Using Electronic Medical Record Data to Characterize the Level of Medication Use by Age-Groups in a Network of Primary Care Clinics

Abstract

Purpose: Our primary aim was to characterize the level of medication use across age-groups by examining electronic medical record data for a large number of patients receiving care in primary care clinics. A secondary aim was to identify factors associated with higher levels of medication use or polypharmacy. Methods: We conducted a retrospective query of electronic medical record data from a clinical data warehouse, evaluating 114 012 patients seen in primary care clinics at least once in the previous 6 months. Medication use was evaluated in 3 different categories: level 1 (0-4 medications), level 2 (5-9 medications), and level 3 (≥10 medications). Multivariate analysis was used to analyze different patient demographics and comorbidities for association with level of medication use. Results: At ages 18 to 24 years, 15% (male) to 23% (female) of patients were taking greater than 5 medications, a trend that continued to increase with older cohorts. Female patients were more likely to have level 2 (odds ratio [OR] = 1.76) and level 3 (OR = 2.73) use compared with men. Level 2 and level 3 use was associated with other patient characteristics, including number of patient encounters (level 2 OR = 2.99; level 3 OR = 8.08 for >7 encounters) and common chronic conditions such as chronic pain (level 2 OR = 2.56; level 3 OR = 6.40), diabetes (level 2 OR = 2.4; level 3 OR = 4.61), heart disease (level 2 OR = 1.99; level 3 OR = 3.65), hypertension (level 2 OR = 2.27; level 3 OR = 2.87), and dyslipidemia (level 2 OR = 1.82; level 3 OR = 2.12). Conclusion: Electronic medical record data may be an important tool for providing more comprehensive information regarding medication usage. Medication usage assessed by electronic medical records, even among the youngest cohort, appears to be greater than other sources of medication usage indicate. Higher levels of medication use were associated with a number of factors, including gender, body mass index, number of patient encounters, and comorbid conditions.