The impact of pharmacist involvement on immunization uptake and other outcomes: An updated systematic review and meta-analysis.

BACKGROUND: The underutilization of immunization services remains a big public health concern. Pharmacists can address this concern by playing an active role in immunization administration.
OBJECTIVE: We performed a systematic review and meta-analysis to assess the impact of pharmacist-involved interventions on immunization rates and other outcomes indirectly related to vaccine uptake.
METHODS: A systematic literature search was conducted using MEDLINE, Embase, and Cochrane Central Register of Controlled Trials databases from inception to February 2022 to identify randomized controlled trials (RCTs) and observational studies in which pharmacists were involved in the immunization process. Studies were excluded if no comparator was reported. Two reviewers independently completed data extraction and bias assessments using standardized forms. Meta-analyses were performed using a random-effects model.
RESULTS: A total of 14 RCTs and 79 observational studies were included. Several types of immunizations were provided, including influenza, pneumococcal, herpes zoster, Tdap, and others in a variety of settings (community pharmacy, hospital, clinic, others). Pooled analyses from RCTs indicated that a pharmacist as immunizer (risk ratio 1.14 [95% CI 1.12-1.15]), advocator (1.31 [1.17-1.48]), or both (1.14 [1.12-1.15]) significantly increased immunization rates compared with usual care or non-pharmacist-involved interventions. The quality of evidence was assessed as moderate or low for those meta-analyses. Evidence from observational studies was consistent with the results found in the analysis of the RCTs.
CONCLUSION: Pharmacist involvement as immunizer, advocator, or both roles has favorable effects on immunization uptake, especially with influenza vaccines in the United States and some high-income countries. As the practice of pharmacists in immunization has been expanded globally, further research on investigating the impact of pharmacist involvement in immunization in other countries, especially developing ones, is warranted.

Background

The underutilization of immunization services remains a big public health concern.

Evidence shows a favourable impact of pharmacist involvement on immunization uptake. However, the existing literature is not comprehensive because of several limitations.

Findings

A systematic review of the literature identified fourteenth randomized controlled trials and seventy-nine observational studies from a range of country settings.

The findings demonstrated that the pharmacist involvement as immunizers or advocators or both significantly increased the immunization uptake, especially strong evidence for influenza vaccine in the United States and some high-income countries.

The findings pose a potential benefit of exploring and expanding the scope of pharmacist practice in immunization in other countries, especially low-and-middle-income countries.

Immunizations are considered one of modern medicine’s greatest achievements and are estimated to save up to 2.5 million lives every year. Despite that, according to the World Health Organization, vaccine hesitancy—the reluctance or refusal to vaccinate despite the availability of vaccines—is one of the top 10 global health threats that result in the underutilization of immunization services. Apart from that, inaccessibility to vaccines or lack of understanding regarding vaccine benefits is also a potential contributor. Consequently, vaccine-preventable diseases continue to be a major concern. Worldwide, seasonal influenza has caused an estimated 3-5 million cases of severe illness and approximately 290,000-650,000 respiratory deaths annually.,

In the United States, pharmacists are one of the most accessible health professionals. In other countries, especially developing ones, pharmacists are often one of the first health professionals sought for care, especially with individuals who do not have access to a primary health care provider. The underutilization of vaccines along with previous influenza pandemics, such as swine flu, has led to an increased demand for immunizations. This has created opportunities for pharmacists to play a role in improving immunization rates and advancing public health. Stakeholders have identified 3 roles for pharmacists in immunization: pharmacists as facilitators (hosting others who vaccinate), pharmacists as advocates (educating and motivating patients), and pharmacists as immunizers (vaccinating patients). In the past few decades, pharmacists have been authorized to administer vaccines in all 50 U.S. states since the American Pharmacists Association established its Pharmacy-Based Immunization Delivery Program in 1996. Statewide protocols are methods to expand pharmacist immunization authority and provide standardized procedures for consistency and safety. These protocols are established by state laws and regulations governing the practice of pharmacy and applicable to qualified immunizing pharmacists. Despite the fact that pharmacist training and guidelines are standardized, there is large variation among statewide protocols. Immunization administration regulations differ by the types of vaccinations allowed, eligible patient age for administration, and allowance of prescriptive authority. Although all states allow pharmacists to administer vaccinations, many do so through statewide protocols, whereas some states do not have such protocols. In this case, these states have chosen to pass laws that allow pharmacists to administer vaccines without a protocol or prescription. The trend in the United States and increasingly in other developed countries (United Kingdom, Canada, Ireland, etc.), has been to grant greater autonomy to pharmacists in the prescribing and administering of vaccinations, which correlates with improved health outcomes and reduced health care costs. In developing countries, the pharmacists’ role in patient care, including immunization, is still limited. The provision of immunizations by pharmacists has had a positive effect in the United States and other developed countries,15, 16, 17, 18, 19, 20, 21, 22 and may have potential applicability to other countries.

With the current coronavirus disease 2019 (COVID-19) epidemic, the Centers for Disease Control and Prevention has called on and highlighted the essential role of pharmacists as part of the response to the pandemic. As a result of it, pharmacist standard operating procedures were expanded to allow pharmacists to administer COVID-19 vaccines. Since then, pharmacists have played a key role in the vaccination process. As of April 2022, pharmacists have administered more than 240 million doses of COVID-19 vaccines. The federal government has recognized the key role they can play and has made them a key part of their COVID-19 vaccination plan.

Several studies have investigated the impact pharmacist-based vaccine administration may have on influenza, pneumococcal, and herpes zoster vaccination rates.16, 17, 21, 24, 25, 26 Previous meta-analyses demonstrated pharmacist involvement in vaccine administration has a statistically significant impact on immunization rates., However, most of the studies included in these meta-analyses used weak designs, including controlled before-after, retrospective cohort, and quasi-experimental, which potentially overestimates the impact on immunization rates. Those analyses also had a large amount of heterogeneity. Moreover, the meta-analyses conducted only included a small number of trials with high risk of bias (ROB) or focused on single outcomes such as immunization uptake or pharmacists as immunizers.

Objective

The purpose of this study is to conduct an updated systematic review and meta-analysis with the most recent trials to explore the impact of pharmacist involvement (i.e., as immunizer or advocator or both) on the immunization rate and other outcomes indirectly related to vaccine uptake. Other outcomes include confidence of pharmacists in vaccine recommendation/administration, perception of patients about vaccination, patient satisfaction, vaccine compliance, etc.

Methods

This meta-analysis followed the Preferred Reporting Items for Systematic Reviews and Meta-Analyses guideline. The systematic review protocol was registered at the International Prospective Register of Systematic Reviews (CRD42021251119).

