Table 2.
Multilevel model of the relation between presence of trimethoprim resistant bacteria in individual patients' urine samples and drug prescribing and other variables at the practice level and the patient level (1995-6)
| Variables | Odds ratio (95% CI) | P value |
|---|---|---|
| Patient level factors | ||
| Older age (+10 years) | 1.15 (1.13 to 1.17) | <0.001 |
| Year (1996 v 1995) | 2.68 (2.14 to 3.35) | <0.001 |
| Sex (male v female) | 0.70 (0.59 to 0.82) | <0.001 |
| Sex × year | 1.42 (1.11 to 1.82) | 0.005 |
| Carstairs deprivation score | 0.99 (0.97 to 1.00) | 0.096 |
| More urine samples sent for analysis (+1) | 1.04 (0.99 to 1.10) | 0.106 |
| More trimethoprim prescriptions (+1) | 1.22 (1.16 to 1.28) | <0.001 |
| More prescriptions of other antibiotics (+1) | 1.18 (1.06 to 1.32) | 0.002 |
| Benzodiazepines prescriptions (yes v no) | 1.12 (0.68 to 1.83) | 0.658 |
| SSRI prescriptions (yes v no) | 0.67 (0.34 to 1.32) | 0.249 |
| HRT prescriptions (yes v no)* | 0.79 (0.64 to 0.97) | 0.028 |
| Oral contraceptive prescriptions (yes v no)* | 1.01 (0.82 to 1.23) | 0.954 |
| Practice level factors† | ||
| Medium practice size (v small) | 1.05 (0.90 to 1.22) | 0.552 |
| Large practice size (v small) | 1.09 (0.93 to 1.27) | 0.282 |
| Greater percentage of male GPs (+10%) | 0.87 (0.65 to 1.18) | 0.381 |
| Fundholding practice (yes v no) | 0.96 (0.84 to 1.09) | 0.545 |
SSRI=selective serotonin reuptake inhibitor. HRT=hormone replacement therapy.
*
Among female patients only.
†
Practice effect μij=0.007 (SE 0.006)