Skip to main content
. 1999 Oct 9;319(7215):943–948. doi: 10.1136/bmj.319.7215.943

Table 1.

Rates of follow up by baseline risk characteristics

Variables 4 months
12 months
Rate (%) % difference (95% CI) Rate (%) % difference (95% CI)
No of patients
Intervention (n=316) 204 (65)  9.9 (−1.3 to 21.0) 169 (54) 8.4 (−4.7 to 21.5)
Control (n=567) 422 (74) 351 (62)
Cigarette smoking
Non-smokers:
 Intervention (n=192) 145 (76)  6.4 (−4.8 to 17.6) 129 (67)  3.5 (−10.5 to 17.6)
 Control (n=287) 235 (82) 203 (71)
Smokers:
 Intervention (n=124)  59 (48) 19.2 (4.1 to 34.4)  40 (32) 20.6 (7.9 to 33.3)
 Control (n=280) 187 (67) 148 (53)
Cholesterol concentration (mmol/l)
<6.5:
 Intervention (n=161)  89 (55) 18.1 (6.7 to 29.4)  68 (42) 15.7 (2.7 to 28.7)
 Control (n=345) 253 (73) 200 (58)
6.5-9.0:
 Intervention (n=153) 115 (75)  3.6 (−8.9 to 16.2) 100 (65) 5.4 (−8.5 to 19.3)
 Control (n=212) 167 (79) 150 (71)
Body mass index and exercise
<25:
 Intervention (n=64)    36 (56.3)  9.6 (−7.0 to 26.1)    33 (51.6)  0.1 (−14.2 to 14.4)
 Control (n=120)    79 (65.8)    62 (51.7)
25-35 with low physical activity:
 Intervention (n=252)   168 (66.7) 10.1 (−0.8 to 20.9)   136 (54.0)  10.7 (−4.3 to 25.7)
 Control (n=447)   343 (76.7)   289 (64.7)

Positive differences indicate greater proportions of control than intervention patients retained in study.