Table 1.
Characteristics and outcome of included randomized controlled trials
| First Author, year [reference] | Sample from population | Vaccine | Intervention category | Intervention vs. comparison | Total sample (intervention/comparison) | Effect measure | Risk of bias |
|---|---|---|---|---|---|---|---|
| Arthur, 2002 [52] | Patients aged over 75 from a large rural general practice in Leicestershire. | Influenza | Reminder/recall & outreach | Personal letter of invitation to attend for vaccination vs. health check at home where vaccine was offered. | 2052 (1372/680) | OR 1.28 (CI 1.03-1.58). | High risk of performance bias (no blinding) and attrition bias (high decline of health checks). |
| Hull, 2002 [53] | Patients aged 65-74 from at 3 general practices in London & Essex. | Influenza | Reminder/recall | Telephone call from receptionist to make appointment vs. letter & info leaflet alone. | 1318 (660/658) | Adj OR 1.29 (CI 1.03-1.62) for phone call (p=0.026) | Low risk of bias. |
| Nuttall, 2003 [54] | Previous non-attenders aged 65-90 from a general practice in Lancashire. | Influenza | Reminder/recall & outreach | Letter vs. leaflet vs. letter & home visit to discuss vaccination. | 90 (30/30/30) | OR 0.84 (0.26-2.74; p=0.77) [50] | Generally low, but risk of performance bias (no blinding) |
| Herrett, 2016 [55] | Patients aged 18-64 in risk groups from 153 general practices in London. | Influenza | Reminder/recall | Sending of pre-defined, recommended text message reminders vs. usual care. | 102257 (77 practices/ 79 practices) | OR 1.12 (CI 1.00-1.25) | Low risk of bias. |
| Porter-Jones, 2009 [58] | All children eligible for 1st MMR from Flintshire, Wales. | MMR | Patient education | Teddy bear with t-shirt directing parents to information sources vs. no bear | 974 (542/432) | OR 1.06 (CI 0.73-1.57) | High risk of selection bias (convenience sampling, no allocation concealment) and performance bias (lack of blinding). |
| aShourie, 2013 & Tubeuf 2014 [59, 74] | First-time parents with a child aged 3-12 months eligible for MMR from 512 general practices in the North of England. | MMR | Patient education | Web-based decision aid vs. leaflet & usual practice vs. usual practice alone. | 220 (48/85/70) | Non-significant difference due to small sample: leaflet vs. usual practice OR 0.14 (CI 0.02-1.14);); and decision aid vs. leaflet OR 10.6 (CI 0.1-188.5); decision aid vs usual practice 2.1 (CI 0.1-52.5) | Generally low, however small groups and lack of blinding. |
| Siriwadena, 2002 [56] | Patients aged >65 and those in eligible risk groups from 30 general practices in Trent region. | Influenza & pneumococcal | Healthworker education | Educational visit to GP practices based on principles of academic detailing lasting one hour vs. provision of information on performance alone. | 30 practices (15/15) | Increases in uptake of pneumococcal in patients with CVD (1.23, CI 1.13-1.34; p=<0.001) and diabetes (1.18, CI 1.08-1.28; p<0.001). No difference in any group for influenza | Generally low, but unclear risk of selection bias (randomisation not described) and high risk of performance bias (analysts not blinded) |
| Mantzari, 2015 [67] | Females aged 16-18 eligible for HPV vaccine in Birmingham and East North health region. | HPV vaccine | Incentive & reminder/recall | Vouchers worth £45 for completing vaccine course of 3 injections and text message reminders vs. invitation letter alone. | 1000 (500/500) | First dose: OR 1.63 (CI 1.08-2.47) for first time and 1.63 (1.075-2.472) for previous non-attenders. Third dose: OR 2.15 (CI 1.32-3.96) for first time and 4.28 (CI 1.92-9.55) for previous non-attenders. | Unclear risk of detection bias (possible for analysis to identify groups) otherwise low. |
MMR measles, mumps and rubella vaccine, HPV human papillomavirus, CI 95% confidence interval, Adj adjusted, OR odds ratio, CVD cardiovascular disease
aboth studies report analysis and results from the same sample