Examining the relationship between antibiotic prescribing and patient-experience survey results for acute adult bronchitis in an outpatient setting

Background: With antibiotic resistance on the rise, there is sustained interest in promoting antibiotic stewardship. Acute adult bronchitis (AAB) is an upper respiratory tract infection primarily of viral etiology. Because of this, evidence-based guidelines do not recommend the use of antibiotics in AAB. Despite this evidence, care providers continue to over-prescribe with concerns about patient expectations and patient satisfaction often cited as contributory factors.

Objectives: The objective of this study is to examine the relationship between antibiotic prescribing and patient-experience survey results.

Methods: Visit-level data were matched to patient satisfaction surveys for the first visit and first survey for each patient. Descriptive statistics and logistic regression were used. The dependent variable was a PS survey item asking patients to rate their care provider from 0 (worst) to 10 (best). A dichotomous variable was created using top box scores (9&10) vs. non-top box scores (0-8). This primary independent variable was used to group patients into three cohorts: (1) those who received an antibiotic prescription; (2) those who received only a non-antibiotic prescription; (3) those who received no prescription.

Results: A total of 49,638 visits coded for AAB were identified. With a match rate of 7.4%, 3,556 visits were included in the final sample. Of these visits, 84% resulted in an antibiotic prescription, 11% in a non-antibiotic prescription, and just 5% in no prescription. Descriptive statistics demonstrated that cohorts differed in age, sex, ethnicity, insurance, visit provider age, and visit provider type. A logistic regression model with multiple imputations for missing data found that receipt of a prescription did not predict receipt of a top box score. The results of this study indicate that receipt of an antibiotic prescription is not a predictor of high provider ratings, but that providers can potentially improve their ratings by expressing concern for the patient’s condition, taking more time with their patients, and spending patient time explaining diagnoses, treatment options, or reasons for non-treatment.