Search strategy

We performed a systematic search via MEDLINE, Embase, and Cochrane Central Register of Controlled Trials from inception to February 28, 2022. Search terms used were immunization, vaccination, pharmacist, pharmacy, pharmaceutical services, and pharmaceutical care. The search strategy is provided in Supplemental Materials, Supplement I. We also searched reference lists of previous systematic reviews. Duplicate studies from multiple sources were removed. Title and abstract screening and full-text screening were performed by 2 reviewers (L.M.L. and D.D.) based on the inclusion and exclusion criteria to identify eligible studies to include in the systematic review and meta-analysis.

Eligibility criteria and study selection

Included studies were either randomized controlled trials (RCTs)/cluster RCTs or observational studies with a comparison group that measured immunization rate and other related outcomes that indirectly improve vaccine uptake such as improvement in vaccine hesitancy and resistance rate, vaccine appropriateness, vaccine compliance, patient’s awareness and attitude toward vaccination, patient satisfaction, etc. Immunization rate is defined as the number of people that have received vaccines divided by the target population. Observational studies included in this review were classified as non-RCTs, controlled before and after studies, before and after studies, retrospective cohorts, and cross-sectional surveys., Interventions needed to involve pharmacists in the immunization process as facilitators, advocators, immunizers or both advocators and immunizers. Comparisons of interest included usual care (defined as routine or standard of care received by patients) or other interventions without pharmacist involvement.

Data extraction

Data extraction was performed by 2 reviewers (D.D. and K.W.) independently using Microsoft Excel (Microsoft Corporation) to record the screening processes and decisions. In cases of discrepancy, a third reviewer (L.M.L.) was consulted to reach a consensus for the inclusion of discrepant studies. Data were extracted using standard data extraction forms. For each included study, data extracted included author name, publication year, study design, setting (community, hospital, clinic, others), population, types of vaccine, intervention, comparators, the role of pharmacists in vaccination (facilitators, immunizers, educators, etc.), number of participants in each arm, total number of cases and participants, follow-up period (mean or median), and country of each study. Data on primary outcomes were extracted using the intention-to-treat analysis principle.

ROB assessment

For RCTs, we used the Cochrane ROB (ROB 2.0) tool (Cochrane) as the framework for assessing the ROB in a signal estimate of an intervention effect. Studies were judged as low ROB, some concern, and high ROB. We used the Cochrane ROB (ROBINS-I) tool (Cochrane) for nonrandomized studies. Studies were judged as low, moderate, serious, or critical ROB or no information.

The Grading of Recommendations, Assessment, Development and Evaluation approach was used to rate the quality of evidence (high, moderate, low, and very low) of estimates derived from meta-analyses., Two reviewers (L.M.L. and S.V.) independently assessed the confidence in effect estimates for all outcomes using the following categories: ROB, inconsistency, indirectness, imprecision, and publication bias.

Data analysis

Meta-analysis was performed using a random-effects model under DerSimonian and Laird method. We estimated pooled risk ratios (RRs) of pharmacists as immunizers, advocators, or both compared with usual care or intervention without pharmacist involvement and 95% CIs incorporating heterogeneity within and between studies. The intention-to-treat principle was used for all analyses. Statistical heterogeneity between trials was assessed using I statistics and Q-statistics. I with values > 50% and Q-statistics with P value < 0.05 indicate substantial levels of heterogeneity. We assessed publication bias using funnel plot asymmetry testing and the Egger regression test. Subgroup analyses based on settings and vaccine type were also performed. To assess the robustness of the findings of our primary outcome, we performed multiple sensitivity analyses based on the following assumptions: (1) exclusion of studies with high ROB, (2) exclusion of studies reporting pharmacists with multiple roles, and (3) using the per-protocol principle (adherence to treatment). All analyses were performed with Stata version 15.0 (StataCorp LLC, College Station, TX).

Patient and public involvement

Patients or the public were not involved in the designing, conducting, reporting, or disseminating of the plans for this research.

Results

Study selection

As shown in Figure 1, a total of 7336 records were identified from our literature search. After removing duplicates and screening titles and abstracts, 309 studies were assessed for eligibility. We excluded a total of 230 studies for the following reasons: abstract only publications (n = 162), no comparator group (n = 31), not intervention of interest (n = 19), not outcome of interest (n = 4), duplicate reports (n =11), or no full-text available (n =3). In total, our initial search yielded 12 RCTs and 67 observational studies.

Figure 1

Preferred Reporting Items for Systematic Reviews and Meta-Analyses flowchart. Abbreviations used: CENTRAL, Cochrane Central Register of Controlled Trials; RCT, randomized controlled trial.

After searching the reference lists of previous meta-analyses, we found 8 eligible RCTs and 27 observational studies. After combining the 2 sources and removing duplicates, we had a total of 14 RCTs and 79 observational studies. Of these, we included 4 RCTs37, 38, 39, 40 that were not included in previous systematic reviews.,

Characteristics of included studies

Table 1 describes the characteristics of the included RCTs. Eleven RCTs15, 16, 17,,,40, 41, 42, 43, 44, 45 assessed the impact of pharmacist involvement on immunization uptake; 1 study in United States investigated the change in the confidence of pharmacists in pneumococcal and herpes zoster vaccine recommendation and administration after tailored training for pharmacists. Another study evaluated the impact of different advocating strategies in a community pharmacy setting using electronic messages and the impact on attitude of patients toward the pneumococcal vaccine in the United States. A study in Jordan evaluated the impact of pharmacist-physician coaching intervention about COVID-19 vaccines via Facebook lives sessions on the COVID-19 vaccine hesitancy and resistance (Table 1).

