There are currently about 49.2 million adults age 65 and older living in the US, but the nation's population is aging; by 2030, one in every five adults will be age 65 or older.1,2 By 2035, this number is projected to increase to 78 million and, for the first time in US history, will exceed the number of children under age 18.1,3 Adults age 65 or older are at an increased risk for acute and chronic diseases, so identifying challenges and developing initiatives to decrease illness in this patient population is imperative. One important strategy is to ensure that all older adults are up to date with the CDC-recommended vaccinations.
According to the CDC, all adults age 65 and older should receive the following routine vaccinations: influenza; tetanus toxoid, diphtheria toxoid, and acellular pertussis vaccine (Tdap); herpes zoster (RZV, ZVL); pneumococcal conjugate (PCV13); and pneumococcal polysaccharide (PPSV23).4 Healthcare providers have a responsibility to ensure that all adults age 65 years or older receive these vaccines.5 Despite universal recommendations, many older adults are underimmunized or unimmunized. In 2015, only 70.4% of US adults age 65 and older received the influenza vaccine, a decline of 3.1% from 2014.6 Only 20.4% of adults age 65 and older report ever having received a Tdap vaccine.6 Although vaccination rates for the herpes zoster vaccine increased by 2.8% in 2015, the overall vaccination rate remains low at 33.4% in the recommended patient population of individuals age 60 and older.6 In 2016, pneumococcal vaccination rates increased by 3.3% in adults age 65 and older to 66.9%.6 By 2020, the Office of Disease Prevention and Health Promotion's Healthy People objectives propose that 90% of adults age 65 and older should receive the influenza and pneumococcal (PCV13 and PPSV23) vaccines, and 30% should receive a herpes zoster vaccine. The pneumococcal vaccines available in the US are the pneumococcal 13-valent conjugate vaccine (Prevnar 13[PCV13]) and the 23-valent Pneumococcal Polysaccharide Vaccine (Pneumovax 23 [PPSV23]). There are also two herpes zoster vaccines available in the US: the zoster vaccine live (Zostavax, [ZVL]) and recombinant zoster vaccine (Shingrix, [RZV]).7 In January 2018, the Advisory Committee on Immunization Practices (ACIP) published the Recommendations of the Advisory Committee on Immunization Practices for Use of Herpes Zoster Vaccines. The recommendations state that the RZV is preferred over the ZVL. RZV is approved by the FDA and recommended by ACIP for the prevention of herpes zoster in adults age 50 and older who are immunocompetent; ACIP also recommends RZV for immunocompetent adults who were previously vaccinated with ZVL. ZVL is approved by the FDA for the prevention of herpes zoster in immunocompetent adults age 50 or older; ACIP recommends ZVL for immunocompetent adults age 60 and older.8
This article examines the design, implementation, and success of a quality improvement project involving a clinical staff educational intervention focused on best vaccination practices and appropriate Medicare Part B and Part D billing conducted in four NP-owned primary care clinics. This quality improvement project aimed to improve vaccination rates in patients age 65 or older over a 3-month period while simultaneously improving the staff's knowledge and attitudes toward vaccinations.
Vaccination benefits in older adults
Vaccinations have numerous benefits. Vaccines are the most cost-effective primary prevention strategy.9 One study examining the impact of 10 important childhood vaccines determined that the return on investment was 16 times greater than the cost of the vaccines and their administration fees.10 Factoring in additional costs, such as the economic impact of vaccine-preventable diseases (VPDs), it is estimated that $44.00 is saved on healthcare for every $1.00 spent on vaccinations.1
VPDs are associated with significant morbidity and mortality. Each year, more than 50,000 deaths occur in the US from VPDs; 99% of these occur in adults.11 Looking specifically at the 65 or older population, annually, more than 300,000 adults are hospitalized from influenza and 23,000 die from complications as a result.12 Influenza and its complications cost the US $87 billion annually.13 Although herpes zoster is rarely fatal, it can result in chronic pain and a decreased quality of life.14 Another 18,000 adults age 65 and older die annually from pneumococcal disease.15 Fifty percent of these deaths could be prevented by pneumococcal vaccination.16
Vaccines also produce significant benefits in the form of herd or community immunity. When most of a population is immunized against a particular disease, those who are unvaccinated reap protective benefit.17 Identifying strategies to optimize the benefits of vaccines while eliminating barriers is essential.
