Hospitals enrolled in the Centers for Disease Control and Prevention’s (CDC’s) National Healthcare Safety Network (NHSN) are required to complete an annual survey on facility practices. The Colorado Emerging Infections Program participates in CDC’s Healthcare-Associated Infections (HAI) and Antimicrobial Use Prevalence Survey (PS). Participating hospitals completed a Healthcare Facility Assessment (HFA) describing facility practices using the most up-to-date information available. A comparison of survey responses was performed to evaluate the surveys’ ability to accurately describe hospital infection control practices (ICPs) and antibiotic stewardship practices (ASPs), as well as to assess changes in hospital ASPs over time.
METHODS:
Survey responses from the 2015 PS HFA were obtained for 16 participating Colorado hospitals. Similarly, responses from the 2015 NHSN annual survey—covering 2014 calendar year data—were obtained for the same hospitals. Survey questions were matched between surveys on language and intent; questions that did not match were excluded. Two methods were used in analysis: 1) responses were evaluated for validity, assuming practices remained the same between survey years; 2) temporal changes in ASPs were assessed from 2014 to 2015, assuming responses from both surveys were accurate. Calculations were performed using SAS 9.3.
RESULTS:
Twenty questions were matched between surveys. Using the first method, 5 (25%) questions had 51-75% concordance among hospitals, and 10 (50%) questions had >75% concordance. Complete survey concordance occurred with questions pertaining to antibiograms and ICPs. Of 16 hospitals, 13 (81%) had 51-75% concordance between surveys, and 3 (19%) had >75% concordance. No hospital had complete concordance across all survey questions. Using the second method, 11 (68.8%) hospitals reported utilizing fewer ASPs in 2015 than in 2014, including fewer hospitals using educational programs for prescribers (n=12, 75%), pharmacist participation (n=9, 56.3%), and administrative support (n=9, 56.3%). More hospitals (n=14, 87.5%) reported regular antibiotic review from 2014 to 2015.
CONCLUSIONS:
Although both surveys evaluated comparable topics, the two methods demonstrate that the surveys were either not reliable in obtaining accurate responses for the same questions, or that a majority of hospitals reduced ASPs from 2014 to 2015. Only 3 (19%) hospitals demonstrated >75% concordance across surveys, and 11 (68.8%) hospitals showed waning participation in ASPs over time. In order to better capture accurate ICP and ASP data, surveys should standardize language and definitions for improved comprehension, and hospitals should designate one person to complete surveys to ensure accurate responses and diminish respondent bias.