Table 1

Characteristics of studies included in meta-analysis

Author and year	Country	Study design	Setting	Intervention and comparison	Type(s) of immunization	Pharmacist role	Outcome	Duration of interventional practice	Age groups	n, treatment/control	Bias assessment
Ginson et al.,15 2000	Canada	Cluster RCT	Hospital	Written or verbal information of vaccines from a pharmacistComparison: usual care	Influenza and pneumococcal	Educator/Advocator	Immunization rate	3 mo	≥ 65	50/52	Some concerns
Higginbotham et al.,16 2012	United States	RCT	Primary health care center	Compare 3 protocols: P1: administered INA only (comparison); P2: identified patients not current on immunization using INA, pharmacist recommended and administered the needed vaccines; P3: INA results and a vaccine education sheet were given to the patients	Influenza, pneumococcal, tetanus, hepatitis A, hepatitis B, MMR, varicella, meningococcal, HPV, and shingles	Immunizer	Immunization rate	4 mo	18–79	P1:P2: 28P3: 36	Some concerns
Klassing et al.,38 2018	United States	RCT	Community pharmacy	Personal phone call, standardized mail letterComparison: usual care	Influenza and pneumococcal	Advocator	Immunization rate	4 mo	≥ 18	Phone call: 276Letter: 277Control: 278	High risk of bias
Marra et al.,29 2014	Canada	Cluster RCT	Community pharmacy	Personalized invitation letters to the eligible clients inviting them to be vaccinated at the pharmacy clinics, pharmacist educate patients on the benefits of vaccinationComparison: usual care	Influenza	Immunizer/Educator/Advocator	Immunization rate	2 y	≥ 65 and < 65	61,623/73,626	Some concerns
Nipp et al.,40 2019	United States	RCT	Pharmacy	Pharmacists perform medication and vaccination history review, evaluate medication indications and recommend medications/vaccinationComparison: usual care	Influenza and pneumococcal	Advocator	Immunization rate	2 mo	≥ 65	29/31	Some concerns
Otsuka et al.,44 2013	United States	RCT	General internal medicine clinic	Information on herpes zoster vaccine by electronic message or letter to personal and nonpersonal health record users, respectively. Pharmacist performed medical record review to confirm herpes zoster vaccine indicationComparison: usual care	Herpes zoster	Advocator	Immunization rate	6 mo	≥ 60	500/2089	Some concerns
Usami et al.,17 2009	Japan	Cluster RCT	Community pharmacy	Pharmacists provide information on the risk of influenza and benefits of influenza vaccination, display posters and leafletComparison: usual care	Influenza	Educator/Advocator	Immunization rate	5 mo	≥ 65	911/952	High risk of bias
Grabenstein et al.,41 1993	United States	RCT	Community pharmacy	Letters explaining the risk of influenza, and availability of vaccineComparison: Letters advising 'poison-proof' their home and return expired or unwanted medications	Influenza	Advocator	Immunization rate	6 mo	≥ 65	299/299	High risk of bias
Kirwin et al.,42 2010	United States	RCT	Hospital-based primary care	Personalized letter from pharmacists to physicians containing treatment recommendation for patientsComparison: usual care	Pneumococcal	Advocator	Immunization rate	1 mo	≥ 65	171/175	High risk of bias
Marrero et al.,43 2006	Puerto Rico	RCT	Community pharmacy	Educational session about influenza and vaccination clinic in a pharmacyComparison: usual care	Influenza	Educator	Immunization rate	12 mo	≥ 65	40/42	High risk of bias
Hasting et al.,37 2020	United States	RCT	Community pharmacy	Immunization training for pharmacists with educational intervention and tailored feedbackComparison: 1-h immunization update	Pneumococcal, Herpes zoster	Immunizer/Advocator	Confidence, Perceived external support, Perceived influence	6 mo	≥18	37/30	High risk of bias
Krueger et al.,39 2020	United States	RCT	Community pharmacy	Message to customers about information on pneumonia prevention, pneumococcal vaccine costs, vaccine safety, community and family duty.Comparison: usual care	Pneumococcal	Advocator	Patients’ attitude toward vaccine	NR	≥18	2211/397	Some concerns
Heaton et al.,45 2022	United States	RCT	Community pharmacy	Pharmacists are provided access to ImmsLink portal to review patient’s immunization history and recommendation.Comparison: usual care	Influenza, Pneumococcal, Herpes zoster, Tdap	Advocator	No. vaccinations	3 mo	≥18	16,750/17,062	Some concerns
Abdel-Qader et al.,46 2022	Jordan	RCT	NA	Pharmacist-physicians coaching intervention about COVID-19 vaccines via Facebook livesComparison: no intervention	COVID-19	Advocator	Vaccine hesitancy/reistance	2 mo	≥18	154/151	Some concerns

Abbreviations used: RCT, randomized controlled trial; INA, immunization needs assessment; MMR, measles, mumps and rubella; HPV, human papillomavirus vaccine; NR, not reported; NA, not applicable; COVID-19, coronavirus disease 2019.

Two studies, assessed the role of pharmacists as immunizers only, whereas 11 studies,,38, 39, 40, 41, 42, 43, 44, 45, 46 assessed pharmacists as advocators only and 1 study for both. No eligible study was found with pharmacists as facilitators. Ten studies,37, 38, 39, 40, 41, 42, 43, 44, were RCTs; 4,,, were cluster randomized trials. Most studies were conducted in high-income countries: 9 studies,37, 38, 39, 40, 41, 42,, from the United States, 2, from Canada, one from Japan, and another one from Puerto Rico. Only one study was conducted in an upper-middle-income economy (Jordan).

From these studies, 17 different kinds of comparator interventions were made against pharmacist-involved interventions. Comparator strategies included pharmacists reviewing medication indication and vaccination history and recommending vaccines to patients/physicians (4 studies,,,, 23.5%), patient education (4 studies,,,, 23.5%), personalized letters (5 studies,,,,, 29.4.%), phone calls (1 study, 5.9%), electronic messages (2 studies,, 11.8%), and trainings (1 study, 5.9%). Studies assessed a variety of different vaccines administered in adult population including influenza (9 studies15, 16, 17,,,,,,), pneumococcal (8 studies,,37, 38, 39, 40,,), herpes zoster (3 studies,,), Td/Tdap (1 study), and COVID-19 (1 study). These studies were conducted primarily in 3 settings: community pharmacy (8 studies,,,37, 38, 39,,, 57.1%), hospital (4 studies,,,, 28.6%), and a primary health care center (1 study, 7.1%).

Table 2 describes the characteristics of the 79 included observational studies. Most studies were conducted in high-income countries: a total of 64 were conducted in the United States (81.0%), 6 in Canada (7.6%), 3 in the Australia (3.8%), 2 in United Kingdom (2.5%), and 1 in Germany (1.3%). Two studies were conducted in upper-middle-income countries: 2 from Turkey (2.5%) and 1 from Jordan (1.3%). The different study designs consisted of 31 before-after studies (39.2%), 18 retrospective cohort studies (22.8%), 17 controlled before-after studies (21.5%), 12 nonrandomized trials (15.2%), and 1 cross-sectional survey (1.3%). There were 15 studies assessing the impact of pharmacists as immunizers (19.0%), 35 as advocators (44.3%), and 2 as facilitators (2.5%). A total of 27 studies (34.1%) investigated the effect of intervention in which pharmacists served more than one role.