Barriers to older adult vaccinations
Numerous barriers exist to adult vaccinations, particularly for older adults. New vaccine recommendations are issued annually by the CDC, although many providers and patients are often unaware of the updates.5 Most states have a pediatric vaccine registry, but the majority do not have an adult vaccination registry.5 As a result, patients and providers are often unsure which, if any, vaccinations have been previously administered.18 When a provider or a patient is unable to verify whether a vaccination has been previously received, both may be reluctant to administer or receive the vaccination.18
The US healthcare system is focused on acute and chronic illnesses. Less attention is paid to preventive services.5 The average primary care visit lasts 16.5 minutes.19 Shorter visits translate to fewer opportunities for healthcare providers to discuss preventive care.19 The strongest predictor of patient vaccination is provider recommendation.10 When vaccinations are not discussed and recommended by a healthcare provider, the patient is unlikely to receive the vaccine.10 Patients are often unaware of the benefits of vaccinations and do not perceive themselves at risk for VPDs.20 Explaining the risks of illness and highlighting the benefits of vaccines increase the likelihood of vaccination acceptance by the patient.20
Most adults age 65 and older are insured by a Medicare plan. In 2016, 56,981,183 individuals were enrolled in the Medicare program.21 Understanding Medicare billing and payment information for vaccinations can be confusing to patients. Although influenza, PCV13, and PPSV23 vaccines are paid for by Medicare Part B, Tdap and herpes zoster vaccinations (ZVL and RZV) are paid for by Medicare Part D.5 Medicare Part D is a pharmacy benefit and historically has only allowed billing by pharmacies. Many are unaware that primary care providers may bill Medicare Part D as well. To enable Medicare Part D billing, practice managers must enroll in a billing service, which facilitates direct billing to Medicare Part D.22 Few clinics have enrolled with these billing services. The patient must also subscribe and pay for a Medicare Part D plan to qualify for coverage of Tdap and herpes zoster vaccinations.5 Given the cost associated with Medicare Part D, patients may choose to forgo this coverage.
The cost of vaccines may also be a significant barrier.5 If an older adult believes they may be responsible for any cost, the individual is less likely to receive the vaccination.18 Many healthcare facilities do not stock vaccines because of the costs associated with purchasing and storage.23 If vaccines are not available onsite, patients are less likely to receive them.23 Although there are many barriers to vaccinating adults 65 years or older, there are numerous strategies that can increase adult vaccination rates.
Best vaccination practices
The National Vaccine Advisory Committee issued a set of recommendations, referred to as best vaccination practices.5 These evidence-based recommendations were designed to assist healthcare providers and healthcare organizations with implementing strategies to ensure that all adults receive the recommended vaccinations.5
It is imperative that healthcare providers remain current with the latest CDC recommendations. Every February, new vaccination recommendations are published, and providers must become familiar with the new guidelines.4 One strategy is to appoint a vaccine champion within the office whose role is to remain up to date on all recommendations and to communicate these updates to clinical staff.24 At every visit, the patient's immunization record should be reviewed and a recommendation should be issued by the healthcare provider to receive the needed vaccinations.5 Patients should be educated at every visit about their vaccine needs; this education should not be reserved only for preventive visits.5 Failing to address vaccinations at every visit results in missed opportunities for providing patient education and recommended vaccinations.5
In 2017, Wright and colleagues published a study demonstrating that adults age 65 and older managed in two NP-owned clinics had higher vaccination rates across five routine vaccinations than current national rates and herpes zoster vaccination rates were higher than Healthy People 2020 objectives.25 Additional studies are needed to assess the impact of NP-managed care on vaccination rates in adults age 65 and older and to identify if strategies such as educational interventions can improve vaccination rates.
