A retrospective cohort study found that adults with diabetes have a significantly higher risk of influenza-related care utilization compared to those without diabetes. Early vaccination provided substantial protection regardless of diabetes status.
The study, published in BMJ Open Diabetes Research & Care, examined 1,117,236 Louisiana adults' electronic medical records from 2012 to 2016. Each influenza season included 137,447 to 409,343 patients, 30% to 35% with diabetes. Participants were predominantly White (62% to 70%) or Black (28% to 36%), with median ages of 48 to 64 years.
Patients with diabetes had 3.7 times more frequent influenza-related inpatient or emergency department visits compared to those without diabetes, after adjusting for demographics and comorbidities. Specifically, influenza events ranged from 0.0106-0.0143 events per person for patients with diabetes, compared to 0.0022-0.0049 for those without diabetes.
Early vaccination (before January 1) reduced the risk of influenza-related health care utilization by 66% for patients with diabetes and 67% for those without diabetes compared to later vaccination. Using an alternative analysis model, the reduction was 66% for those with diabetes and 78% for those without diabetes.
The study utilized two novel methodologies to assess vaccination effects while avoiding healthy user bias. In the first model, patients vaccinated early versus late were matched based on demographics and comorbidities. The second model compared vaccinated to not-yet-vaccinated patients within each month of the influenza season.
Influenza events were identified using ICD-9 and ICD-10 codes for influenza and pneumonia diagnoses during inpatient or emergency room visits. Events peaked during typical influenza seasons and correlated with CDC influenza-like illness surveillance data.
Women and adults aged 65 and older had higher rates of influenza-related health care utilization in both groups of patients with and without diabetes. Patients with diabetes tended to be older (46% to 49% over 65 years versus 20% to 24% over 65 years without diabetes) and had a higher proportion of males (44% versus 33% to 36% in patients without diabetes).
The protective effect of vaccination was consistent across age groups and sexes. For all-cause inpatient or emergency department visits, the relative risk reduction ranged from 13% to 36% across different influenza seasons. The reduction ranged from 41% to 75% for respiratory-related visits, and for influenza-specific visits, 43% to 82%.
The study period covered the switch to high-dose vaccination in adults 65 years or older during the 2015 to 2016 influenza season. However, no noticeable changes in vaccine efficacy were observed using the study's methodology.
Limitations included potential over- or underestimation of influenza events based on ICD coding and inability to directly compare vaccinated to unvaccinated patients due to incomplete vaccination records outside the studied health systems. The study population from Louisiana may not be representative of other areas.
The findings provide quantitative evidence supporting targeted early influenza vaccination campaigns for patients with diabetes to minimize severe disease outcomes and health care costs. Further research is needed to examine interactions between diabetes, obesity, and influenza vaccination impacts.
Funding for this study was provided by NIH grants and conducted in partnership with REACHnet; one author disclosed consulting relationships with pharmaceutical companies.