Table 2

Characteristics of included observational studies

Category	No. studies (N = 79), n (%)
Country
United States	64 (81.0)
Canada	6 (7.6)
Australia	3 (3.8)
United Kingdom	2 (2.5)
Turkey	2 (2.5)
Germany	1 (1.3)
Jordan	1 (1.3)
Study design
Before-after study28	31 (39.2)
Retrospective cohort study29	18 (22.8)
Controlled before-after study28	17 (21.5)
Nonrandomized trial28,29	12 (15.2)
Cross-sectional survey29	1 (1.3)
Pharmacist role
Immunizer	15 (19.0)
Facilitator	2 (2.5)
Advocator	35 (44.3)
Immunizer/advocator	23 (29.1)
Facilitator/advocator	2 (2.5)
Immunizer/facilitator/advocator	2 (2.5)
Study setting
Community pharmacy	25 (31.6)
Hospital	16 (20.3)
Medical center	10 (12.7)
Clinic setting	17 (21.5)
Other	7 (8.9)
Unspecified	4 (5.1)
Interventiona
Patient education and motivation (pharmacist counseling, letter, phone call)	32 (40.5)
Immunization eligibility/need assessment	26 (32.9)
Pharmacists authorized to administer vaccine	12 (15.2)
Pharmacist training	9 (11.4)
Pharmacist immunization service	8 (10.1)
Others	2 (2.5)
Type of vaccineb
Influenza	40 (50.6)
Pneumococcal	42 (53.2)
Herpes zoster	12 (15.2)
Tdap	10 (12.7)
Other	12 (15.2)
Unspecified	1 (1.3)
Outcomec
Immunization rate	64 (81.0)
Knowledge/awareness/perception	8 (10.1)
Vaccine errors/appropriateness/failure	4 (5.1)
Satisfaction/barrier	8 (10.1)
Completion of dose/structure	2 (2.5)
Cost	3 (3.8)

Some studies incorporated several interventions (for example: medication review was performed together with patient education).

Some studies assessed more than 1 vaccine.

Some studies reported more than 1 outcome.

A detailed description of the ROB assessment among included RCTs is presented in Supplemental Materials, Supplement II. Six studies,,,41, 42, 43 were found to have an overall high ROB and 8 studies,,,,,44, 45, 46 had some concerns. No studies were at low ROB. ROB related to the generation of the allocation sequence and allocation concealment was low for most studies except one, which used an alternative randomization method. By the nature of the intervention, it was impossible to blind the participants or pharmacists to the assigned intervention group. However, we did not think that it affected the deviations from the intended interventions for 7 studies.15, 16, 17,,,, Five other studies,,41, 42, 43 had a loss to follow-up, so we assessed them as high ROB. Five studies,,41, 42, 43 had high or some concerns of ROB on missing outcome data given that their analyses were performed for the sample size after exclusion of loss to follow-up. Three studies,, were concerned about the bias of measurement of the outcome given that the outcome accessors were not blinded. All the studies had unclear risk about the selective reporting bias (Table 1; Supplemental Materials, Supplement II).

ROB assessment for observation studies is presented in Supplemental Materials, Supplements III and IV. Overall ROB was assessed as low, moderate, serious, and critical for 2 studies (2.8%), 25 studies (31.6%), 26 studies (32.9%), and 25 studies (31.6%), respectively. One study (1.3%) did not have sufficient information to assess ROB.

Evidence from RCTs

Pharmacists as immunizers

Pooled analysis of 2 RCTs, (n = 135,350) (Figure 2) for pharmacists as immunizers demonstrated a statistically significant increase in immunization rate (RR 1.14 [95% CI 1.12–1.15]), favoring the intervention compared with usual care or other intervention without pharmacist involvement. No heterogeneity (P = 0.786, I = 0%) was observed in this analysis owing to the small number of studies included.

Figure 2

Impact of pharmacist involvement in immunization rate of all types of vaccine by role. Abbreviation used: DL, DerSimonian and Laird method.

Pharmacists as advocators

Pooled analysis of 10 RCTS,,,,40, 41, 42, 43, 44, 45 (n = 175,550) (Figure 2) for pharmacists as advocators demonstrated a statistically significant increase in the immunization rate (RR 1.31 [95% CI 1.17–1.48]) compared with usual care or other intervention without pharmacist involvement. However, a high heterogeneity was observed in this analysis (P < 0.001, I = 89.6%).

Subgroup analyses

For influenza vaccine, pharmacist involvement as immunizer (RR 1.14 [95% CI 1.11–1.17], I = 0%) or advocator (1.19 [1.07–1.32], I = 83.0%) also significantly increased the immunization rates (Table 3; Supplemental Materials, Supplement V). For community pharmacy setting, pooled analysis of 6 studies,,,,, (n = 172,453) (Figure 3) demonstrated that immunization intervention programs with pharmacist involvement as advocators or both advocators and immunizers significantly increased the immunization rate (1.17 [1.06–1.28], I = 86.3%). Subgroup analysis for influenza vaccine in community setting based on 6 RCTs,,,,, also demonstrated a statistically significant increase in vaccine rate (1.17 [1.06–1.28], I = 84.5%) (Table 4; Supplemental Materials, Supplement VI). For hospital setting, pooled analysis of 4 RCTs,,, (n = 3097) (Figure 3) with pharmacists as advocators demonstrated a statistically significant increase in immunization rate (2.82 [1.13–7.03], I = 92.6%). High heterogeneity was observed in this analysis attributed to the inclusion of a study with high ROB. The removal of this study resulted in statistically significant results (3.74 [2.67–5.22], I = 0%) with no evidence of heterogeneity. A similar finding was observed for influenza vaccine rate at a hospital setting (4.78 [0.93–24.58], I = 40.1%) (Table 4). There was only one study at primary health care center, and no statistically significant impact was found for this setting.

Table 3

Summary of meta-analyses by pharmacist role

Role of pharmacist	Primary analysis		Sensitivity analyses
			#1—excluding high ROB trial		#2—excluding trial with both role		#3—adherence to treatment
	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)
All vaccines
Immunizer	2	(1.14 [1.12–1.15]; 0%)	2	(1.14 [1.12–1.15]; 0%)	1	NAa	NAb	NAb
Advocator	10	(1.31 [1.17–1.48]; 89.6%)	5	(1.47 [1.19– 1.81]; 93.3%)	9	(1.52 [1.25–1.85]; 90.5%)	6	(1.28 [1.05–1.56]; 79.1%)
Influenza vaccines
Immunizer	2	(1.14 [1.12–1.15]; 0%)	2	(1.14 [1.12–1.15]; 0%)	1	NAa	NAb	NAb
Advocator	8	(1.19 [1.07–1.32]; 83.0%)	4	(1.13 [1.0–1.28]; 76.5%)	7	(1.31 [1.08–1.59]; 84.2%)	5	(1.37 [1.01–1.85]; 89.3%)

Abbreviations used: RR, risk ratio; ROB, risk of bias.

Meta-analysis was not performed owing to limited number of studies.

No available data for adherence-to-treat principle.

Figure 3

Impact of pharmacist involvement in immunization rate of all types of vaccine by study setting. Abbreviation used: DL, DerSimonian and Laird method.