Growing evidence maintains that the language used to recommend vaccinations may influence vaccination acceptance.26 Providers who use presumptive language rather than participatory language are more likely to see a patient accept the recommended vaccinations.26 One example of presumptive vaccine language is, “You are due for the flu and pneumococcal vaccines. It is safe to give them at the same visit, and I highly recommend them.” An example of participatory language is, “How do you feel about receiving the flu shot?” In a videotaped study of 111 interactions between healthcare providers and parents of children, presumptive language resulted in 74% of parents accepting the vaccination recommendation and vaccinating their children during the visit.26 Conversely, when participatory language was used, 83% of parents resisted the vaccinations, and only 3% fully accepted the healthcare provider's recommendation for the vaccinations.26
All routinely administered vaccines should be stocked in the office.5 If vaccines are not available onsite, the patient should be referred to a nearby location such as a pharmacy, public health department, or another healthcare facility to receive the vaccination.5 Providers must also be familiar with what constitutes a valid contraindication or reason to defer a vaccination.5 A minor illness is not a valid reason to defer, whereas a moderate to severe illness and/or a history of anaphylaxis to an ingredient in the vaccine are valid contraindications to vaccine administration.27
When a patient refuses a vaccination, this refusal should be documented in the electronic health record (EHR). To serve as a future reminder, providers can include an ICD-10 code, such as Z28.21 (vaccination not carried out because of patient refusal) or Z28.82 (vaccination not carried out because of caregiver refusal) in the EHR.5 If a patient needs a vaccination that cannot be administered on the same day, such as the second dose of the herpes zoster vaccine, which must be given 2 to 6 months after the first dose, the appointment should be scheduled in the EHR system so that the patient receives a reminder notification. Electronic clinical decision prompts in EHR systems and reminder/recall systems have been shown to increase vaccination rates.5
Sample. Two different samples were used in all phases of this quality improvement project. The first sample included the medical records of all patients receiving care in four NP-owned clinics (n = 808). The second sample consisted of all NPs, RNs, licensed practical nurses (LPNs), and medical assistants (MAs) who administer vaccinations and are employed in the four clinics (n = 18).
Ethical considerations. Patients in this project had previously consented to allow their EHR to be used for quality improvement and research purposes. NPs and clinical staff who participated consented to their participation, and no identifying information was collected. Confidentiality was maintained throughout the project with minimal risks to participants. The project was approved by the University of Alabama Institutional Review Board.
Setting. This project was conducted in four NP-owned primary care clinics located in New Hampshire.
Data collection and intervention
Phase I. A report of the medical record number, gender, and age for all active patients age 65 years or older was generated by the practice manager of each clinic. The immunization records of 808 patients were reviewed. Preintervention data were collected from July 1, 2018, through July 31, 2018. Documentation on the receipt of the vaccinations (Tdap, PCV13, PPSV23, herpes zoster [ZVL, RZV], and influenza) was recorded in an Excel spreadsheet. Influenza vaccine was considered to have been received if it was administered to the patient between August 1, 2017, and November 1, 2017; PCV13 and PPSV23 were considered to have been administered if received when the patient was age 65 or older; and Tdap, ZVL, and RZV were considered vaccine adherent if they had ever been received.
Phase II. In July 2018, the Vaccination Knowledge and Attitudes Survey was sent to the practice manager of each clinic. All clinical staff who administer vaccinations were asked to sign the consent form, separate the consent from the survey, and complete the anonymous survey. The questionnaire consisted of 10 items developed by the primary author of this article (see Vaccination Knowledge and Attitudes Survey). Respondents were asked to indicate the extent of their agreement with each statement using a five-point Likert scale with response options that ranged from “Strongly Disagree” to “Strongly Agree.” The questionnaire was designed to assess knowledge regarding vaccination practices and commitment to the vaccination process. The consent forms and questionnaires were then returned to the investigator prior to commencing with the educational intervention.
After the preintervention vaccination rates were collected and the preintervention Vaccination Knowledge and Attitude Survey was completed, a 45-minute educational intervention was conducted in late July 2018 for the NPs and clinical staff at the four clinics. This presentation included a slide deck with information on best vaccination practices and strategies to incorporate these practices into the care of patients age 65 or older. Specific instructions on billing Medicare Part B and Part D plans were also included, as well as preintervention immunization rates from each clinic with comparisons to the 2016 US vaccination rates and Healthy People 2020 objectives.7
Phase III. Three months after the completion of the educational intervention, the vaccination records of the same 808 patients were reviewed. Data were collected on receipt of the same vaccinations and entered into an electronic spreadsheet. Influenza vaccine was considered to have been received if administered between August 1 and November 1, 2018; PCV13 and PPSV23 if received at age 65 or older; and Tdap, ZVL, and RZV, if ever received. Patient medical record numbers from the pre- and postintervention periods were matched, enabling the use of repeated measures statistical analyses.
