Table 4.
The risk of involvement in a motor vehicle collision by medication exposure
| Medication | Number of individuals taking medication (N) | 1:4 Matched control periodsa | |||
|---|---|---|---|---|---|
| Model 1 OR (95 % CI) | Model 2 OR (95 % CI) | ||||
| Anticholesteremic | |||||
| Simvastatin | 12 | 1.00 | (0.12, 8.17) | 0.42 | (0.03,07.23) |
| Anticoagulants | |||||
| Clopidogrel | 10 | 10.73 | (1.19, 96.67) | 7.62 | (0.48, 122.10) |
| Warfarin | 10 | 0.72 | (0.06, 9.04) | 0.30 | (0.08, 1.22) |
| Anticonvulsants | |||||
| Gabapentin | 15 | 2.69 | (0.59, 12.32) | 1.32 | (0.24, 7.17) |
| Antidepressants | |||||
| Citalopram | 10 | 3.01 | (0.31, 29.65) | 3.21 | (0.24, 42.50) |
| Antihyperglycemics | |||||
| Insulin | 12 | 15.76 | (1.78, 139.61) | 2.63 | (0.14, 48.72) |
| Antihypertensive | |||||
| Furosemide | 12 | 0.50 | (0.08, 3.22) | 0.81 | (0.06, 10.62) |
| Hydrochlorothiazide | 13 | 12.35 | (1.35, 113.06) | 15.01 | (0.76, 296.60) |
| Lisinopril | 25 | 1.56 | (0.43, 5.69) | 0.27 | (0.05, 1.60) |
| Metoprolol | 29 | 5.29 | (1.31, 21.38) | 1.16 | (0.23, 5.79) |
| Muscle relaxants | |||||
| Albuterol | 11 | 0.44 | (0.06, 3.20) | 0.25 | (0.02, 3.34) |
| Narcotic analgesics | |||||
| Hydrocodone | 15 | 1.32 | (0.36, 4.92) | 0.37 | (0.04, 3.79) |
| Tramadol | 11 | 10.56 | (1.17, 95.51) | 11.41 | (1.27, 102.15) |
| Sleep medications | |||||
| Zolpidem | 10 | 4.20 | (0.73, 24.13) | 1.42 | (0.66, 3.00) |
| Steroids | |||||
| Fluticasone | 11 | 0.56 | (0.08, 3.89) | 0.41 | (0.04, 4.85) |
| Prednisone | 12 | 0.12 | (0.01, 1.07) | 0.19 | (0.02, 1.82) |
| Vasodilators | |||||
| Nitroglycerin | 12 | 2.54 | (0.42, 15.23) | 1.27 | (0.07, 23.82) |
| Other drugs | |||||
| Alendronate | 10 | 1.15 | (0.22, 6.04) | 0.13 | (0.01, 1.78) |
| Combination drugs | |||||
| Oxycodone and Acetaminophen | 16 | 0.61 | (0.18, 2.00) | 0.17 | (0.02, 1.63) |
aConditional logistic regression was used to calculate the odds ratios and 95 % CI. Each case’s medication exposure during the 14 day risk period immediately before the crash was matched to four separate control periods up to 1 year before the collision to assess if medication use during the risk period was associated with an increase of motor vehicle collision compared to control periods. Model 1 is the crude estimate (i.e. unadjusted) while Model 2 was adjusted for the number of medications a person was taking during each case and control period