Do practices exposed to virology sampling training and a dashboard achieve higher rates of flu vaccination?

Problem

Seasonal influenza has a substantial health and economic impact. Influenza vaccination programs are implemented over a short period of time and are the primary prevention strategy. As part of the Public Health England flu surveillance system, we monitor both vaccination rates and influenza like symptoms in a nationally representative sample of primary care practices. Increasing vaccination rates has large impact on morbidity and mortality. We were able to compare vaccination rates between practices exposed to virology sampling training and provision of an online dashboard and those that were not.The data are current and representative of the influenza vaccination program in England.Aim: To determine whether if there was a difference between risk groups or overall between the two types of practice.

Approach

Patient influenza vaccinations for the period September to October 2018 were identified from Royal College of General Practitioners (RCGP) Research and Surveillance Centre (RSC), primary care sentinel network.Vaccination rates were compared between the practices exposed to virology sampling training and dashboard, those in the national representative network of practices not.

Findings

The analysis of vaccination rates in the virology sampling trained practices showed a difference in the rate of influenza vaccination for the whole population, 24.52% (178,788.00/729,090.00 : 24.13%-24.92%, 95% C.I.) for sampling practices compared to 17.98% (447,749.00/2,490,117.00 : 17.73%-18.23%, 95% C.I.) for practices that are not part of the influenza virology sampling scheme. This difference in rate was significant (p < 0.001). Similar differences where observed in the over 65 years population, 69.36% (104,216.00/150,253.00 : 69.07%-69.65%, 95% C.I.) and 55.32% (237,335.00/429,019.00 : 55.09%-55.55%, 95% C.I.), sampling practices and non-sampling practices respectively, (p <0.001).For at high risk patients, there was a similar difference in vaccination rate, 29.84% (32,095.00/107,552.00 : 28.91%-30.78%, 95% C.I.) vs. 28.86% (79,231.00/274,494.00 : 28.27%-29.46%, 95% C.I.) for sampling practices and non-sampling practices respectively.A similar difference in vaccination rate in children age 2 to 9 years was observed, however practices reported data quality issues resulting from variation in provision of vaccination records from local authorities.

Consequences

The rates achieved in the sampling practices compared to non-sampling practices indicates that significant improvement in vaccination rates in certain groups is feasible in primary care. In Europe 75% vaccination is stated as the level required for herd immunity, whilst in the USA they say 80%-90%. Process evaluation of these differences could lead to improvement of influenza vaccine coverage is achievable would reduce burden on the seasonal burden on both primary and secondary care.