Challenges in outcomes ascertainment

Pharmacoepidemiologists often rely on diagnostic or procedure codes in defining their study outcome. However, whether these codes or algorithms are valid or not needs further exploration. “Indeed, outcome definition is a pivotal step in designing research using claims databases and a very relevant point commonly raised by journal reviewers” said Adel Alrwisan, M.S., a graduate student at department of pharmaceutical outcomes and policy at the University of Florida College of Pharmacy.

In the literature, several studies have overcome this concern by cross validating identified outcomes with medical records. “Obtaining medical records could be a very lengthy process and prohibitively costly; thus might hinder researcher from attempting their study outcome validation” said Almut Winterstein, Ph.D., a professor and chair of pharmaceutical outcomes and policy at the UF College of Pharmacy.

Researchers at University of Florida used another approach by examining how concordant were the incidence of eardrum perforations identified by using Medicaid claims data and incidence estimates from longitudinal studies, as a gold standard. A study of tympanostomy tube complications in 47,000 children, eligible for Medicaid, revealed up to 3 percent of children suffered from persistent eardrum perforation. Furthermore, the study authors showed the impact of different definitions of persistent eardrum perforations on the sensitivity and specificity of the study outcome. The incidence varied from 0.5 to 12 percent with most and least restrictive outcome definition, respectively. The study was recently published in the journal of Otology and Neurotology.

“This study may open the door for more research in the area of medication safety and otic preparations, as we were able to demonstrate an algorithm for identifying persistent eardrum perforations that produce results consistent with those of longitudinal studies.” said Patrick Antonelli, M.D., professor and chair of the department of otolaryngology at the UF College of Medicine.

“To date, most of the validation studies focus has been on cardiovascular outcomes and deaths with limited attention to other outcomes such as otological events” Alrwisan said.

Alrwisan is working on Winterstein’s research team. He earned his master’s degree from University of Aberdeen in 2010 and joined the department of pharmaceutical outcomes and policy in 2012. His research interests mainly focuses on pharmacoepidemiology and health outcomes of antibiotics.