Skip to main content
. 2018 Jul 27;33(11):1877–1884. doi: 10.1007/s11606-018-4592-6

Table 2.

Odds Ratio Estimates and 95% Confidence Intervals from Logistic Regression Models Using Lasso-Selected Variables Only for Fast and Timely Filling

Variable Fast filling (≤ 7 days); OR (95% CI) Timely filling (≤ 30 days); OR (95% CI)
Therapeutic area (antihypertensive) 1.20 (1.08, 1.33) NA
Therapeutic area (asthma/COPD) NA 0.72 (0.65, 0.79)
Asthma 0.69 (0.62, 0.77) 0.81 (0.75, 0.88)
Diabetes 0.71 (0.64, 0.79) 0.66 (0.61, 0.71)
Hypertension NA 0.83 (0.77, 0.90)
Congestive heart failure NA 0.80 (0.69, 0.94)
Hyperlipidemia 0.85 (0.77, 0.94) 0.93 (0.87, 1.00)
Number of refills 1.03 (1.01, 1.05) NA
Number of providers NA 0.93 (0.92, 0.95)
Number of current medications 1.30 (1.27, 1.33) 1.27 (1.25, 1.29)
Total number of medications NA 1.05 (1.04, 1.07)
Average out-of-pocket amount (per $10) 0.97 (0.96, 0.99) 0.98 (0.97, 0.99)
Age (per 5 years) NA 1.04 (1.02, 1.06)
Low-income subsidy in at least 1 month NA 1.28 (1.17, 1.39)
Median household income in zip code (per $10,000) NA 1.02 (1.01, 1.03)
Quantity (per 10 pills) 0.97 (0.96, 0.98) NA
New to therapeutic area NA 0.80 (0.68, 0.94)
Race: White NA 0.89 (0.81, 0.98)
Primary language: unavailable NA 0.69 (0.51, 0.94)
Body mass index: > 30 NA 0.82 (0.76, 0.88)
Prescriber is a primary care physician NA 1.08 (1.01, 1.15)
Dispense as written on index prescription NA 0.34 (0.26, 0.43)
Medication frequency: daily NA 1.77 (1.56, 2.01)
Medication frequency: > daily NA 1.09 (0.95, 1.24)
Interaction terms
 Female × relationship status: not in a relationship vs. others NA 1.05 (0.99, 1.13)
Abnormal HbA1c* × therapeutic area: diabetes NA 1.65 (1.22, 2.27)
Abnormal LDL × therapeutic area: hyperlipidemia NA 1.44 (1.22, 1.71)
Abnormal SBP × therapeutic area: hypertension 1.53 (1.25, 1.88) 1.54 (1.36, 1.75)
Number of refills × quantity (per 100 pills) NA 0.97 (0.95, 0.98)
Quantity (per 10 pills) × med frequency: > daily NA 0.96 (0.95, 0.97)

NA indicates that the variable was not selected by lasso for that outcome and was therefore not included in the relevant model

*Defined as HbA1c ≥ 8.0

†Defined as LDL ≥ 130

‡Defined as SBP ≥ 150