened world of primary care and the likely continued increase in demand for vaccination services, it may be time to look more closely at the partnership between primary care and pharmacists. Pharmacists participating in this study proved every bit as diligent in following recommended ACIP and CDC adult vaccination safety guidelines and provided vaccinations at lower cost than primary care providers.
Prescribing physicians and pharmacists already are partners in providing healthcare. Acceptance of pharmacists as “alternative immunizers” could be further fostered through efforts to document and share immunization records across the healthcare system, potentially through the use of community-based registries.