After the educational presentation, the Vaccination Knowledge and Attitudes Survey was completed by the same clinical staff for a second time. The responses were anonymous, so participants' pre- and postintervention responses could not be matched for data analysis.
IBM SPSS Statistics Version 25 was used for data analysis. Clinics 2, 3, and 4 were combined into one clinic for data analysis because their patients had been seen in more than one of the clinics. For this quality improvement project, these three combined clinics will be referred to as Clinic 2.
The vaccination data were analyzed separately for the two clinics due to differences in their preintervention vaccination rates. Descriptive statistics for patients' age and gender were computed for each clinic (see Patient demographics by clinic). The percentage of patients receiving each vaccine was calculated for the pre- and postintervention periods (see Clinic 1 preintervention and postintervention vaccination rates and Clinic 2 preintervention and postintervention vaccination rates). Vaccination rates were also compared with US rates and Healthy People 2020 objectives (see Clinic vaccination rates compared with national rates and Healthy People 2020 objectives).7
Means and standard deviations for all 10 items on the Vaccination Knowledge and Attitudes Survey were computed for the pre- and postintervention periods. Higher means are indicative of more agreement, and lower means are indicative of less agreement. Given the small sample size and skewed distributions, the nonparametric Mann Whitney test was used to compare preintervention and postintervention responses for each question.
Phase I (preintervention demographics and vaccination rates). There were 658 patient records from Clinic 1. All preintervention vaccination rates were below the 2016 US vaccination rates and Healthy People 2020 objectives, except ZVL. There were 150 patient records included from Clinic 2. All preintervention vaccination rates were below the 2016 US vaccination rates except Tdap and all were also below Healthy People 2020 objectives.
Phase II (preintervention responses to the Vaccination Knowledge and Attitudes Survey). A total of 18 NPs and clinical staff completed the Vaccination Knowledge and Attitudes Survey (Clinic 1, n = 8; Clinic 2, n = 10). (See Vaccine Knowledge and Attitudes Survey results [pre- and postintervention]). The responses did not differ significantly between the two clinics and, therefore, were combined for data analysis. The means for each question ranged from 3.11 to 4.78.
Phase III (postintervention vaccination rates). There were 658 patient records from Clinic 1 included in the postintervention analysis. All preintervention patients were included in the postintervention data collection and analysis. Postintervention vaccination rates were Tdap: 22.2%; Influenza: 39.4%; PPSV23: 45.9%; PCV13: 55.6%; ZVL: 38.4%; RZV Dose 1: 8.5%; RZV Dose 2: 2.9%. All postintervention vaccination rates increased from preintervention and all, except Influenza and ZVL, were statistically significant. Postintervention Tdap rates exceeded the 2016 US vaccination rates, and ZVL exceeded both the 2016 US vaccination rates and Healthy People 2020 objectives.
There were 150 patient records from Clinic 2 included in the postintervention analysis; these were the same records included in the preintervention analysis. Postintervention vaccination rates were Tdap: 27.3%; Influenza: 42.7%; PPSV23: 32.7%; PCV13: 15.3%; ZVL: 13.3%; RZV Dose 1: 2.0%; RZV Dose 2: 0.0%. All postintervention vaccination rates improved from baseline. Although pre- versus postintervention rates of RZV could not be compared because of the small sample size, Dose 1 did increase slightly. All were statistically significant except RZV Dose 1 and Dose 2 and Influenza. Postintervention Tdap rates exceeded the 2016 US vaccination rates while the remainder of vaccinations were below the 2016 US vaccination rates and Healthy People 2020 objectives.
Postintervention survey. The same 18 NPs and clinical staff completed the postintervention Vaccination Knowledge and Attitudes Survey. The means of each question ranged from 3.11 to 4.83. All means increased from the preintervention questionnaire with three exceptions: Question 1 (I advocate for adults age 65 years and older to become better vaccinated) decreased from 4.61 to 4.39; Question 2 (I know how to access current information from the CDC to inform my vaccination practice) and Question 10 (My patients age 65 years and older often refuse vaccinations) remained unchanged at 4.56 and 3.11, respectively. The largest improvement was Question 9 (I understand the difference between presumptive versus participatory language), which increased from 3.61 to 4.33. None of the changes were statistically significant.
