Effect of adherence to the written action plan on exacerbation recovery time and unscheduled healthcare utilisation
| Study definition* (n=217 exacerbations) | Adherence | Crude | Adjusted¶ | |||||
| Yes (n=87) | No (n=130) | B/OR | 95% CI | p Value | B/OR | 95% CI | p Value | |
| Exacerbation total recovery time, days | 10.9 (5.6) | 16.0 (10.7) | −5.08 | −7.62 to −2.55 | 0.0001 | −5.84 | −8.44 to −3.23 | 0.0001 |
| Treated recovery time, days | 10.5 (5.6) | 12.3 (9.0) | −1.8 | −4.03 to 0.42 | 0.112 | −2.21 | −4.58 to 0.16 | 0.072 |
| Unscheduled healthcare utilisation, yes§ | 42 (49.4) | 65 (51.2) | 0.93 | 0.54 to 1.61 | 0.801 | 0.94 | 0.49 to 1.83 | 0.864 |
| Additional definition† (n=217 exacerbations) | Yes (n=87) | No (n=130) | ||||||
| Exacerbation total recovery time, days | 11.2 (6.8) | 18.6 (11.1) | −7.43 | −9.91 to −4.96 | 0.0001 | −7.36 | −9.85 to −4.87 | 0.0001 |
| Treated recovery time, days | 10.7 (6.7) | 13.1 (9.3) | −2.38 | −4.66 to 0.1 | 0.040 | −2.04 | −4.37 to 0.29 | 0.090 |
| Unscheduled healthcare utilisation, yes§ | 39 (50.6) | 65 (50.0) | 0.97 | 0.67 to 1.72 | 0.928 | 0.99 | 0.52 to 1.96 | 0.978 |
| Worst case analysis‡ (n=288 exacerbations) | Yes (n=87) | No (n=201) | ||||||
| Exacerbation total recovery time, days | 10.9 (5.6) | 15.0 (10.2) | −4.09 | −6.46 to −1.72 | 0.001 | −4.87 | −7.31 to −2.43 | 0.0002 |
| Unscheduled healthcare utilisation, yes§ | 42 (49.4) | 109 (55.1) | 0.8 | 0.48 to 1.33 | 0.384 | 0.78 | 0.43 to 1.41 | 0.412 |
↵* Action plan adherence was defined as the initiation of both prednisone and antibiotics within 3 days of exacerbation onset. All other actions were defined as non-adherence.
↵† In the additional definition, adherence was defined as the initiation of prednisone and/or antibiotics within 3 days of exacerbation onset. All other actions were defined as non-adherence.
↵‡ In the worst case analysis, all reported exacerbations were included considering the exacerbations with missing data on the use of antibiotics and/or prednisone as failure to adherence. Adherence was defined as the initiation of prednisone and antibiotics within 3 days of exacerbation onset. Treated recovery time was not calculated as the exacerbations that were added had missing start dates of treatment.
↵§ Five exacerbations were excluded due to missing data concerning the use of unscheduled healthcare services.
↵¶ Adjusted for age, sex, post-bronchodilator forced expiratory volume in 1 s as percentage predicted, smoking status, exacerbation severity, contacting the case manager, cardiac disease, use of long-acting bronchodilators and inhaled corticosteroids, influenza vaccination, exacerbation sequence during follow-up and randomisation group in the randomised controlled trial.
B, regression coefficient (ie, change in recovery time with adherence).