Table 4.
Covariate balance before and after PS matching ibuprofen-paracetamol cohorts
| Before PS matching | After PS matching | |||||
|---|---|---|---|---|---|---|
| Ibuprofen cohort (%) | Paracetamol cohort (%) | ASMD | Ibuprofen cohort (%) | Paracetamol cohort (%) | ASMD | |
| Demographics | ||||||
| Female | 64 | 60.6 | 0.07 | 62.3 | 62.8 | − 0.01 |
| Age, years | ||||||
| 18–19 | 1.4 | 0.3 | 0.11 | 0.8 | 0.7 | 0.01 |
| 20–24 | 3.8 | 1.1 | 0.18 | 2.3 | 2.1 | 0.01 |
| 25–29 | 5 | 1.6 | 0.19 | 3.4 | 3.2 | 0.01 |
| 30–34 | 6.4 | 2.5 | 0.19 | 4.9 | 4.8 | 0 |
| 35–39 | 7.1 | 3.5 | 0.16 | 6.1 | 6.2 | 0 |
| 40–44 | 7.8 | 4.6 | 0.13 | 7.3 | 7.5 | 0 |
| 45–49 | 8.8 | 6.1 | 0.1 | 8.8 | 8.8 | 0 |
| 50–54 | 10.8 | 8.6 | 0.07 | 11.2 | 11.2 | 0 |
| 55–59 | 12.5 | 11.8 | 0.02 | 13.5 | 13.6 | 0 |
| 60–64 | 11.8 | 13.5 | − 0.05 | 13.2 | 13.4 | − 0.01 |
| 65–69 | 9.6 | 13.7 | − 0.13 | 10.9 | 11.1 | 0 |
| 70–74 | 6.9 | 12.3 | − 0.19 | 8 | 7.9 | 0 |
| 75–79 | 4.1 | 8.9 | − 0.2 | 4.8 | 4.8 | 0 |
| 80–84 | 4 | 11.5 | − 0.28 | 4.8 | 4.8 | 0 |
| Conditions | ||||||
| Amenorrhea, any time prior | 6.2 | 3 | 0.15 | 4.9 | 4.9 | 0 |
| Hypertension, any time prior | 51.4 | 68.3 | − 0.35 | 55.6 | 56.4 | − 0.02 |
| Hip pain, any time prior | 16.5 | 24.2 | − 0.19 | 17.6 | 18.3 | − 0.02 |
| Hyperlipidemia, any time prior | 39.9 | 53.7 | − 0.28 | 42.9 | 43.9 | − 0.02 |
| Nicotine dependence, any time prior | 16.6 | 16.3 | 0.01 | 16.2 | 16.9 | − 0.02 |
| Cataracts, any time prior | 13.2 | 21.5 | − 0.22 | 14.6 | 15.3 | − 0.02 |
| OA, any time prior | 19.9 | 29.1 | − 0.22 | 21.8 | 23 | − 0.03 |
| Hip OA, any time prior | 7.6 | 13.5 | − 0.19 | 8.6 | 8.8 | − 0.01 |
| Knee OA, any time prior | 27.3 | 39 | − 0.25 | 29.9 | 30.7 | − 0.02 |
| Osteoporosis, any time prior | 7.9 | 13.5 | − 0.18 | 8.9 | 9.1 | − 0.01 |
| Lumbar spine stenosis, any time prior | 13.5 | 26.9 | − 0.34 | 15.4 | 16.3 | − 0.02 |
| Low back pain, 6 months prior | 43 | 36.6 | 0.13 | 41.3 | 41.3 | 0 |
| Hip OA, 6 months prior | 3.6 | 6.5 | − 0.13 | 4.1 | 4.1 | 0 |
| Knee OA, 6 months prior | 13.9 | 18 | − 0.11 | 15.1 | 15.1 | 0 |
| Knee OA, 1 month prior | 5.9 | 9.3 | − 0.13 | 6.5 | 6.4 | 0 |
| Cerebrovascular disease | 1.9 | 4.1 | − 0.13 | 2.1 | 2.3 | − 0.01 |
| Chronic obstructive lung disease | 10.9 | 19.9 | − 0.25 | 12.4 | 13 | − 0.02 |
| Diabetes mellitus | 0.7 | 0.9 | − 0.03 | 0.7 | 0.8 | − 0.01 |
| Ischaemic heart disease | 0.2 | 0.3 | − 0.03 | 0.2 | 0.2 | 0 |
| Obesity | 25.9 | 26.1 | 0 | 25.6 | 26.1 | − 0.01 |
| Malignant neoplastic disease | 0.2 | 0.3 | − 0.03 | 0.2 | 0.2 | 0 |
| Procedures and visits | ||||||
| Knee arthroscopy, any time prior | 2.6 | 6.7 | − 0.2 | 3 | 3.2 | − 0.01 |
| DXA scan, any time prior | 13.3 | 20.5 | − 0.19 | 14.7 | 15.2 | − 0.01 |
| ECG, any time prior | 37.6 | 46.7 | − 0.18 | 39.1 | 40.1 | − 0.02 |
| Knee X-ray, any time prior | 14.9 | 21.7 | − 0.18 | 15.8 | 16.5 | − 0.02 |
| Manual/ physiotherapy, any time prior | 36.8 | 45.7 | − 0.18 | 38.1 | 39.1 | − 0.02 |
| Total knee replacement, any time prior | 0.4 | 2.1 | − 0.16 | 0.4 | 0.5 | − 0.01 |
| Arthrocentesis, 6 months prior | 5.8 | 8.2 | − 0.09 | 6.4 | 6.6 | − 0.01 |
| Emergency visit, 6 months prior | 15.4 | 13.5 | 0.05 | 12.8 | 13.4 | − 0.02 |
| Clinical index scoresa | ||||||
| CHADS2VASc | 2.02 | 2.85 | − 0.48 | 2.15 | 2.19 | − 0.03 |
| Diabetes Comorbidity Severity Index | 1.43 | 2.49 | − 0.41 | 1.57 | 1.65 | − 0.03 |
| Charlson index - Romano adaptation | 2 | 3.19 | − 0.41 | 2.14 | 2.27 | − 0.05 |
Select characteristics before and after PS matching, showing the (weighted) percentage of subjects with the characteristics in the Ibuprofen versus comparator cohorts, as well as the ASMD. Data shown from the OpenClaims, on treatment, February to October 2020 analysis. A complete covariate balance list can be found in the accompanying supplementary files, including drug covariates that were balanced at the ingredient level
ASMD absolute standardised mean difference, DXA dual-energy X-ray absorptiometry, ECG electrocardiogram, OA osteoarthritis, PS propensity score
aData for these covariates are given as a mean score for the cohort, not % of the cohort