WHO consolidated guidelines on tuberculosis: module 1: prevention: infection prevention and control: web annex B: GRADE evidence summary tables

Web Annex B. GRADE evidence summary tables WHO consolidated guidelines on tuberculosis Module 1: Prevention Infection prevention and control WHO consolidated guidelines on tuberculosis: Module 1: Prevention - infection prevention and control: Web Annex B. GRADE evidence summary tables ISBN 978-92-4-005591-9 (electronic version) © World Health Organization 2022 Some rights reserved. This work is available under the Creative Commons Attribution-NonCommercial-ShareAlike 3.0 IGO licence (CC BY-NC-SA 3.0 IGO; https://creativecommons.org/licenses/by-nc-sa/3.0/igo). Under the terms of this licence, you may copy, redistribute and adapt the work for non-commercial purposes, provided the work is appropriately cited, as indicated below. In any use of this work, there should be no suggestion that WHO endorses any specific organization, products or services. The use of the WHO logo is not permitted. If you adapt the work, then you must license your work under the same or equivalent Creative Commons licence. If you create a translation of this work, you should add the following disclaimer along with the suggested citation: “This translation was not created by the World Health Organization (WHO). WHO is not responsible for the content or accuracy of this translation. The original English edition shall be the binding and authentic edition”. Any mediation relating to disputes arising under the licence shall be conducted in accordance with the mediation rules of the World Intellectual Property Organization (http://www.wipo.int/amc/en/mediation/rules/). Suggested citation. Web Annex B. GRADE evidence summary tables. In: WHO consolidated guidelines on tuberculosis: Module 1: Prevention - infection prevention and control. Geneva: World Health Organization; 2022. Licence: CC BY-NC-SA 3.0 IGO. Cataloguing-in-Publication (CIP) data. CIP data are available at http://apps.who.int/iris. Sales, rights and licensing. To purchase WHO publications, see http://apps.who.int/bookorders. To submit requests for commercial use and queries on rights and licensing, see https://www.who.int/copyright. Third-party materials. If you wish to reuse material from this work that is attributed to a third party, such as tables, figures or images, it is your responsibility to determine whether permission is needed for that reuse and to obtain permission from the copyright holder. The risk of claims resulting from infringement of any third-party-owned component in the work rests solely with the user. General disclaimers. The designations employed and the presentation of the material in this publication do not imply the expression of any opinion whatsoever on the part of WHO concerning the legal status of any country, territory, city or area or of its authorities, or concerning the delimitation of its frontiers or boundaries. Dotted and dashed lines on maps represent approximate border lines for which there may not yet be full agreement. The mention of specific companies or of certain manufacturers’ products does not imply that they are endorsed or recommended by WHO in preference to others of a similar nature that are not mentioned. Errors and omissions excepted, the names of proprietary products are distinguished by initial capital letters. All reasonable precautions have been taken by WHO to verify the information contained in this publication. However, the published material is being distributed without warranty of any kind, either expressed or implied. The responsibility for the interpretation and use of the material lies with the reader. In no event shall WHO be liable for damages arising from its use. This publication forms part of the WHO consolidated guidelines on tuberculosis: Module 1: Prevention - infection prevention and control. It is being made publicly available for transparency purposes and information, in accordance with the WHO handbook for guideline development, 2nd edition (2014). Design by Inís Communication Web Annex B. GRADE evidence summary tables WHO consolidated guidelines on tuberculosis Module 1: Prevention Infection prevention and control WHO Guidelines on Tuberculosis Infection 3 Prevention and Control 2019 UPDATE ONLINE ANNEXES Annex 4 – GRADE evidence summary tables Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can triage of people with TB signs, symptoms or with confirmed TB disease, reduce TB transmission to health care workers (HCW) (including community HCWs) when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment № of patients Effect Certainty Importance № of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Triage No triage Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settingsa 6 1,2,3,4,5,6,b,c,d,e,f observational studiesg serious h not serious very seriousi seriousj none 1966/24852 (7.9%) 1350/9647 (14.0%) RR 0.57 (-- to --) 60 