Table 4

Summary of meta-analyses by study setting

Study setting	Primary analysis, # studies		Sensitivity analyses
			#1—excluding high ROB trial		#2—excluding trial with both role		#3—adherence to treatment
	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)	# studies	(RR [95% CI]; I2)
All vaccines
Community pharmacy	6	(1.17 [1.06–1.28]; 86.3%)	2	(1.09 [0.99–1.2]; 89.1%)	5	(1.24 [1.03–1.50]; 88.9%)	3	(1.19 [0.95–1.48]; 77.8%)
Hospital	4	(2.82 [1.13–7.03]; 92.6%)	3	(3.74 [2.67–5.22]; 0%)	4	(2.82 [1.13–7.03]; 92.6%)	3	(2.59 [0.81–8.21]; 86.5%)
Primary health care center	1	NAa	1	NAa	1	NAa	NAb	NAb
Influenza vaccines
Community pharmacy	6	(1.17 [1.06–1.28]; 84.5%)	2	(1.11 [1.03–1.19]–73.8%)	5	(1.23 [1.02–1.47]; 86.5%)	3	(1.16 [0.88–1.53]; 92%)
Hospital	2	(4.78 [0.93–24.58]; 40.1%)	2	(4.78 [0.93–24.58]; 40.1%)	2	(4.78 [0.93–24.58]; 40.1%)	2	(5.16 [1.38–19.20]; 26.2%)
Primary health care center	1	NAa	1	NAa	1	NAa	NAb	NAb

Abbreviations used: NA, not applicable; RR, risk ratio; ROB, risk of bias.

Meta-analysis was not performed owing to limited number of studies.

No available data for adherence-to-treat principle.

Sensitivity analyses

For pharmacists as immunizers, when excluding the study with pharmacists serving as both immunizers and advocators, no sensitivity analyses were performed because of a limited number of studies. For pharmacists as advocators, sensitivity analyses by excluding the high ROB trials,,,, and a study with pharmacists as both immunizers and advocators resulted in statistically significant findings (RR, 21.47 [95% CI, 1.19–1.81], 93.3%) and (1.52 [1.25–1.85], 90.5%) respectively that continued to favor the intervention with pharmacists as advocators (Table 3). Sensitivity analysis by applying adherence-to-treatment principle to 6 RCTs,,,40, 41, 42 also demonstrated an increase in immunization rate (1.28 [1.05–1.56], 79.1%) (Table 3).

The effects of pharmacist involvement on other outcomes

One RCT assessed the impact of a 6-month tailored training program for pharmacists in community pharmacies on confidence of pharmacists in pneumococcal and herpes zoster vaccine recommendation and administration and pharmacists' perceived support for these vaccines. The study indicated that the intervention had a positive impact on the outcomes which in turn may increase the immunization activities within the community setting. Another RCT assessed the effect of different messaging strategies (pneumonia prevention, pneumonia vaccine costs, vaccine safety, community, and family duty) for the pneumococcal vaccination to the population on the patients’ favorable attitude toward vaccines and intent to consult with the pharmacist about them. Results indicated that the message on fatality, safety, and duty to family and community increased the intent to vaccinate by 25%. A study assessed impact of a pharmacist-physician online coaching intervention about COVID-19 vaccines on vaccine hesitancy and resistance. Findings demonstrated that the intervention improved participants’ attitude and knowledge toward COVID-19 vaccines and reduced significantly the vaccine hesitancy and resistance rate.

Quality of evidence from RCT

Overall, the quality of evidence was low to moderate (Table 5). The quality of evidence for pharmacists as immunizers was graded as moderate quality. We also found a moderate quality of evidence for pharmacists as advocators in the community setting but only for the influenza vaccine.

Table 5

GRADE for primary and subgroup analyses

Outcome	Pharmacist role	Illustrative comparative risksa		RR (95% CI)	No participants, (No studies)	Quality of the evidence (GRADE)
		Assumed riska	Corresponding riska
		Usual care	Pharmacist intervention
Primary analysis by pharmacist role at any setting for all vaccines
Immunization rate of all vaccines	Immunizer	34 per 1000	39 per 1000	RR 1.14 (1.12–1.15)	135,350 (2 studies)	⨁⨁⨁◯b MODERATE
	Advocator	69 per 1000	90 per 1000	RR 1.31 (1.17–1.48)	175,550 (10 studies)	⨁⨁◯◯c LOW
Subgroup analysis by pharmacist role at community pharmacy
Immunization rate of all vaccines	Immunizer	NAd	NAd	NAd	NAd	NAd
	Advocator	43 per 1000	50 per 1000	RR 1.17 (1.06–1.28)	172,453 (6 studies)	⨁⨁⨁◯e MODERATE
Immunization rate of influenza vaccines	Immunizer	NAd	NAd	NAd	NAd	NAd
	Advocator	49 per 1000	56 per 1000	RR 1.17 (1.06–1.28)	150,946 (6 studies)	⨁⨁⨁◯e MODERATE
Subgroup analysis by pharmacist role at hospital
Immunization rate of all vaccines	Immunizer	NAd	NAd	NAd	NAd	NAd
	Advocator	7 per 1000	26 per 1000	RR 2.82 (1.13–7.03)	3097 (4 studies)	⨁⨁⨁◯e MODERATE
Immunization rate of influenza vaccines	Immunizer	NAa	NAa	NAa	NAa	NAd
	Advocator	7 per 1000	33 per 1000	RR 4.78 (0.93–24.58)	162 (2 studies)	⨁⨁◯◯f LOW

Abbreviations used: GRADE, Grading of Recommendation Assessment, Development and Evaluation; RR, risk ratio; NA, not applicable.

The basis for the assumed risk (eg. the median or mean of control group risk across studies). The corresponding risk is based on the assumed risk in the comparison group and the relative effect of the intervention.

Moderate-quality evidence is caused by indirectness (different study setting – community pharmacy and primary health care).

Low-quality evidence is caused by inconsistency (high heterogeneity) and publication bias.

Meta-analysis was not performed owing to limited number of studies.

Moderate-quality evidence is caused by inconsistency (high heterogeneity).

Low-quality evidence is caused by imprecision (very small sample size).

