Table 3.
Before-after studies without a control evaluating effects of ACF on tuberculosis case notifications
| Country, population | Case-finding method | Diagnostic method | Co-interventions | Type of tuberculosis |
Person-years |
Number of tuberculosis cases |
CNR |
Reported estimates | |||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Baseline | Endline | Baseline | Endline | Baseline | Endline | CNR ratio | |||||||
| Corbett et al (2010)11 | Zimbabwe, general population | Door to door or mobile clinics in vans | Sputum smear if symptoms | None | Microbiologically confirmed | 55 216 | 322 093 | 154 | 1142 | 278·9 | 354·6 | 1·27 | No effect estimate provided for effect of ACF on CNR |
| Fatima et al (2014)27 | Pakistan, informal urban | Community mobilisation, mobile clinics | Sputum smear and clinician assessment if symptoms | Financial incentives to local providers, training to private general practitioners | Microbiologically confirmed | 9 067 658 | 9 067 658 | 8933 | 11 392 | 98·5 | 125·6 | 1·28 | No effect estimate provided for microbiologically confirmed cases; the proportion of smear-negative cases was reported to be significantly higher during the intervention |
| Lorent et al (2014)28 | Cambodia, informal urban | Door to door, community health workers collecting and transporting sputum | Sputum tests if symptoms (mainly smear, some culture or Xpert); clinician assessment with or without chest x-ray for some people | Laboratory strengthening | Microbiologically confirmed | 1 445 582 | 1 445 582 | 1610 | 2075 | 111·4 | 143·5 | 1·29 | Case notifications of bacteriologically confirmed tuberculosis increased from 1610 to 2075 (29% increase) |
| John et al (2015)29 | Nigeria, indigenous groups | Community mobilisation, mobile clinics | Sputum smear if symptoms; Xpert if negative sputum smear and symptoms persist | None | Microbiologically confirmed | 7 400 000 | 7 400 000 | 2436 | 3479 | 32·9 | 47·0 | 1·43 | New smear-positive notifications increased by 49·5% compared with the expected number based on historical trends |
| Maggard et al (2015)30 | Zambia, people in prison | Education within prison, mobile chest x-ray clinic | Chest x-ray and sputum smear regardless of symptoms | Laboratory strengthening, radiology equipment | All types | 5775 | 5775 | 138 | 409 | 2390 | 7082 | 2·96 | No effect estimate provided for effect of ACF on CNRs |
| Degner et al (2016)31 | USA, people in prison (compared two forms of ACF) | At entry to prison | Chest x-ray for all; sputum culture if chest x-ray abnormal; in baseline period, tuberculin skin test for all | None | All types | 30 000 | 35 000 | 8 | 37 | 26·7 | 105·8 | 3·96 | No effect estimate provided for effect of ACF on CNRs |
| Fatima et al (2016)32 | Pakistan, informal urban (neighbourhood contacts) | Door to door | Sputum smear if symptoms; Xpert if negative sputum smear and symptoms persist | Contact tracing | Microbiologically confirmed | 36 000 000 | 36 000 000 | 28 159 | 30 066 | 78·2 | 83·52 | 1·07 | Case detection of bacteriologically confirmed tuberculosis increased by 6·8% with intervention |
| Mallick et al (2017)33 | India, people in prison | Education, community mobilisation within prison | Sputum smear if symptoms | None | Microbiologically confirmed | 16 199 | 16 199 | 316 | 412 | 1951 | 2543 | 1·30 | CNR for all forms of tuberculosis increased by 38% in endline period compared with control period |
| Karamagi et al (2018)34 | Uganda, people in prison | Community mobilisation, door to door, community health workers collecting and transporting sputum | Sputum smear if symptoms | Contact tracing, facility-based screening | Microbiologically confirmed | NA* | NA | NA | NA | 171 | 212† | 1·24 | No effect estimate provided for effect of ACF on CNRs |
| Ford et al (2019)35 | India, remote rural | Community mobilisation, mobile chest x-ray units | Chest x-ray and sputum if symptoms | Change to national tuberculosis programme guidelines | Microbiologically confirmed | NA | NA | 3111 | 3058 | NA | NA | 0·98‡ | Increase in new smear-positive tuberculosis CNR during 2015–16 (p=0·003)§ |
CNR=case notification rate. ACF=active case-finding. NA=not applicable.
*
The population denominator estimate and numbers of tuberculosis cases are not stated.
†
Mean of tuberculosis CNR for two quarters in which intervention was ongoing.
‡
No population denominator stated; CNR was calculated assuming the underlying population remained the same.
§
In the study, it is not clear how this p value was calculated or whether it is adjusted for clustering.