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in low TB burden settingsk 5 2,3,4,5,6,b,c,f,l observational studiesg serious h not serious very seriousi seriousm none 206/22035 (0.9%) 322/8045 (4.0%) RR 0.23 (-- to --) 31 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in high TB burden settingsn 1 1,d observational studies serious o not serious p seriousq not serious none 1760/2817 (62.5%) 1028/1602 (64.2%) RR 0.97 (-- to --) 19 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in primary care - not measured - - - - - - - - Reduction in LTBI incidence/prevalence in secondary/tertiary carer 6 1,2,3,4,5,6,b,c,d,e,f observational studiesg serious h not serious very seriousi seriousj none 1966/24852 (7.9%) 1350/9647 (14.0%) RR 0.57 (-- to --) 60 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in all settingss 2 7,8,t,u,v observational studies serious w not serious very seriousx seriousy none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in low TB burden settings 1 9 observational studies not serious not serious p not serious seriousz none RR 0.32 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in high TB burden settingsaa 2 7,8,u,v observational studies serious w not serious very seriousx seriousy none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in primary care - not measured - - - - - - - - Reduction in active TB incidence/prevalence in secondary/tertiary carebb 2 7,8,t,u,v observational studies serious w not serious very seriousx seriousy none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio Web Annex B. GRADE evidence summary tables Online annexes Annex 1. Methods and Expert panels Annex 2. GRADE Summary of Evidence Tables Annex 3. GRADE Evidence to Decision Tables 1WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 4 Prevention and Control 2019 UPDATE Explanations a. Please note: The total number of studies measuring the effect of triage on the incidence of LTBI in all settings was 10. Four studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; 3) Louther, 1997; and 4) Yanai, 2003. Please see separate footnotes that summarise the results of these studies. b. Study reporting outcome, but not included in summary assessments. Baussano, 2007: incidence rate of TST conversions of 106/4034 person-years before TBIC interventions were implemented, vs. 42 TST conversions per 4463 person-years after implementation (crude rate ratio 0.36 after vs. before). c. Study reporting outcome, but not included in summary assessments. Blumberg, 1998 (some overlap with 1995 paper): TST conversion rate of 5.98/100 person-years in 1992 (pre-intervention) to 1.09/100 person-years from 1993–1997 (after the intervention was implemented; crude incidence rate ratio 0.18, after vs. before [derived from data presented]; authors report a p-value comparing the two time periods: <0.001). d. Study reporting outcome, but not included in summary assessments. Yanai, 2003: TST conversions from 9.3 per 100 person-years (95% CI 3.3–15.3) before the implementation of TBIC measures (in 1995–1997) to 6.4 per 100 person-years (95% CI 1.5–11.4) and 2.2 per 100 person-years (95% CI 0–5.1), after implementation, in 1998 and 1999, respectively. Unadjusted rate ratio 0.9 (95% CI 0.4–2.2) for 1998 vs. 1995–1997 and 0.03 (95% CI 0.01–0.2) for 1999 vs. 1995–1997; adjusted rate ratio 0.4 (95% CI 0.1–1.6) and 0.01 (95% CI 0–0.04) for 1998 and 1999 vs. 1995–1997, respectively). e. Definitions of triage varied widely between the six studies: Bangsberg - "all patients known HIV+, with HIV risk factors, or homelessness presenting with pneumonia/evidence of TB were isolated on presentation at the emergency room"; Blumberg 1995 - "expanded respiratory isolation policy"; Holzman - not defined; Roth - "rapid diagnosis and treatment"; Welbel - "revised policy (based on CDC guidelines) for isolation [CDC 1994: "in hospitals and other inpatient facilities, any patient suspected of having or known to have infectious TB should be placed in a TB isolation room"]; and Wenger - "higher index of suspicion for TB and stricter application of isolation criteria" f. Study reporting outcome, but not included in summary assessments. Louther, 1997: 7.2 TST conversions per 100 person-years before the implementation of infection control measures, compared with 3.3 per 100 person-years after the implementation (crude rate ratio 0.46 [derived from data presented]; authors report p-value comparing the two groups: 0.001). g. A mix of before/after, during/after, and prospective and retrospective cohort studies. h. All studies are observational. Several studies have high risk of bias, with loss to follow-up, or incomplete ascertainment and/or reporting of outcomes of interest i. Indirectness exists in the wide variation in types of triage and the descriptions of their implementation, as well as the implementation of a large number of infection control measures at one time. Please see assessment of directness for details. j. Low number of