Evidence from observational studies

Vaccination rate was increased in all 64 studies comparing the pharmacists’ involvement with usual care or other intervention without pharmacists. We performed pooled analyses from observational studies to investigate the impact of pharmacists in immunization activities on vaccine uptake and found consistent findings with meta-analyses from RCTs (Supplemental Materials, Supplement VII). Pooled analysis of 9 observational studies47, 48, 49, 50, 51, 52, 53, 54, 55, 56 (n = 2,676,385) for pharmacists as immunizers demonstrated a statistically significant increase in immunization rate (RR 2.17 [95% CI 1.71–2.75], I = 97.7%), favoring the intervention compared with usual care or other intervention without pharmacist involvement. Pooled analysis of 17 studies,,,50, 51, 52, 53, 54, 55, 56, 57, 58, 59, 60, 61, 62, 63 (n = 2,643,385) for pharmacists as advocators demonstrated a statistically significant increase in the immunization rate (2.01 [1.66–2.44], I = 99%) compared with usual care or other intervention without pharmacist involvement. A similar finding was observed for influenza vaccine rate with pharmacists as immunizers (3 studies49, 50, 51, n =5561) (2.15 [1.16–4.02], I = 98.2%) and advocators (6 studies,,,,,, n = 11,331) (2.02 [1.37–2.98], I = 92.1%). Pooled analyses by study setting indicated that pharmacist participation in immunization at any setting significantly increased the immunization rate of all type of vaccines (Supplemental Materials, Supplement VIII). The quality of evidence was assessed from low to very low (Supplemental Materials, Supplement IX). For other outcomes, interventions with the participation of pharmacists had an impact on the improvement of vaccine appropriateness (selection of vaccine based on individual patient criteria), vaccine compliance, patients’ awareness and attitude toward vaccination, and patients’ satisfaction. A study estimating cost per vaccine administration indicated that the average direct costs per adult immunization were lower in pharmacies compared with physician offices or other medical settings by 16%-26% and 11%-20%, respectively. A reference list of included observational studies can be found at Supplemental Materials, Supplement X.

Heterogeneity exploration

We explored the source of heterogeneity by conducting subgroup analyses by vaccine types (all vaccines vs. influenza vaccines) and study settings (Tables 3 and 4; Supplemental Materials, Supplements VII and VIII) but could not identify the reasons. Sources of heterogeneity were also explored using univariate meta-regression for RCTs and observational studies of the following variables: pharmacist role, intervention, comparison, study setting, type of vaccines, number of vaccines investigated, country, and ROB. None of them were found to explain heterogeneity with the I ranging from 97%-99% in meta-regressions (Supplemental Materials, Supplements XI and XII). We also generated a L’Abbé plot (Supplemental Materials, Supplements XIII and XIV) and found that differences in the relative effect of intervention across studies were not associated with differences across control groups.

Discussion

Vaccination is one of the most effective public health interventions to mitigate the burden of disease and saves millions of lives per year. Several studies have demonstrated the favorable impact of pharmacists, one of the most accessible health professionals, on immunization.,,,,66, 67, 68, 69 We performed a systematic review and meta-analysis of pharmacist involvement in immunization rates (facilitators, advocators, or immunizers) compared with usual care or other intervention/services without pharmacists. Findings from both RCTs and observational studies indicated that the involvement of pharmacists in the immunization process regardless of their roles or vaccine provided had a substantial impact on immunization rate. Evidence from RCTs demonstrated that pharmacist participation in vaccination activity in both community and hospital settings had a positive impact on immunization rate, particularly for influenza vaccine in community settings. Evidence from observational studies indicated that pharmacist interventions improved the immunization uptake of all vaccines including influenza vaccine at any setting. We performed a series of sensitivity analyses for both RCTs and observational studies and observed that our results were robust. All sensitivity analyses indicated statistically significant impact of pharmacist involvement on immunization rate that were in line with the main finding (Tables 3 and 4; Supplemental Materials, Supplements VII and VIII). Pharmacist involvement also had a favorable effect on other outcomes such as vaccine appropriateness, vaccine compliance, vaccine hesitancy, patient awareness and attitude toward immunization, and patient satisfaction, which in turn may increase the vaccine uptake.

The results of this study using data from the most recent trials were consistent with the findings from previous meta-analyses, and addressed the importance of pharmacists’ participation in improving public health issues such as immunization. Although the evidence was from high-income countries, mostly from the United States, the positive impact of pharmacists on vaccine uptake suggested the benefit of expanding the scope of pharmacist practice in terms of vaccine administration and immunization advocating activities at a variety of settings at the global scale. The global shifting trend from providing product-centered services to patient-centered services such as immunization has been happening in pharmacist practice for years.

Pharmacist involvement in immunization process varies globally. The United States is advanced in involving pharmacists in immunization process such as hosting patients (which means providing venue for patients coming for vaccination), storing vaccines, or communicating with physicians and nurses since the mid-1800s, but it was more than a decade that pharmacists’ role has progressed from being vaccine facilitator and advocator to becoming vaccine immunizers. Today, all states allow trained pharmacists to administer vaccines. Following the United States, some developed countries, including Canada, United Kingdom, New Zealand, Portugal, Ireland, and Australia, have authorized pharmacists to administer vaccines. Recently, in 2018, health professionals from 20 countries gathered in a conference on “Pharmacy-based interventions to increase vaccine uptake” in Venice to present evidence-based review of vaccination administration authorization for pharmacists and discuss opportunities for further expansion of pharmacist role in immunization globally. They reported higher immunization rates in countries that authorized pharmacists to administer vaccines such as United States, Canada, United Kingdom, Ireland, and Portugal compared with countries that did not authorize. Other countries such as Estonia, Croatia, Spain, and Malta shared their benefits from the additional participation of pharmacists in immunization activities.72, 73, 74 These countries are accompanied with the high visibility and accessibility of community pharmacies, and pharmacists are one of the first health professionals individuals turn to when seeking health care. Thus, in these countries, pharmacists can play a critical role in the prevention, control, and management of high-incidence vaccine-preventable infections and to assist during disease outbreaks and pandemics. They can easily identify patients at higher risk and specific target groups for vaccination, providing necessary counseling and actively participating in reminder and recall systems to ensure that vaccination schedules are met., Although such services have been provided by community pharmacists in these high-income countries, other countries, especially low- and middle-income countries (LMICs), have started to explore and expand the scope of pharmacist practice in public health, especially in immunization. In these LMICs, although pharmacists have participated in patient management care in community pharmacy or primary health care settings, the role of pharmacists in immunization has not been well defined as yet, mostly owing to regulatory obstacles, financial shortage, and lack of professional training programs on immunization for pharmacists., This finding favors the implementation of pharmacy-based immunization in other countries, especially ones in low-resource settings where immunization programs rely on public health institutions (hospitals, medical centers or clinics, etc.) or individual physicians to deliver vaccines. Although providing immunization services at public health institutions is effective in reaching children, older people, or people with chronic diseases within their medical visits, adult populations in remote areas who do not receive routine health care services are likely to have poor accessibility to immunization services. A lack of coordination between vaccination and curative health services and incomplete vaccination during vaccination visits were reported as the causes of missed opportunities for vaccines in public health facilities. Fees for immunizations create an important barrier to vaccinations in public facilities in low-resource setting. Moreover, in these developing countries, hospitals and clinics are often overburdened leading to long waits and missed opportunities for vaccination. Those countries would benefit from additional participation by pharmacists in the immunization process.,

The 2021-2025 Global Alliance for Vaccines and Immunization 5-year strategy has 4 goals, in which the equity goal focus is to “help countries extend immunization services to regularly reach under-immunized and zero-dose children to build a stronger primary health care platform.” Based on our review, given their accessibility, qualification, and experience in patient management care, pharmacists, especially those who work in community pharmacies or primary health care settings in developing countries, would play the critical role to achieve this global immunization effort.

