Table 3.
Risk of bias assessment for studies reporting on incidence and prevalence of active TB disease and latent TB infection in health care workers
| Study ID | Risk of selection biasa | Risk of detection/information bias for each outcomeb | Risk of confoundingc |
|---|---|---|---|
| Prospective cohort study | |||
| Adams: Prevalence and determinants of TB infection in health care workers, unpublished | UNCLEAR: Participation in the study was voluntary. 505/764 HCWs participated in the study. | Incidence and prevalence of LTBI and TB disease: LOW (TB diagnosed using standard procedures) Risk factors for TB: HIGH (HCWs completed questionnaires which are prone to information bias) |
UNCLEAR: Community vs. occupational exposure to TB; HIV status of all of the HCWs not known |
| McCarthy et al. [22] (linked to Van Rie et al. [23]) |
UNCLEAR: Participation in the study was voluntary. 120/450 eligible HCWs and 79/296 eligible medical students participated in the study. | Incidence and prevalence of LTBI: LOW (latent TB infection diagnosed using standard procedures) Risk factors for TB: HIGH (HCWs completed questionnaires which are prone to information bias) |
UNCLEAR: Community vs. occupational exposure to TB. However, HIV status of all participants was assessed. |
| Cross sectional study | |||
| Ayuk et al. [20] | UNCLEAR: Not all HCWs completed questionnaire | Incidence and prevalence of TB disease: LOW Risk factors for TB: HIGH (HCWs completed questionnaires which are prone to information bias) |
UNCLEAR: Community vs. occupational exposure to TB; HIV status of all of the HCWs not known |
| Claassens et al. [15] | LOW: Although authors report that health care facilities were randomly selected there is no explanation of how the randomisation process was conducted. | Incidence of TB disease: HIGH (In each health care facility a questionnaire was completed by the facility manager to indicate the number of HCWs who were registered in that facility and who had been on TB treatment from January 2006 to December 2008. | UNCLEAR: Community vs. occupational exposure to TB; HIV status of all of the HCWs not known |
| Dwadwa et al: Health worker access to HIV/TB prevention, treatment and care services in Africa: situational analysis and mapping of routine and current best practices, unpublished | UNCLEAR: Six of the facilities were randomly selected although there is no explanation of how this was conducted. Four of the facilities were specifically selected based on current best practice as recommended by the Department of Health and AIDS and TB Directorates. | Number of TB cases: HIGH (Data obtained from questionnaires and interviews | UNCLEAR: Community vs. occupational exposure to TB |
| Kranzer et al. [6] | UNCLEAR: It is not clear how the community health workers (CHWs) were selected, if all of the CHW were selected to participate or if participation was voluntary | Prevalence of TB disease: LOW (standard TB diagnostic techniques used) | UNCLEAR: HIV status of all of the HCWs not known |
| Naidoo et al. [24] | UNCLEAR: Although authors state that a randomly selected sample of dentists was approached to participate in the study, it is not clear how this randomisation process was conducted. Only 78 of the 100 dentists participated | Prevalence of LTBI: LOW (LTBI diagnosed with Mantoux tests) | UNCLEAR: It is not clear where the dentists practiced or the demographics of their patients; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| Retrospective cohort study | |||
| Balt et al. [26] | LOW: Detailed review of health staff records at the four dedicated TB centres | Incidence of TB disease: LOW | UNCLEAR: Community vs. occupational exposure to TB |
| Malangu et al. [5] | LOW: A pre-designed data collection form was used to extract data from claims submitted to the Compensation Commissioner from January 2007 to December 2009 | Cases of TB disease: LOW (Data based on reported cases of TB. However, it is well known that there is underreporting of TB among HCWs with regards to occupational diseases. This may introduce detection bias and affect the external validity of the study) | UNCLEAR: possible underreporting of TB cases; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| Jarand et al. [25] | LOW: Retrospective case record review of all patients with XDR-TB in Eastern and Western Cape province from 1996 to 2008 | UNCLEAR: it is not known how study authors determined whether patients were health care workers | UNCLEAR: Community vs. occupational exposure to TB |
| Mehtar et al. [29] | LOW: Retrospective review of occupationally acquired TB case reports | Cases of TB disease: LOW (Data based on reported cases of OATB. However, it is well known that there is underreporting of TB among HCWs with regards to occupational diseases. This may introduce detection bias and affect the external validity of the study) | UNCLEAR: Underreporting of TB cases; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| Naidoo et al. [16] | LOW: Retrospective record review. All HCW with TB treated at 8 specified public sector hospitals were included if records confirmed HCW status | Incidence of TB disease: LOW (However, it is possible that HCWs seeking TB treatment may not have stated their occupation, resulting in underreporting of TB cases and information bias) | UNCLEAR: Underreporting of TB cases; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| O’Donnell et al. [17] | UNCLEAR: Although hospital database was used to identify MDR-TB and XDR-TB admissions the study relied on HCW self-reporting their occupation. | Incidence of TB disease: LOW (However, it is possible that HCWs seeking TB treatment may not have stated their occupation, resulting in underreporting of TB cases and information bias) | UNCLEAR: Underreporting of TB cases; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| Tudor et al. [18] | LOW: Data abstracted from occupational health employee medical charts using a standardized chart audit form | Incidence of TB disease: LOW | UNCLEAR: Underreporting of TB cases; Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
| Van Rie et al. [23] | UNCLEAR: Participation in the study was voluntary. 120/450 eligible HCWs and 79/296 eligible medical students participated in the study. | Prevalence of LTBI: LOW LTBI was diagnosed using TST and IGRAs |
UNCLEAR: Community vs occupational exposure to TB. HIV status of all participants was assessed. |
| Wilkinson et al. [21] | LOW: Staff TB data was extracted confidentially from the anonymized tuberculosis control programme register. | Incidence of TB disease: LOW (Case ascertainment is known to be complete because tuberculosis treatment cannot be obtained anywhere else in the district and all staff illness episodes are recorded in personnel files) | UNCLEAR: Community vs occupational exposure to TB; HIV status of all of the HCWs not known |
LOW low risk of bias, HIGH high risk of bias UNCLEAR unclear risk of bias
aSelection bias refers to systematic differences between baseline characteristics of the groups that are compared or characteristics of those who participate in the study and those who don’t. It is important that the group described is representative of the population of interest
bDetection bias/information bias refers to systematic differences between groups in how outcomes are determined. Participant’s self-reported outcomes are usually associated with a high risk of detection or information bias
cConfounding factors can cause or prevent the outcome of interest, are not intermediate variables, and are not associated with the factor(s) under investigation. Confounding factors result in situations in which the effects of two processes are not separated, or the contribution of causal factors cannot be separated, or the measure of the effect of exposure or risk is distorted because of its association with other factors influencing the outcome of the study