Baseline demographic and health characteristics of the community-dwelling cohort, before and after propensity score weighting
| Baseline characteristics | Before propensity score weighting | After propensity score weighting | ||||
| New opioid users | Controls | Standardised difference# | New opioid users | Controls | Standardised difference# | |
| Subjects | 89 327 | 41 652 | 89 224 | 41 930 | ||
| Age | 76.9±7.0 | 77.3±7.6 | 0.06 | 77.0±7.0 | 77.0±7.6 | 0.00 |
| Female | 47.3 | 47.5 | 0.00 | 47.4 | 47.1 | 0.01 |
| Low income based on ODB flag within 1 year prior to the index date | 24.2 | 21.8 | 0.06 | 23.4 | 23.4 | 0.00 |
| Income quintile | ||||||
| 1 (lowest) | 23.4 | 22.5 | 0.02 | 23.1 | 23.4 | 0.01 |
| 2 | 22.3 | 21.9 | 0.01 | 22.1 | 22 | 0.00 |
| 3 | 19.5 | 20.0 | 0.01 | 19.7 | 19.7 | 0.00 |
| 4 | 18.0 | 18.4 | 0.01 | 18.1 | 17.9 | 0.01 |
| 5 (highest) | 16.5 | 17.0 | 0.01 | 16.7 | 16.6 | 0.00 |
| Missing data | 0.3 | 0.3 | 0.00 | 0.3 | 0.3 | 0.00 |
| Rural setting | 16.7 | 17.4 | 0.02 | 16.9 | 16.9 | 0.00 |
| COPD exacerbation frequency in past year | ||||||
| 0 exacerbations | 62.2 | 60.9 | 0.03 | 61.9 | 61.7 | 0.00 |
| ≥1 outpatient exacerbations | 19.4 | 17.7 | 0.04 | 18.8 | 18.7 | 0.00 |
| ≥1 exacerbations requiring presentation to hospital | 18.4 | 21.3 | 0.08 | 19.3 | 19.7 | 0.01 |
| COPD exacerbation in past 30 days | 7.9 | 12.4 | 0.15 | 9.2 | 9.0 | 0.01 |
| Duration of COPD prior to index date | ||||||
| <2 years | 22.5 | 38.9 | 0.37 | 27.7 | 28.0 | 0.01 |
| 2–5 years | 19.6 | 15.9 | 0.10 | 18.4 | 18.5 | 0.00 |
| >5 years | 57.9 | 45.2 | 0.26 | 53.9 | 53.5 | 0.01 |
| Medications in past 180 days | ||||||
| Short/long-acting β agonists | 36.0 | 34.6 | 0.03 | 35.6 | 36.1 | 0.01 |
| Short/long-acting anticholinergics | 39.2 | 37.9 | 0.03 | 39.0 | 39.5 | 0.01 |
| Inhaled corticosteroids | 14.0 | 13.6 | 0.01 | 13.9 | 13.9 | 0.00 |
| Combination inhaled corticosteroid-long acting beta agonist inhalers | 33.2 | 30.1 | 0.07 | 32.3 | 32.7 | 0.01 |
| Oral corticosteroids | 13.8 | 11.8 | 0.06 | 13.2 | 13.5 | 0.01 |
| Theophylline | 2.4 | 1.9 | 0.03 | 2.3 | 2.3 | 0.00 |
| Respiratory antibiotics | 44.9 | 40.7 | 0.09 | 43.6 | 44.2 | 0.01 |
| Any outpatient visits in the past 12 months | 99.6 | 99.4 | 0.03 | 99.5 | 99.5 | 0.00 |
| Any hospitalisations in the past 12 months | 46.2 | 41.7 | 0.09 | 45.0 | 46.5 | 0.03 |
| Any ICU admissions in the past 12 months | 6.6 | 6.7 | 0.00 | 6.8 | 7.5 | 0.03 |
| Any surgery in the past 12 months | 14.1 | 5.0 | 0.29 | 11.3 | 11.8 | 0.02 |
| Total number of non-opioid drugs received in the past year | 11.4±5.9 | 10.2±5.7 | 0.20 | 11.1±5.8 | 11.2±6.1 | 0.03 |
| Incident benzodiazepine receipt in past 30 days | 1.9 | 2.6 | 0.05 | 2.1 | 2.2 | 0.00 |
| Prevalent benzodiazepine receipt in past 30 days | 13.2 | 10.8 | 0.07 | 12.5 | 12.8 | 0.01 |
| Charlson score | ||||||
| 0 | 29.8 | 29.6 | 0.00 | 29.6 | 28.8 | 0.02 |
| 1 | 18.8 | 17.6 | 0.03 | 18.5 | 18.4 | 0.00 |
| 2 | 13.7 | 12.7 | 0.03 | 13.5 | 13.7 | 0.01 |
| ≥3 | 22.0 | 20.1 | 0.05 | 21.6 | 22.6 | 0.03 |
| Missing data | 15.7 | 19.9 | 0.11 | 16.9 | 16.4 | 0.01 |
| Ischaemic heart disease¶ | 47.5 | 42.9 | 0.09 | 46.1 | 46.5 | 0.01 |
| Congestive heart failure¶ | 26.8 | 27.2 | 0.01 | 27.0 | 27.5 | 0.01 |
| Any malignancy¶ | 13.0 | 7.7 | 0.17 | 11.4 | 12.1 | 0.02 |
| Musculoskeletal/connective tissue disease¶ | 86.9 | 78.2 | 0.24 | 84.2 | 84.4 | 0.01 |
| Osteoporosis¶ | 16.0 | 15.1 | 0.02 | 15.8 | 15.7 | 0.00 |
| Psychotic psychiatric disease¶ | 5.5 | 5.6 | 0.01 | 5.6 | 5.7 | 0.01 |
| Non-psychotic psychiatric disease¶ | 50.5 | 47.4 | 0.06 | 49.6 | 49.8 | 0.00 |
| Sleep disorder¶ | 55.2 | 53.0 | 0.04 | 54.6 | 54.9 | 0.01 |
| Dementia¶ | 11.2 | 13.8 | 0.08 | 12.0 | 12.1 | 0.00 |
| Cohort entry in flu season | 38.1 | 38.7 | 0.01 | 38.2 | 38.1 | 0.00 |
Data are presented as mean±sd or %, unless otherwise stated. ODB: Ontario Drug Benefit; COPD: chronic obstructive pulmonary disease; ICU: intensive care unit. #: standardised differences of >0.10 are thought to indicate potentially meaningful differences. ¶: presence of all comorbidities was based on 5-year look-back from the index date.