In the context of COVID-19 pandemic, pharmacies worldwide are one of the few places that are kept open for public service even during the strict lockdowns. Community pharmacists are a vital health care provider during the outbreak and are highlighted the essential role of pharmacists as part of the response to the pandemic. To addressing the vaccination efforts, pharmacists were the first profession targeted for expanded scope of practice in the United States. The Federal Retail Pharmacy Program for COVID-19 Vaccination is a collaboration among the federal government, states and territories, and 21 national pharmacy partners and independent pharmacy networks to increase access to COVID-19 vaccination across the United States.

Strengths of this review include the comprehensive search strategy with the inclusion of studies in previous meta-analyses and the most data from recent trials and evidence from observational studies to provide updated findings. Moreover, besides the main outcome, which was immunization uptake, this study assessed the impact of pharmacist involvement in the immunization process on other related outcomes.

Limitations

There are several limitations to this review. Owing to a small number of RCTs for other related outcomes, we were unable to pool the findings. Another limitation was the high heterogeneity of the advocating interventions. We explored the source of inconsistency by the study setting, type of vaccine, or type of intervention. Heterogeneity was affected by these factors but, overall, it was still high for pharmacist as advocators. In addition, there were several studies that were graded as having a high or critical ROB. However, we do not think it affected our findings given that all sensitivity analyses excluding these studies resulted in statistically significant findings that continued to favor the intervention with pharmacists for both RCTs and observational studies. Finally, the findings may not apply for all health care systems given that the data for this study were mostly from the United States and some high-income and upper-middle-income countries. The impact may vary for different countries, but it poses the potential benefit of the implementation of pharmacy-based immunization for other countries, especially low-resource ones.

Conclusion

Pharmacist involvement as immunizer, advocator, or both roles has favorable effects on immunization uptake, especially strong evidence for influenza vaccine. In addition, interventions with pharmacist involvement also had an impact on other related outcomes (patient attitude toward vaccines, pharmacist confidence in vaccine recommendation and administration, vaccine compliance and appropriateness, and patient satisfaction), which indirectly improves the vaccine coverage. Pharmacists could play a key role in public health responses, such as what they have demonstrated with the COVID-19 epidemic, help address concerns with vaccine hesitancy, and have a positive impact on immunization uptake during any future pandemics. Pharmacists have the potential to play an important role in increasing access to vaccines and improving coverage, yet evidence of their role in vaccinations remains limited in LMICs. LMICs should try to expand the role of pharmacists as advocators or immunizers to offer vaccination services based on the present findings. Greater documentation of pharmacist involvement in vaccination services in LMICs is needed to demonstrate the value of successful integration of pharmacists in immunization programs.

No	Search term	Ovid Medline	Embase
1	exp Pharmacist/	17448	83013
2	Pharmacist$.tw.	34668	77231
3	exp Pharmacy/	8483	20311
4	Pharmacy.tw.	42434	87045
5	exp Pharmacies/	8125	20311
6	Pharmacies.tw.	11672	21010
7	exp Pharmaceutical Services/	71991	20311
8	(Pharmaceutical service).ti,ab.	186	387
9	exp care, pharmaceutical/	71991	20148
10	(Pharmaceutical cares).ti,ab	4	18
11	exp Community Pharmacy Services/	4801	20311
12	exp pharmacy service/	71991	26341
13	exp Pharmacy Services, Hospital/	11811
14	(Pharmac$ adj2 care$).ti,ab.	5935	12450
15	exp Immunization/	179753	332651
16	exp Vaccination/	87558	196139
17	exp Vaccines/	234480	382516
18	immunis$.tw.	12429	16608
19	immuniz$.tw.	134352	173562
20	vaccin$.tw.	325136	409780
21	exp Mass vaccination/	3123	4273
22	exp immunotherapy, active/	89041	270
23	exp immunization, secondary/	8225	174
24	exp immunization schedule/	10945	332651
25	or/1-14	128414	213909
26	or/15-24	503354	671021
27	25 AND 26	2630	4410
28	Limit 27 to humans	2192	3856
29	Limit 28 to (dt=19460101-20210131)	2190

No.	Search term	CENTRAL
1	Mh Pharmacist	20
2	Pharmacist:ti,ab,kw	3793
3	Mh Pharmacy	165
4	Pharmacy:ti,ab,kw	4875
5	Mh Pharmacies	3
6	Pharmacies:ti,ab,kw	1070
7	Mh “Pharmaceutical Services”	1
8	“Pharmaceutical service”:ti,ab,kw	9
9	Mh “pharmaceutical care”	3
10	“pharmaceutical care”:ti,ab,kw	745
11	Mh “Community Pharmacy Services”	2
12	“Community pharmacies”:ti,ab,kw	494
13	Mh “pharmacy service”	3
14	Mh “Pharmacy Services,/Hospital”	0
15	“Pharmacy service”:ti,ab,kw	214
16	Mh Immunization	85
17	“Immunization schedule”:ti,ab,kw	1351
18	“Immunization, Secondary”:ti,ab,kw	917
19	“Immunotherapy, Active”:ti,ab,kw	103
20	Mh Vaccination	171
21	Mh Vaccines	156
22	“Mass vaccination”:ti,ab,kw	118
23	Mh immunis∗	4
24	immuniz∗:ti,ab,kw	8276
25	vaccin∗:ti,ab,kw	25711
26	Mh vaccin∗	171
27	or/1-15	8047
28	or/16-26	27274
29	27 AND 28	138

Supplement VII

Summary of meta-analyses of observational studies by pharmacist role

Role of pharmacist	Primary analysis		Sensitivity analyses
			#1- excluding critical ROB studies		#2 - excluding studies with both role
	# studies	(RR [95%CI]; I2)	# studies	(RR [95%CI]; I2)	# studies	(RR [95%CI]; I2)
All vaccines
Immunizer	9	(2.17 [1.71, 2.75]; 97.7%)	6	(2.19 [1.47, 3.26]; 98%)	3	(2.74 [1.77,4.23]; 97%)
Advocator	17	(2.01 [1.66, 2.44]; 99%)	12	(1.86 [1.50, 2.31]; 96.8%)	11	(2.15 [1.65, 2.81]; 97.1%)
Both	6	(1.78 [1.47,2.15]; 92.4%)	3	(1.73 [1.02,2.93]; 93.7%)	-	-
Influenza vaccines
Immunizer	3	(2.15 [1.16, 4.02]; 98.2%)	2	(2.76 [1.39, 5.49]; 98.5%)	2	(2.76 [1.39,5.49]; 98.5%)
Advocator	6	(2.02 [1.37, 2.98]; 92.1%)	3	(1.97 [1.03, 3.75]; 91.5%)	5	(2.37 [1.45, 3.82]; 93.2%)
Both	1	NA	0	NA	-	-

