The Source of Purchased Medications and Its Impact on Medication Mistakes and Hospitalizations.

The study was observational, so it cannot show that medication source caused mistakes or hospitalizations; hospitalizations may have occurred before multiple sources were used. Medication difficulty, mistakes, and hospitalizations were self-reported, and admitting mistakes may have been underreported. Important confounders were not measured, including number and types of medications and reasons for hospitalization, and health literacy was not objectively measured. Few participants reported medication mistakes, and the medication-mistake model failed an assumption test, so those estimates were less stable. The analysis excluded people in nursing homes, proxy respondents, people who always needed help with medications, and people with dementia. The final sample was 3899 of 6312 initial individuals after exclusions and missing data, so the results apply to the selected group.

The article lists research support from NIH extramural and non-U.S. government sources. No author conflict-of-interest declaration is provided in the supplied text.

Do not conclude that using both pick-up and delivery causes medication mistakes or hospitalizations. The study was observational, hospitalizations may have preceded multiple medication sources, number and types of medications were not measured, and the medication-mistake association was marginal (p=0.05).