Appendix 1b.
Results of Multinomial Logistic Regression Model for Predictors of Quarterly Undersupply and Oversupply of Antihyperlipidemic Medications (n=8,770)
| Odds Ratios and 95% Confidence Intervals | APP (95% CI) | ||
|---|---|---|---|
| Undersupply | Oversupply | Oversupply | |
| Sex | |||
| Male (ref) | -- | -- | 6.6% (5.0-8.2) |
| Female | 1.15 (0.96-1.39) | 0.74 (0.51-1.08) | 4.9% (3.8-5.9) |
| Race/ethnicity | |||
| White (ref) | -- | -- | 5.6% (4.6-6.5) |
| Black | 1.95 (1.32-2.89)** | 0.68 (0.24-1.95) | 3.1% (0.2-6.1) |
| Other/Unknown | 1.64 (1.08-2.50)* | 1.91 (0.79-4.63) | 8.4% (2.5-14.3) |
| Age and Insurance Status | |||
| HMO coverage only, any age (ref) | -- | -- | 2.5% (1.3-3.6) |
| Age <65 in Medicare only | 0.98 (0.68-1.42) | 3.64 (1.70-7.79)** | 8.0% (4.3-11.8) |
| Age <65 in Medicare and Medicaid | 0.97 (0.69-1.36) | 2.27 (1.08-4.77)* | 5.3% (2.7-7.9) |
| Age ≥65 in Medicare only | 0.87 (0.69-1.10) | 2.41 (1.34-4.32)** | 5.8% (4.3-7.2) |
| Age ≥65 in Medicare and Medicaid | 0.69 (0.47-1.01) | 5.15 (2.43-10.90)** | 11.7% (6.6-16.8) |
| Baseline ACG-PM score | |||
| Quartile 1 (ref) | -- | -- | 7.5% (4.7-10.3) |
| Quartile 2 | 1.02 (0.84-1.24) | 0.73 (0.45-1.16) | 5.7% (3.9-4.5) |
| Quartile 3 | 1.08 (0.86-1.35) | 0.85 (0.51-1.43) | 6.4% (4.8-8.1) |
| Quartile 4 | 1.33 (1.01-1.76)* | 0.56 (0.31-1.02) | 4.2% (3.0-5.4) |
| Diabetes Complications | |||
| No (ref) | -- | -- | 5.6% (4.5-6.6) |
| Yes | 1.24 (0.97-1.59) | 1.10 (0.68-1.77) | 5.7% (3.6-7.7) |
| Depression | |||
| No (ref) | -- | -- | 5.7% (4.7-6.8) |
| Yes | 1.15 (0.86-1.54) | 0.89 (0.50-1.57) | 4.9% (2.7-7.2) |
| Dementia | |||
| No (ref) | -- | -- | 5.5% (4.6-6.4) |
| Yes | 1.20 (0.76-1.90) | 1.37 (0.68-2.75) | 6.9% (3.0-10.8) |
| Alcohol or drug abuse | |||
| No (ref) | -- | -- | 5.7% (4.7-6.6) |
| Yes | 1.41 (0.73-2.72) | 0.44 (0.12-1.62) | 2.4% (0-5.4%) |
| Psychoses | |||
| No (ref) | -- | -- | 4.7% (3.9-5.6) |
| Yes | 0.64 (0.45-0.91)* | 2.77 (1.59-4.82)** | 12.6% (7.6-17.5) |
| Total number of diabetes-related medications at baseline | |||
| 1-2 (ref) | -- | -- | 5.5% (2.6-8.4) |
| 3-4 | 0.68 (0.52-0.91)** | 0.96 (0.49-1.86) | 6.0% (4.4-7.5) |
| 5-6 | 0.64 (0.47-0.87)** | 0.92 (0.46-1.84) | 5.8% (4.4-7.3) |
| ≥7 | 0.68 (0.46-0.99)* | 0.63 (0.27-1.43) | 4.1% (2.3-5.9) |
| Total number of unique medications (all types) at baseline | |||
| 0-5 (ref) | -- | -- | 3.3% (1.7-4.9) |
| 6-10 | 1.01 (0.78-1.30) | 1.78 (1.02-3.10)* | 5.6% (4.3-6.9) |
| 11-15 | 1.11 (0.81-1.53) | 1.83 (0.87-3.83) | 5.6% (3.4-7.8) |
| ≥16 | 0.89 (0.59-1.34) | 3.09 (1.40-6.80)** | 9.3% (5.5-13.1) |
| K-index for study period | |||
| Quartile 1 (ref) | -- | -- | 4.7% (2.9-6.5) |
| Quartile 2 | 0.99 (0.77-1.28) | 1.48 (0.87-2.53) | 6.7% (4.8-8.5) |
| Quartile 3 | 0.90 (0.70-1.15) | 1.09 (0.63-1.90) | 5.2% (3.7-6.8) |
| Quartile 4 | 0.73 (0.56-0.95)* | 1.13 (0.64-2.00) | 5.7% (3.8-7.6) |
| Insulin use, %(n) | |||
| No | -- | -- | 5.1% (4.0-6.1) |
| Yes | 0.93 (0.76-1.14) | 1.37 (0.93-2.01) | 6.8% (5.0-8.6) |
| Number of unique providers seen for E&M visits per quarter | |||
| 0 (ref) | -- | -- | 5.2% (3.9-6.6) |
| 1 | 0.84 (0.73-0.96)** | 1.03 (0.77-1.37) | 5.6% (4.5-6.8) |
| 2 | 0.94 (0.80-1.10) | 1.03 (0.74-1.42) | 5.5% (4.2-6.7) |
| ≥3 | 0.95 (0.79-1.14) | 1.12 (0.77-1.64) | 5.9% (4.5-7.3) |
| Any days in inpatient setting | |||
| No (ref) | -- | -- | 5.8% (4.8-6.7) |
| Yes (ref) | 2.35 (1.95-2.83)** | 0.97 (0.65-1.45) | 4.2% (2.7-5.7) |
| Any 90-day prescription fills in the quarter | |||
| No (ref) | -- | -- | 4.3% (3.5-5.2) |
| Yes (ref) | 0.79 (0.62-1.01) | 3.42 (2.40-4.89)** | 13.1% (9.9-16.3) |
| Last available LDL prior to quarter | |||
| <100 mg/dl (ref) | -- | -- | 6.1% (5.0-7.2) |
| 100-129 mg/dl | 1.52 (1.25-1.86)** | 1.12 (0.74-1.68) | 6.0% (4.0-8.0) |
| ≥130 mg/dl | 3.98 (3.02-5.24)** | 0.56 (0.23-1.35) | 2.1% (0.4-3.8) |
| Not tested | 1.48 (1.11-1.96)** | 0.91 (0.56-1.49) | 5.1% (3.0-7.1) |
*
p<.05
**
p<.01