Table 4.
Factors associated with revising a prescription in response to an alert
| Alerts seen=807 | Descriptive statistics |
Regression analysis* |
||||||||
|---|---|---|---|---|---|---|---|---|---|---|
| Over-ride |
Revise |
Bivariate |
Multivariate |
|||||||
| (n=727) |
(n=80) |
OR | 95% CI | p Value | OR | 95% CI | p Value | |||
| Alert characteristics | ||||||||||
| Severity of alert | n | % | n | % | ||||||
| Most severe | 264 | 36 | 51 | 64 | 2.21 | 1.48 to 3.29 | <0.01 | 2.03 | 1.39 to 2.98 | <0.01 |
| Moderate or less severe | 463 | 64 | 29 | 36 | Reference | Reference | Reference | Reference | Reference | Reference |
| Type of alert | ||||||||||
| Drug–age contraindication | 263 | 36 | 30 | 38 | Ref | Ref | Ref | Ref | Ref | Ref |
| Cumulative side effects | 104 | 14 | 8 | 10 | 0.97 | 0.73 to 1.30 | 0.86 | 1.06 | 0.70 to 1.60 | 0.79 |
| Drug–disease contraindication | 128 | 18 | 16 | 20 | 0.94 | 0.64 to 1.38 | 0.75 | 0.74 | 0.44 to 1.23 | 0.24 |
| Therapy duplication | 94 | 13 | 8 | 10 | 0.97 | 0.57 to 1.67 | 0.92 | 1.14 | 0.47 to 2.79 | 0.77 |
| Drug interaction | 42 | 6 | 1 | 1 | 0.67 | 0.49 to 0.93 | 0.02 | 0.49 | 0.22 to 1.08 | 0.08 |
| Excess dose | 75 | 10 | 13 | 16 | 0.98 | 0.50 to 1.92 | 0.95 | 0.67 | 0.34 to 1.34 | 0.26 |
| Renal dose adjustment | 10 | 1 | 2 | 3 | 1.55 | 1.07 to 2.25 | 0.02 | 1.64 | 1.09 to 2.46 | 0.02 |
| Drug or potential cross allergy | 8 | 1 | 1 | 1 | 0.93 | 0.49 to 1.74 | 0.81 | 0.66 | 0.37 to 1.19 | 0.17 |
| Other | 3 | 1 | 1 | 1 | 1.42 | 0.67 to 3.02 | 0.36 | 1.41 | 0.80 to 2.49 | 0.24 |
| Patient characteristics | ||||||||||
| Female | 475 | 65 | 50 | 63 | 1.02 | 0.70 to 1.47 | .94 | – | – | – |
| Male | 252 | 35 | 30 | 38 | Ref | Ref | Ref | |||
| Mean | SD | Mean | SD | |||||||
| Age (years) | 74.5 | 6.2 | 74.5 | 6.5 | 1.02 | 0.99 to 1.04 | 0.25 | – | – | – |
| Risk factors for fall-related injuries | n | % | n | % | ||||||
| Fall-related Injury | 32 | 4.4 | 5 | 6.3 | 0.89 | 0.35 to 2.26 | 0.80 | – | – | – |
| Gait and balance problems | 212 | 29.2 | 29 | 36.3 | 1.18 | 0.78 to 1.78 | 0.44 | – | – | – |
| Lower extremity weakness | 61 | 8.4 | 7 | 8.8 | 1.31 | 0.68 to 2.51 | 0.42 | – | – | – |
| Cognitive impairment | 103 | 14.2 | 12 | 15 | 1.67 | 1.03 to 2.70 | 0.04 | 1.95 | 1.13 to 3.36 | 0.02 |
| Number of ambulatory visits (year prior to visit) (OR per 5 visit increase) | 16.2 | 23.2 | 18.6 | 34.1 | 1.02 | 0.99 to 1.06 | 0.19 | 1.05 | 1.004 to 1.09 | 0.03 |
| Number of active medications (at visit) | 10.8 | 4.9 | 10.9 | 4.8 | 1.01 | 0.97 to 1.06 | 0.52 | |||
| Physician characteristics | ||||||||||
| Demographic characteristics: | n | % | n | % | ||||||
| Female | 119 | 16 | 25 | 31 | 1.63 | 0.73 to 3.65 | 0.23 | – | – | – |
| Male | 608 | 84 | 55 | 69 | Reference | Reference | Reference | |||
| Mean | SD | Mean | SD | |||||||
| Practice experience (OR per 5 year increase) | 24.2 | 5.4 | 24.7 | 6.7 | 0.96 | 0.60 to 1.55 | 0.88 | – | – | – |
| Practice Characteristics | ||||||||||
| Daily practice volume (OR per 5 patient increase) | 20.4 | 6.1 | 19.3 | 6.8 | 088 | 0.65 to 1.02 | 0.38 | – | – | – |
| Percentage of patients 65 years and older (OR per 10% increase) | 19.5 | 9.1 | 19 | 8.3 | 0.83 | 0.44 to 1.56 | 0.56 | – | – | – |
| Experience and skills related to electronic prescribing | ||||||||||
| Electronic prescription speed (minutes per 3 e-RXs) (per minute increase) | 5.5 | 1.6 | 5.8 | 1.8 | 1.18 | 0.90 to 1.55 | 0.23 | – | – | – |
| Electronic prescription rate (e-RXs per 100 visits) (per 10 e-RX increase) | 31.4 | 12.4 | 28.8 | 15.9 | 0.84 | 0.65 to 1.10 | 0.20 | 0.86 | 0.66 to 1.13 | 0.27 |
| Alert level setting | n | % | n | % | ||||||
| View severe alerts only | 323 | 44.4 | 48 | 60 | 1.80 | 0.91 to 3.57 | 0.09 | 1.11 | 0.54 to 2.28 | 0.78 |
| View serious and severe alerts only, or view all alerts | 404 | 55.6 | 32 | 40 | Reference | Reference | Reference | Reference | Reference | Reference |
*The above dataset contained 807 alerts clustered within 303 patients, who were clustered within 47 physicians. For patients clustered within physicians, minimum to maximum cluster size was 1 to 141, respectively. Alternating logistic regression was used to account for two levels of clustering (alerts within patients; patients within physicians).