Note: ROB, risk of bias; NA, insufficient data

Supplement VIII

Summary of meta-analyses of observational studies by study setting

Study setting	Primary analysis		Sensitivity analyses
			#1- excluding critical ROB studies		#2 - excluding studies with both role
	# studies	(RR [95%CI]; I2)	# studies	(RR [95%CI]; I2)	# studies	(RR [95%CI]; I2)
All vaccines
Community pharmacy	6	(2.97 [1.92, 4.58]; 98.4%)	4	(2.91 [2.18, 3.88]; 88.6%)	3	(2.79 [1.93, 4.03]; 92.4%)
Hospital/Medical center	9	(2.04 [1.53, 2.72]; 97.7%)	7	(1.75 [1.31, 2.35]; 97.6%)	8	(2.37 [1.67, 3.36]; 97.6%)
Clinic	4	(1.67 [1.21, 2.31]; 89.2%)	4	(1.67 [1.21, 2.31]; 89.2%)	3	(1.89 [1.33, 2.68]; 89.9%)
Other	1	NA	1	NA	0	NA
Influenza vaccines
Community pharmacy	2	(3.32 [2.11, 5.22]];65.5%	2	(3.32 [2.11, 5.22]];65.5%	2	(3.32 [2.11, 5.22]];65.5%
Hospital/Medical center	3	(3.84 [1.66, 8.88]; 91.3%)	1	NA	3	(3.84 [1.66, 8.88]; 91.3%)
Clinic	2	(1.53 [0.95, 2.46]; 95.6%)	2	(1.53 [0.95, 2.46]; 95.6%)	2	(1.53 [0.95, 2.46]; 95.6%)
Other	1	NA	1	NA	0	NA

Note: ROB, risk of bias; NA, insufficient data

Supplement IX

Grading of Recommendation Assessment, Development and Evaluation (GRADE) for analyses of observational studies

Outcome	Pharmacist role	Illustrative comparative risks1		Relative risk(95% CI)	Number of participants(No. of studies)	Quality of the evidence (GRADE)
		Assumed risk1Usual care	Corresponding risk1Pharmacist intervention
Primary analysis by pharmacist role at any setting for all vaccines
Immunization rate of all vaccines	Immunizer	16 per 1000	34 per 1000	2.17(1.71, 2.75)	2,676,385(9 studies)	⨁⨁◯◯2LOW
	Advocator	15 per 1000	30 per 1000	2.01(1.66, 2.44)	2,643,474(17 studies)	⨁⨁◯◯2LOW
Subgroup analysis by pharmacist role at any setting for influenza vaccine
Immunization rate of influenza vaccines	Immunizer	104 per 1000	223 per 1000	2.15(1.16, 4.02)	5,561(3 studies)	⨁⨁◯◯2LOW
	Advocator	22 per 1000	44 per 1000	2.02(1.37, 2.98)	11,331(6 studies)	⨁◯◯◯3VERY LOW

Note: CI, confidence interval; RR, risk ratio.

The basis for the assumed risk (eg. the median or mean of control group risk across studies). The corresponding risk is based on the assumed risk in the comparison group and the relative effect of the intervention.

Low quality evidence is due to inconsistency (high heterogeneity) and large effect size.

Very low quality evidence is due to inconsistency (high heterogeneity & risk of bias).

Supplement XI

Meta regression of all RCTs of pharmacist involvement on immunization rate of all vaccines compared to usual care/non-pharmacist-involved intervention

Analysis	No. of RCTs	No. of comparisons	RR	95%CI	I-squared
Immunizer	2	2	1.14	1.12, 1.15	0%
Advocator	10	10	1.31	1.17,1.48	89.6%
Variable (univariate)	No. of studies	No. of comparisons	RR	Standard error	p-value	I-squared
Pharmacist role1	11	12	1.37	1.59	0.513	88.4%
Intervention2	11	12	0.89	1.42	0.751	88.3%
Comparison3	11	12	0.95	1.66	0.930	88.2%
Study setting4	11	12	1.40	1.32	0.257	87.1%
No of vaccine investigated in study5	11	12	1.09	1.54	0.835	88.2%
Country6	11	12	1.07	1.42	0.847	87.9%
Risk of bias7	11	12	0.85	1.41	0.636	86.6%
Vaccine type8	11	12	0.73	1.53	0.480	87.7%

Note:

Pharmacist role as immunizer or advocator;

Intervention: pharmacist direct vs indirect involvement with patients;

Comparison: Usual care or other control without pharmacist;

Study setting: community pharmacy or hospital;

Number of vaccines investigated by the study;

Country: USA or non-USA;

Risk of bias assessed as Moderate, Some concerns or High;

Vaccine type: Influenza or Non-influenza vaccine.

Supplement XII

Meta regression of all observational studies of pharmacist involvement on immunization rate of all vaccines compared to usual care/non-pharmacist-involved intervention

Analysis	No. of studies	No. of comparisons	RR	95%CI	I-squared
Immunizer	9	9	2.17	1.71, 2.75	97.7%
Advocator	17	17	2.01	1.66,2.44	99.0%
Both	6	6	1.78	1.47,2.15	92.4%
Variable (univariate)	No. of studies	No. of comparisons	RR	Standard error	p-value	I-squared
Pharmacist role1	20	26	1.34	1.62	0.551	98.7%
Intervention2	20	26	1.93	1.73	0.243	97.7%
Comparison3	20	26	1.56	1.41	0.209	98.7%
Study setting4	20	26	0.72	1.23	0.133	97.7%
No of vaccine investigated in study5	20	26	0.73	1.85	0.614	98.9%
Country6	20	26	0.96	2.19	0.957	98.9%
Risk of bias7	20	26	1.23	1.35	0.496	98.2%
Vaccine type8	20	26	0.83	1.62	0.698	98.8%

Note:

Pharmacist role as immunizer or advocator or both;

Intervention: pharmacist direct vs indirect involvement with patients;

Comparison: Usual care or other control without pharmacist;

Study setting: community pharmacy or hospital/medical center or clinic or other;

Number of vaccines investigated by the study;

Country: USA or non-USA;

Risk of bias assessed as Moderate, Serious or Critical;

Vaccine type: Influenza or Non-influenza vaccine.