events (<300) in almost all studies and two studies (Bangsberg and Wenger) have fewer than 20 events. The exception is the study by Roth et al., which has a total 2,878 events. k. Please note: The total number of studies estimating the effect of triage on the incidence of LTBI in low TB burden settings was eight. Three studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; and 3) Louther, 1997. Please see separate footnotes that summarise the results of these studies. l. Definitions of triage varied widely between the five studies: Bangsberg - "all patients known HIV+, with HIV risk factors, or homelessness presenting with pneumonia/evidence of TB were isolated on presentation at the emergency room"; Blumberg 1995 - "expanded respiratory isolation policy"; Holzman - not defined; Welbel - "revised policy (based on CDC guidelines) for isolation [CDC 1994: "in hospitals and other inpatient facilities, any patient suspected of having or known to have infectious TB should be placed in a TB isolation room"]; and Wenger - "higher index of suspicion for TB and stricter application of isolation criteria" m. All studies have small numbers of events (<300; two had <20 events) and moderate overall sample sizes (except for Blumberg et al.) n. Please note: The total number of studies estimating the effect of triage on the incidence of LTBI in high TB burden settings was two. One study was excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because it did not report results in a format suitable for aggregation. This was (first author, year published): 1) Yanai, 2003. Please see the separate footnote that summarises the results of this study. o. High loss to follow-up. p. Cannot comment on inconsistency as data from only one study included. q. Very different definitions of triage used, population not well described, differences in background risk, and triage implemented along with other infection control measures. Please see assessment of directness for details. r. Please note: The total number of studies measuring the effect of triage on the incidence of LTBI in secondary/tertiary care settings was 10. Four studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; 3) Louther, 1997; and 4) Yanai, 2003. Please see separate footnotes that summarise the results of these studies. s. Please note: The total number of studies measuring the effect of triage on the incidence of TB disease in all settings was four. Two studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Jacobson, 1957; and 2) O'Hara, 2017. Please see separate footnotes that summarise the results of these studies. t. Study reporting outcome, but not included in summary assessments. Jacobson, 1957: incidence rate of 78 episodes of TB disease among healthcare workers in 38,331 person-years in the control group (1942–51, before the intervention was implemented) to 12 episodes in 18,229 person-years after the implementation of triage (1952–55; crude incidence rate ratio 0.32, after vs. before). u. Definitions of triage differed between the two studies: Harries - "priority to patients with chronic cough; rapid collection of sputum specimens" and Yanai - "triage/isolation and expedited diagnosis training for health care workers" v. Study reporting outcome, but not included in summary assessments. O'Hara, 2017: Unadjusted odds ratio (OR) for TB disease in HCW at facilities with a higher administrative score was 0.94 (95% CI 0.87–1.02; p = 0.12). Adjusted OR (adjusted for environmental score, PPE score, miscellaneous score, and number of TB patients) 0.97 (95% CI 0.90–1.04; p = 0.36). w. Under-ascertainment of outcomes in at least one study; poor reporting of loss to follow-up. x. Very serious indirectness exists in terms of the population studied and the nature and implementation of the intervention. Please see assessment of directness for details. y. Small numbers of events in both studies. z. Small number of outcomes in before (n = 78) and after (n = 12) periods. aa. Please note: The total number of studies measuring the effect of triage on the incidence of TB disease in high TB burden settings was three. One study was excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because it did not report results in a format suitable for aggregation. This was (first author, year published): 1) O'Hara, 2017. Please see the separate footnote that summarises the results of this study. bb. Please note: The total number of studies measuring the effect of triage on the incidence of TB disease in secondary/tertiary care settings was four. Two studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Jacobson, 1957; and 2) O'Hara, 2017. Please see separate footnotes that summarise the results of these studies. References 1. Roth VR, Garrett DO,Laserson KF,Starling CE,Kritski AL,Medeiros EAS,Binkin N,Jarvis WR. A multicenter evaluation of tuberculin skin test positivity and conversion among health care workers in Brazilian hospitals.. Int J Tuberc Lung Dis; 2005. 