All vaccination rates in the two clinics improved over the 3 months following the educational intervention, except RZV Dose 2 in Clinic 2. In addition, the majority were statistically significant. These improvements are notable given the short time frame.
Although the percentage of adults receiving the influenza vaccine increased in both clinics, the results were not statistically significant. Because of the timing of the project, postintervention data collection was conducted 4 months into the 2018-2019 influenza season. To allow for a similar comparison of influenza vaccination rates year to year, recipients were considered to have received the influenza vaccine if it was administered from August 1, 2017 to November 1, 2017 (preintervention) and August 1, 2018 to November 1, 2018 (postintervention). In a typical year, influenza vaccines are administered from August 1 to April 30. This shortened time frame may have negatively impacted the results.
In December 2017, the CDC recommended two doses of RZV for all adults over age 50. They encouraged all patients previously vaccinated with ZVL to be vaccinated with RZV and made RZV the preferred herpes zoster vaccine. With this recommendation, the use of ZVL in the US was essentially eliminated. ZVL vaccination rates did not significantly improve in either clinic over the course of this study. This was likely a result of ZVL no longer being routinely administered. Although there was a slight increase in ZVL vaccination rates in both clinics, this may be attributed to the clinics receiving vaccination records from the patient's previous provider, or providers and pharmacists administering the ZVL vaccine because they had previously purchased supplies of the vaccine and had some remaining doses.
Following the CDC's recommendation for RZV vaccination, patient demand was significant, resulting in a shortage of the vaccine. The clinics had difficulty obtaining the vaccine from the manufacturer, which likely reduced vaccination rates. To administer and bill for herpes zoster vaccinations, clinics must set up accounts with the various vaccine manufacturers, order the vaccinations, and enroll in a Medicare Part D billing service. This process can take months and may have exceeded the time frame allowed for this project.
The postintervention means from all but three items on the Vaccination Knowledge and Attitudes Survey improved from the preintervention. The largest increase was seen with Question 9 (I understand the difference between presumptive versus participatory language). This language was specifically addressed in the educational intervention and may explain the score increase for this item. Most of the means on the preintervention questionnaire were high (4.00 or higher), which meant there was not much room for improvement. Given the small sample, producing statistically significant changes was difficult.
There appears to be a disconnect between the self-reported knowledge and attitudes toward vaccinations and the actual vaccination rates. Whereas the attitudes and knowledge were reported as high, actual vaccination rates were below national averages and Healthy People 2020 objectives. This may be due to a patient, system, or process issue within the clinics that does not allow staff knowledge to be applied. For instance, vaccinations might not be available from the suppliers and/or the clinic systems are not equipped to handle the vaccination processes. Patients may also not perceive themselves at risk and refuse vaccinations.
There were limitations to this project. Clinics 2, 3, and 4 were combined for data analysis because some patients were seen in more than one of the clinics. Given the timeline associated with this project, the influenza vaccine data had to be collected only 4 months into influenza season, which may have limited the accuracy of the data. Purchasing vaccines and enrolling with Medicare Part D billing services can take significant time. Clinics that did not already have these processes in place may not have had enough time to be able to demonstrate their results for inclusion in this study, thus negatively impacting the results. This project was conducted in NP-owned primary care clinics. It is not known if this same intervention would work in larger healthcare systems, physician-owned practices, or specialty clinics. The Vaccination Knowledge and Attitudes Survey was developed by this investigator and has not been tested for reliability or validity. The small sample size (N = 18) meant there was relatively less power when conducting statistical analysis of pre- versus post differences on the Vaccination Knowledge and Attitudes Survey. It is also not possible to determine the sustainability of interventions on practice over time.
