Does influenza vaccination status change physician ordering patterns for respiratory viral panels? Inspection for selection bias

Abstract

ABSTRACTPurpose: Hospitalized patients with an acute respiratory illness (ARI) were compared to determine if demographic characteristics, timing or influenza vaccination biased who received clinical respiratory viral panel (RVP) testing.Methods: 171 enrollees in an influenza vaccine effectiveness study and a sample of non-enrollees (N = 1029) admitted to a community hospital with ARI during December 2015 through April 2016 comprised the study sample. Those who received clinical RVP testing (n = 292) were compared to those who did not by age, sex, influenza vaccination status, and period (pre-peak influenza season vs. peak/post peak influenza season), using Chi square- and t-tests, and logistic regression.Results: Mean age of participants was 70 years, 58% was female and 45% had been vaccinated against influenza in the 2015–2016 season. Those with clinical RVP testing were significantly younger (67 years) than those without RVP (71 years; P < 0.001), but did not differ with respect to sex or vaccination status. The odds of clinical RVP testing were significantly (P = 0.004) related to younger age (< 65 years) (Odds ratio (OR) = 1.51; 95% CI = 1.14–2.00) and to later period (peak/post peak influenza season; OR = 2.64; 95% CI = 1.84–3.79) but were unrelated to influenza vaccination status or the interaction of time and vaccination status.Conclusion: RVP testing was significantly higher among younger hospitalized patients with an ARI and during the peak and post peak influenza periods than earlier in the season, but influenza vaccination status was not a significant factor. Studies that enroll based on clinical RVP testing should account for potential differences by age.