One hundred fifty-four patients were recruited over 3 years from 373 adults admitted with asthma. The most common reason for exclusion was that patients were admitted for < 24 h or the clinical researcher was not notified of the admission. Other reasons for exclusion are shown in Figure 1. Seventy-six patients were randomized to the doctor group, and 78 patients were randomized to the specialist nurse group. this
Analysis was by intention to treat. At clinic follow-up, 13 patients were found not to have asthma (many patients were not seen by a respiratory physician during their hospital stay), 4 patients withdrew from the study, and 1 patient was excluded because of participation in the pilot study. Thus, 66 patients (86.8%) in the doctor group and 70 patients (89.7%) in the nurse group were analyzed. When measurements were missing for particular variables, the patients were not included in that part of the analysis. The median age was 33 years (range, 17 to 83 years) in the nurse group, compared to 37 years (range, 17 to 91 years) in the doctor group (p = 0.17). The median time from inpatient admission to hospital discharge was 3 days in both groups (range, 1 to 14 days; p = 0.23). Similar numbers of patients had a general practitioner attendance or a referral to a psychologist in both groups.
Complete data were available for 65 patients in the doctor group and 68 patients in the nurse group. Thirty-two patients (49.2%) in the doctor group and 31 patients (45.6%) in the nurse group had exacerbations over the 6-month follow-up period, an odds ratio of 0.86 (95% CI, 0.44 to 1.71; p = 0.674). Altogether, there were 174 exacerbations recorded, 76 in the doctor group and 98 in the nurse group. There was no association between group and the number of exacerbations (p = 0.368, x2 test). The number of exacerbations per patient are shown in Figure 2; 10.5% (14 of 133 patients) had four or more exacerbations. The mean number of exacerbations per patient was 1.44 (98 exacerbations in 68 patients) in the nurse group, compared with 1.17 (76 exacerbations in 65 patients) in the doctor group. Using Poisson regression, this gives a relative risk of 1.23 (95% CI, 0.91 to 1.66; p = 0.368).
Figure 1. Patient flow through the study.
Figure 2. The number of exacerbations per patient in each study group.