In recent years physicians have made greater use of newer, more expensive antibiotics to treat ear infections in children. This trend is of concern to public health policymakers because 1) it increases costs and 2) the new drugs, which are active against a wider range of bacterial species, may lead to the rapid development of antimicrobial resistance. Using individual-level data, we use a mixed multinomial logit model to examine the impact of drug and individual characteristics on physicians' antibiotic choice. Our dataset covers the period from 1980 to 1998, allowing us to observe how prescribing patterns have changed over time in response to resistance. We find that the new drugs are better along a number of dimensions valued by physicians and patients, but their superior characteristics cannot account entirely for their increased use. An important omitted factor driving use is antimicrobial resistance, and we estimate that resistance increases antibiotic spending on ear infections by $28 million annually. Many previous studies have attempted to quantify the negative consequences associated with antimicrobial resistance by looking at patient morbidity and death. Our study highlights the fact that even in the absence of adverse impacts on patient health, resistance may burden the health care system by inducing physicians to substitute towards more expensive drugs.
Learning Objectives: 1. Understand the impact of antimicrobial resistance on antibiotic utilization. 2. Identify factors driving use of new, expensive drugs. 3. Develop familiarity with new statistical methods for modeling treatment choice.
Keywords: Antibiotic Resistance, Prescription Drug Use Patterns
Presenting author's disclosure statement:
Organization/institution whose products or services will be discussed: None
I do not have any significant financial interest/arrangement or affiliation with any organization/institution whose products or services are being discussed in this session.