Table 3

Secondary analysis of numbers and incident rate ratios for uptake of screening for tuberculosis on the mobile X-ray unit at intervention and control homeless hostels

Control
Eligible (number screened)
Intervention
Eligible (number screened)
TotalUnadjusted intervention group risk ratio (95% CI)Adjusted* intervention group risk ratio (95% CI)
Per protocol—peer educators who attended intervention hostel on day of screening1192 (503)1051 (432)2243 (935)0.97 (0.75 to 1.26)0.97 (0.78 to 1.22)
Hostel that did not participate effectively†748 (267)444 (137)1192 (404)0.86 (0.67 to 1.11)0.88 (0.67 to 1.14)
Hostel size
 43 or less beds362 (176)338 (134)700 (310)0.82 (0.60 to 1.11)0.80 (0.60 to 1.06)
 Greater than 43 beds830 (327)812 (334)1642 (661)1.04 (0.76 to 1.43)1.08 (0.82 to 1.42)
Historical screening uptake
 50% or less694 (272)718 (241)1412 (513)0.86 (0.64 to 1.14)0.86 (0.65 to 1.14)
 Greater than 50%498 (231)432 (227)930 (458)1.13 (0.85 to 1.51)1.13 (0.85 to 1.51)
  • Data are median (IQR) unless otherwise stated.

  • *Analysis adjusted for historical uptake rates and hostel bed size.

  • †One hostel (in intervention arm) did not have data collected on participation effectiveness. A score of greater than 13 was considered effective. The rating of hostel performance (results described in table 1) included scoring from 0 to 2 for elements of their effectiveness, including how involved the manager of the hostel was during screening, how much encouragement there was from staff, whether posters advertising the screening session were on display, how well-organised the screening event was and whether incentives were provided on the screening day. See online supplementary appendix for full details of each element included in the hostel effectiveness scoring system.