The results from this quality improvement project suggest that an educational intervention including current practice immunization rates, best vaccination practices, and information on billing for adult vaccinations can produce improvements in vaccination rates over a 3-month period, the majority of which were statistically significant. (See Vaccination best practices.) This intervention also improved knowledge and attitudes regarding adult vaccinations. Quality improvement projects such as this can improve vaccination rates, even over a short time period. This quality improvement project should serve as a pilot for more generalizable research conducted over a longer period, in a variety of settings, and with control and test groups for comparison.
Vaccination best practices5
- Appoint a vaccine champion within the office to update the staff and communicate changes in vaccine recommendations.
- Make sure all staff are regularly educated about vaccines and their importance.
- Promote a provaccine culture within the facility.
- Issue a strong recommendation to patients at every visit, not just at preventive visits, about the need for vaccinations.
- Use presumptive language not participatory language when discussing vaccinations.
- Mild illness is not a reason to defer a vaccination and the vaccination should be provided, when appropriate.
- Stock all vaccinations within the facility.
- Enroll in billing service that can enable a facility to bill for Medicare Part D vaccines.
- Use reminder/recall systems for vaccines.
- Use clinical decision-making prompts within the EHR.
- Vaccination refusal should be documented in the clinical note and EHR system.
- Set reminders for needed vaccinations within the EHR.
- Update and utilize vaccine registries, if available within the facility or state.
- If unable to confirm the receipt of a vaccine, revaccinate (when in doubt, revaccinate).
- Develop relationships with other vaccine providers within the community to provide needed vaccines if unavailable within the facility.
- Run the office/healthcare system vaccination rates.
- Implement quality improvement projects within the facility to improve vaccination rates.
4. Kim DK, Hunter P. Recommended adult immunization schedule, United States, 2019. Ann Intern Med
5. National Vaccine Advisory Committee. Recommendations from the National Vaccine Advisory Committee: standards for adult immunization practice. Public Health Rep
8. Dooling KL, Guo A, Patel M. Recommendations of the Advisory Committee on Immunization Practices for Use of Herpes Zoster
Vaccines. MMWR Recomm Rep
9. Ventola CL. Immunization in the United States: recommendations, barriers, and measures to improve compliance: Part 1: childhood vaccinations
. P T
10. Sevin AM, Romeo C, Gagne B, Brown NV, Rodis JL. Factors influencing adults' immunization practices: a pilot survey study of a diverse, urban community in central Ohio. BMC Public Health
11. Ventola CL. Immunization in the United States: recommendations, barriers, and measures to improve compliance: Part 2: adult vaccinations
. P T
. 2016;41(8):492–497, 502-506.
12. Rüttimann RW, Bonvehí PE, Vilar-Compte D, Isturiz RE, Labarca JA, Vidal EI. Influenza
among the elderly in the Americas: a consensus statement. Pan Am J Public Health
13. Lang PO, Aspinall R. Vaccination in the elderly: what can be recommended. Drugs Aging
14. Johnson RW, Alvarez-Pasquin MJ, Bijl M, et al Herpes zoster
epidemiology, management, and disease and economic burden in Europe: a multidisciplinary perspective. Ther Adv Vaccines
16. Prevention of pneumococcal disease: recommendations of the advisory committee on immunization practices (ACIP). MMWR Recomm Rep
19. Young RA, Burge SK, Kumar KA, Wilson JM, Ortiz DF. A time-motion study of primary care physicians' work in the electronic health record era. Fam Med
20. Turner M, Parks C, Murphy F, et al Pneumococcal vaccination: identifying barriers and strategies to improve administration rates. AAACN Viewpoint
23. Anderson EL. Recommended solutions to the barriers to immunization in children and adults. Mo Med
24. Mazzoni SE, Brewer SE, Pyrzanowski JL, et al Effect of a multi-modal intervention on immunization rates in obstetrics and gynecology clinics. Am J Obstet Gynecol
25. Wright WL, Morrell E, Lee J, Cuellar NG, White P. Comparison of immunization rates of adults ages 65 years and older managed within two nurse practitioner-owned clinics with national immunization rates. J Am Assoc Nurse Pract
26. Opel DJ, Zhou C, Robinson JD, et al Impact of childhood vaccine discussion format over time on immunization status. Acad Pediatr
27. Kroger AT, Sumaya CV, Pickering LK, Atkinson WL. General recommendations on immunization recommendations of the advisory committee on immunization practices (ACIP). MMWR Recomm Rep