Rea 2004.
| Study characteristics | ||
| Methods | Cluster RCT; follow‐up: 12 months; control: conventional care | |
| Participants | Eligible: 158 Randomised: 135; I: 83, C: 52 Completed: 117 Mean age of both groups: 68 years Sex (% male) of both groups: 41.5 Inclusion criteria: COPD diagnosed by ICD‐9‐CM codes and GP records for a clinical diagnosis of moderate to severe COPD Major exclusion criteria for patients: chronic asthma, bronchiectasis, comorbidity more significant than COPD, unable to give informed consent, prognosis < 12 months, long‐term oxygen therapy or too unwell, deceased Major exclusion criteria GP: no longer enrolled with participating GP practice or moved out of area, unable to contact patient, insufficient practice nurse resource |
|
| Interventions | Chronic disease management programme was implemented including ‐ An action plan, which was implemented by patient's own GP and practice nurse, with advice from respiratory nurse and specialist physician. The plan comprised a timetable for regular maintenance checks and achievable goals set for lifestyle changes ‐ Patients visited the nurse monthly, the GP 3‐monthly and at other times if worsening symptoms demanded more visits ‐ Patients received education about smoking cessation, medication. Annual influenza vaccination and pulmonary rehabilitation were recommended Intervention duration: 12 months Disciplines involved: GP, nurse, pulmonologist |
|
| Outcomes | Health status, SF‐36, CRQ, shuttle walk test, spirometry, hospital admissions, medication, courses of oral steroids, courses of antibiotics, smoking cessation Randomisation at cluster level, analysis at patient level |
|
| Notes | Dominant component: self‐management/action plan and structured follow‐up by GP/nurse | |
| Risk of bias | ||
| Bias | Authors' judgement | Support for judgement |
| Random sequence generation (selection bias) | Low risk | Quote: "practices were randomised, using a set of computer‐generated numbers" |
| Allocation concealment (selection bias) | Unclear risk | Comment: no information available |
| Blinding of participants and personnel (performance bias) All outcomes | High risk | Comment: participants and healthcare providers not likely to have been blinded to group allocation |
| Blinding of outcome assessment (detection bias) All outcomes | High risk | Comment: healthcare providers involved in the programme administered outcome measurements at visit |
| Incomplete outcome data (attrition bias) All outcomes | Low risk | Comment: missing outcome data balanced between groups, with similar reasons for missing data across groups |
| Selective reporting (reporting bias) | Low risk | Comment: all outcomes reported |
| Recruitment bias | Low risk | Quote: "written information about the trial was provided to patients and consent was obtained before patients knew whether they belonged to an intervention or control practice" |
| Baseline imbalance between groups | High risk | Comment: no stratified or pair‐matched randomisation was used, resulting in baseline imbalance of 99 eligible patients in the intervention group and 59 patients in the control group |
| Loss to follow‐up of clusters | High risk | Quote: "after randomisation, two practices declined to participate, and in three, changes of either GP's or practice nurses prevented participation before enrolment had begun" |
| Adequate analysis methods for CRT | High risk | Comment: inadequate methods of analysis: randomisation done at level of GP practice, analysis performed at level of patients |