Does increasing demand, vaccination access, and provider activity increase influenza vaccination rates in people aged 60 years and older living in the community?
Vaccination rates vary across countries and socioeconomic and health risk groups.
The evidence is current to 7 December 2017.
We included three new trials (15,993 participants) for this update; the review now includes a total of 61 trials with 1,055,337 participants. All participants were aged 60 years or older, living in the community.
Study funding sources
Government health organisations funded 33 studies; foundations funded 9 studies; organisations that provided healthcare services in the studies funded 3 studies; and a pharmaceutical company offering free vaccines funded 1 study. Fifteen studies did not report any funding source.
Increasing community demand for vaccination (12 strategies, 41 trials, 767,460 participants)
Effective interventions consisted of reminders/recalls using letters and leaflets, and nurses or pharmacists educating and nurses vaccinating patients. Individual effective studies consisted of client outreach by retired teachers, receptionists, nurses, and medical students.
It was not possible to combine some interventions for analysis as they were too varied: 17 studies of simple reminders (11 with significant results); 16 studies of personalised reminders (12 with significant results); two studies of customised letters versus form letters (both with significant results); and four studies of health risk appraisals plus vaccination recommendations (all with significant results).
Improving vaccination access (6 strategies, 8 trials, 9353 participants)
Effective interventions consisted of home visits, client group clinic visits, and free vaccine offers.
Improving provision by providers or the healthcare system (11 strategies, 15 trials, 278,524 participants)
Effective interventions that could be combined for analysis included physician payment, physician reminders, clinic posters encouraging physician competition, and chart reviews plus benchmarking to rates of the top 10% of physicians. We could not analyse some groups of interventions: physician reminders (four studies, two of which were effective) and facilitator vaccination encouragement (three studies, two of which were effective).
Individual studies that were not effective consisted of posters plus postcards versus posters alone, educational reminders to physicians compared to mailed educational materials, educational outreach plus feedback to teams versus written feedback, and increasing staff vaccination rates.
No studies measured if interventions reduced illness or hospital admissions or reported societal-level interventions.
Quality of the evidence
Overall, we assessed the included studies as at moderate risk of bias. The overall GRADE assessment of the evidence was high to moderate quality.
We identified interventions that demonstrated significant positive effects of low (postcards), medium (personalised phone calls), and high (home visits, facilitators) intensity that increase community demand for vaccination, enhance access, and improve provider/system response. The overall GRADE assessment of the evidence was moderate quality. Conclusions are unchanged from the 2014 review.
The effectiveness of interventions to increase influenza vaccination uptake in people aged 60 years and older varies by country and participant characteristics. This review updates versions published in 2010 and 2014.
To assess access, provider, system, and societal interventions to increase the uptake of influenza vaccination in people aged 60 years and older in the community.
We searched CENTRAL, which includes the Cochrane Acute Respiratory Infections Group's Specialised Register, MEDLINE, Embase, CINAHL, and ERIC for this update, as well as WHO ICTRP and ClinicalTrials.gov for ongoing studies to 7 December 2017. We also searched the reference lists of included studies.
Randomised controlled trials (RCTs) and cluster-randomised trials of interventions to increase influenza vaccination in people aged 60 years or older in the community.
We used standard methodological procedures as specified by Cochrane.
We included three new RCTs for this update (total 61 RCTs; 1,055,337 participants). Trials involved people aged 60 years and older living in the community in high-income countries. Heterogeneity limited some meta-analyses. We assessed studies as at low risk of bias for randomisation (38%), allocation concealment (11%), blinding (44%), and selective reporting (100%). Half (51%) had missing data. We assessed the evidence as low-quality. We identified three levels of intervention intensity: low (e.g. postcards), medium (e.g. personalised phone calls), and high (e.g. home visits, facilitators).
Increasing community demand (12 strategies, 41 trials, 53 study arms, 767,460 participants)
One successful intervention that could be meta-analysed was client reminders or recalls by letter plus leaflet or postcard compared to reminder (odds ratio (OR) 1.11, 95% confidence interval (CI) 1.07 to 1.15; 3 studies; 64,200 participants). Successful interventions tested by single studies were patient outreach by retired teachers (OR 3.33, 95% CI 1.79 to 6.22); invitations by clinic receptionists (OR 2.72, 95% CI 1.55 to 4.76); nurses or pharmacists educating and nurses vaccinating patients (OR 152.95, 95% CI 9.39 to 2490.67); medical students counselling patients (OR 1.62, 95% CI 1.11 to 2.35); and multiple recall questionnaires (OR 1.13, 95% CI 1.03 to 1.24).
Some interventions could not be meta-analysed due to significant heterogeneity: 17 studies tested simple reminders (11 with 95% CI entirely above unity); 16 tested personalised reminders (12 with 95% CI entirely above unity); two investigated customised compared to form letters (both 95% CI above unity); and four studies examined the impact of health risk appraisals (all had 95% CI above unity). One study of a lottery for free groceries was not effective.
Enhancing vaccination access (6 strategies, 8 trials, 10 arms, 9353 participants)
We meta-analysed results from two studies of home visits (OR 1.30, 95% CI 1.05 to 1.61) and two studies that tested free vaccine compared to patient payment for vaccine (OR 2.36, 95% CI 1.98 to 2.82). We were unable to conduct meta-analyses of two studies of home visits by nurses plus a physician care plan (both with 95% CI above unity) and two studies of free vaccine compared to no intervention (both with 95% CI above unity). One study of group visits (OR 27.2, 95% CI 1.60 to 463.3) was effective, and one study of home visits compared to safety interventions was not.
Provider- or system-based interventions (11 strategies, 15 trials, 17 arms, 278,524 participants)
One successful intervention that could be meta-analysed focused on payments to physicians (OR 2.22, 95% CI 1.77 to 2.77). Successful interventions tested by individual studies were: reminding physicians to vaccinate all patients (OR 2.47, 95% CI 1.53 to 3.99); posters in clinics presenting vaccination rates and encouraging competition between doctors (OR 2.03, 95% CI 1.86 to 2.22); and chart reviews and benchmarking to the rates achieved by the top 10% of physicians (OR 3.43, 95% CI 2.37 to 4.97).
We were unable to meta-analyse four studies that looked at physician reminders (three studies with 95% CI above unity) and three studies of facilitator encouragement of vaccination (two studies with 95% CI above unity). Interventions that were not effective were: comparing letters on discharge from hospital to letters to general practitioners; posters plus postcards versus posters alone; educational reminders, academic detailing, and peer comparisons compared to mailed educational materials; educational outreach plus feedback to teams versus written feedback; and an intervention to increase staff vaccination rates.
Interventions at the societal level
No studies reported on societal-level interventions.
Study funding sources
Studies were funded by government health organisations (n = 33), foundations (n = 9), organisations that provided healthcare services in the studies (n = 3), and a pharmaceutical company offering free vaccines (n = 1). Fifteen studies did not report study funding sources.