2. Wenger PN, Otten J,Breeden A,Orfas D,Beck-Sague CM,Jarvis WR. Control of nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis among healthcare workers and HIV-infected patients. Lancet; 1995. 3. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 4. Blumberg HM, Watkins DL,Berschling JD,Antle A,Moore P,White N,Hunter M,Green B,Ray SM,McGowan Jr. J E. Preventing the nosocomial transmission of tuberculosis. Ann Intern Med; 1995. 5. Bangsberg DR, Crowley K,Moss A,Dobkin JF,McGregor C,Neu HC. Reduction in tuberculin skin-test conversions among medical house staff associated with improved tuberculosis infection control practices. Infect Control Hosp Epidemiol; 1997. 6. Holzman, RS. A comprehensive control program reduces transmission of tuberculosis to hospital staff. Clin Infect Dis; 1995. 7. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 8. Harries AD, Hargreaves NJ,Gausi F,Kwanjana JH,Salaniponi FM. Preventing tuberculosis among health workers in Malawi. Bull WHO; 2002. 9. Jacobson G, Hoyt DD,Bogen E. Tuberculosis in hospital employees as affected by an admission chest X-ray screening program. Dis Chest; 1957. 2WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 5 Prevention and Control 2019 UPDATE Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can triage of people with TB signs, symptoms or with confirmed TB disease, reduce TB transmission to other persons attending healthcare settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Triage No triage Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settings (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in all settings (n = 2 studies) 2 1,2,a observational studies serious b not serious very serious c serious d none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in low TB burden settings (n = 2 studies) 2 1,2,a observational studies serious b not serious very serious c serious d none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in high TB burden settings (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in primary care (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in secondary/tertiary care (n = 2 studies) 2 1,2,a observational studies serious b not serious very serious c serious d none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in HIV-negative individuals (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in HIV-positive individuals (n = 2 studies) 2 1,2,a observational studies serious b not serious very serious c serious d none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. Please note that meta-analysis was *not* conducted - all summary estimates and measures of effect are crude estimates. b. Serious risk of bias, probable to alter the results: exposure is different for each study between before and after groups; and not a clear differentiation of intervention vs. no intervention. c. Multiple interventions were introduced at the same time. In addition, 'triage' was poorly defined in both studies, as targeting people with "respiratory disease and fever'" but with no mention of expedited diagnosis, or as an "increased index of suspicion for TB" without description of how this was implemented. Please see also assessment of directness. d. Both studies had small sample sizes. The total at-risk population was 543; a total 50 events were included. References 1. Stroud LA, Tokars JI Grieco MH Crawford JT Culver DH Edlin BR Sordillo EM Woodley CL Gilligan ME Schnieder N Williams J Jarvis WR. Evaluation of infection control measures in preventing the nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis in a New York city hospital. Infect Control Hosp Epidemiol; 1995. 2. Moro ML, Errante I Infuso A Sodano L Gori A Orcese CA Salamina G D'Amico C Besozii G Caggese L. Effectiveness of infection 3WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 6 Prevention and Control 2019 UPDATE Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can respiratory isolation/separation of people with presumed or demonstrated infectious TB reduce TB transmission to HCWs (including community HCWs) when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment № of patients Effect Certainty Importance № of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Respiratory isolation No respiratory isolation Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settingsa 12 1,2,3,4,5,6,7,8,9,10,11,12,b,c,d,e,f,g,h observational studies very seriousi not serious very serious j seriousk none 2413/91397 (2.6%) 1914/40097 (4.8%) RR 0.55 (-- to --) 21 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in low TB burden settingsl 11 1,2,4,5,6,7,8,9,10,11,12,b,c,d,f,h observational studies very seriousm not serious very serious j seriousk none 653/88580 (0.7%) 886/38495 (2.3%) RR 0.32 (-- to --) 16 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in high TB burden settingsn 1 3,e,g observational studies seriouso not seriousp seriousj not serious none 1760/2817 (62.5%) 1028/1602 (64.2%) RR 0.97 (-- to --) 19 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in primary care - not measured - - - - - - - - Reduction in LTBI incidence/prevalence in secondary/tertiary careq 12 1,2,3,4,5,6,7,8,9,10,11,12,b,c,d,e,f,g,h observational studies very seriousi not serious very serious j seriousk none 2413/91397 (2.6%) 1914/40097 (4.8%) RR 0.55 (-- to --) 21 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in all settingsr 2 13,14,s,t observational studies serious u not serious very seriousv seriousw none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reductions in active TB incidence/prevalence in low TB burden settings - not measured - - - - - - - - Reductions in active TB incidence/prevalence in high TB burden settingsx 2 13,14,s,t observational studies seriousu not serious very seriousv seriousw none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reductions in active TB incidence/prevalence in primary care 1 15,y observational studies very seriousz not serious p very seriousaa serious bb none OR 1.09 (0.99 to 1.19) 1 fewer per 1,000 (from 1 fewer to 1 fewer) ⨁◯◯◯ VERY LOW CRITICAL Reductions in active TB incidence/prevalence in secondary/tertiary carecc 2 13,14,t observational studies seriousu not serious very seriousv seriousw none 110/6216 (1.8%) 129/7161 (1.8%) RR 0.98 (-- to --) 0 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio; OR: Odds ratio 4WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 7 Prevention and Control 2019 UPDATE Explanations a. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of LTBI in all settings was 19. Seven studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; 3) Bryan, 1983; 4) da Costa, 2009; 5) Louther, 1997; 6) Sinkowitz, 1996; and 7) Yanai, 2003. Please see separate footnotes that summarise the results of these studies. b. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Baussano, 2007: incidence rate of TST conversions of 106/4034 person-years before TBIC interventions were implemented, vs. 42 TST conversions per 4463 person-years after implementation (crude rate ratio 0.36 after vs. before). c. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Blumberg, 1998; some overlap with 1995 paper): TST conversion rate of 5.98/100 person-years in 1992 (pre-intervention) to 1.09/100 person-years from 1993-1997 (after the intervention was implemented; crude incidence rate ratio 0.18, after vs. before [derived from data presented]; authors report a p-value comparing the two time periods: <0.001). d. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Bryan, 1983: TST conversion of 4.5% of HCWs in 1976, before the implementation of TBIC measures, vs. 5.1%, 1.5%, 0.85%, and 0.59% in the four years after implementation (crude risk ratio 1.13, 0.33, 0.19, and 0.13 for 1977–1981, respectively). e. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. da Costa, 2009: TST conversions incidence rate from 5.8 per 1,000 person-months (95% CI 4.9–6.7), to 3.7 per 1,000 person-months (95% CI 2.8–4.6); rate ratio 0.46 (95% CI 0.23– 0.89) after vs. before, p = 0.006; adjusted rate ratio (adjusted for exposure and occupation) 0.24 (95% CI 0.10–0.54). f. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Sinkowitz, 1996: TST conversion in 0%, 8.0%, and 5.1% of bronchoscopists in hospitals without IC measures and zero TB patients, 1–5 TB patients, and ≥6 TB patients, vs. 3.3%, 8.3%, and 5.7% in hospitals with the same numbers of TB patients but which had implemented four IC measures (crude risk ratio 1.04 and 1.12 [IC vs. no IC] for hospitals with 1–5 TB patients and ≥6 TB patients, respectively). In other HCW, TST conversion in 0.49%, 0.64%, and 0.76% in hospitals without IC measures and zero TB patients, 1–5 TB patients, and 6 TB patients, vs. 0.53%, 0.69% and 0.90% in hospitals with the same numbers of TB patients but which had implemented four IC measures (crude risk ratio 1.08, 1.08, and 1.18 [IC vs. no IC] for hospitals with zero, 1–5 and ≥6 TB patients, respectively). g. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Yanai, 2003: TST conversions from 9.3 per 100 person-years (95% CI 3.3–15.3) before the implementation of TBIC measures (in 1995–1997) to 6.4 per 100 person-years (95% CI 1.5– 11.4) and 2.2 per 100 person-years (95% CI 0–5.1), after implementation, in 1998 and 1999, respectively. Unadjusted rate ratio 0.9 (95% CI 0.4–2.2) for 1998 vs. 1995–1997 and 0.03 (95% CI 0.01–0.2) for 1999 vs. 1995–1997; adjusted rate ratio 0.4 (95% CI 0.1–1.6) and 0.01 (95% CI 0–0.04) for 1998 and 1999 vs. 1995–1997, respectively). h. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Louther, 1997: 7.2 TST conversions per 100 person-years before the implementation of infection control measures, compared with 3.3 per 100 person-years after the implementation (crude rate ratio 0.46 [derived from data presented]; authors report p-value comparing the two groups: 0.001). i. Most studies included here have a high or unclear risk of bias. All are observational studies, some with high rates of loss to follow-up (e.g., Roth), low or unclear levels of participation, or incomplete reporting of outcomes (e.g., Blumberg). Two studies do not report results correctly or have missing results. j. Indirectness was primarily through the implementation of multiple infection control measures together with isolation. Please see assessment of directness for details. k. Imprecision exists: all except two studies (Fridkin and Roth) have fewer than 300 outcomes and three studies (Bangsberg, Behrman, and Wenger) have fewer than 20 outcomes. l. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of LTBI in low TB burden settings was 16. Five studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; 3) Bryan, 1983; 4) Louther, 1997; and 5) Sinkowitz, 1996. Please see separate footnotes that summarise the results of these studies. m. Most studies included here have a high or unclear risk of bias. All are observational studies, some have incomplete reporting of outcomes (e.g., Blumberg), and two studies do not report results correctly or have missing results. n. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of LTBI in high TB burden settings was three. Two studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) da Costa, 2009 and 2) Yanai, 2003. Please see separate footnotes that summarise the results of these studies. o. High proportions were lost to follow-up; those lost to follow-up may have been at higher risk of disease (more likely to be physicians). p. Cannot comment on inconsistency as data from only one study are included. q. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of LTBI in secondary/tertiary care settings was 19. Seven studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Baussano, 2007; 2) Blumberg, 1998; 3) Bryan, 1983; 4) da Costa, 2009; 5) Louther, 1997; 6) Sinkowitz, 1996; and 7) Yanai, 2003. Please see separate footnotes that summarise the results of these studies. r. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of active TB disease in all settings was four. Two studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Claassens, 2013 and 2) O'Hara, 2017. Please see separate footnotes that summarise the results of these studies. s. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. Claassens, 2013: Unadjusted odds ratio for smear-positive TB among health care workers in facilities where administrative controls were implemented vs.facilities without (or with fewer) administrative controls 1.09 (95% CI 0.99–1.19), p = 0.07. t. STUDY REPORTING OUTCOME BUT NOT INCLUDED IN SUMMARY ASSESSMENTS. O'Hara, 2017: Unadjusted odds ratio (OR) for TB disease in HCW at facilities with a higher administrative score was 0.94 (95% CI 0.87–1.02; p = 0.12). Adjusted OR (adjusted for environmental score, PPE score, miscellaneous score, and number of TB patients) 0.97 (95% CI 0.90–1.04; p = 0.36). u. Under-ascertainment of outcome in at least one study. All studies implemented isolation/spatial separation in addition to a number of other TBIC interventions; the effect of isolation/separation on the outcome of interest cannot be determined. Poor reporting of loss to follow- up. v. Very serious indirectness exists, for populations studied and in the nature of and fidelity to the intervention. Please see assessment of directness for details. w. Both studies had fewer than 200 events; one had fewer than 100 events. x. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of active TB disease in high TB burden settings was four. Two studies were excluded from the summary analysis (certainty estimates and crude summaries of findings [meta- analysis was NOT conducted]) because they did not report results in a format suitable for aggregation. These were (first author, year published): 1) Claassens, 2013 and 2) O'Hara, 2017. Please see separate footnotes that summarise the results of these studies. y. Please note that the odds ratio quoted for this study is for the development of smear-positive TB among healthcare workers at facilities classified by their implementation of infection control measures (i.e., the authors reported slightly increased odds of developing smear- positive TB in healthcare workers in facilities where administrative controls were implemented compared with facilities without or with fewer administrative controls). z. High likelihood of under-ascertainment of outcome (smear-positive disease in HCW), as only routine records used, without verification or any additional efforts to estimate numbers of cases. In addition, high variability in implementation intervention across different facilities, with isolation only implemented in ~50% of facilities. Most importantly, the study used the facilities as the base unit for assessing risk of TB disease (so reduced TB incidence to a binary of 'any' vs. 'no' HCW developing TB at a particular facility) - individual HCW data not analysed. aa. aa. Indirectness is severe. Please see assessment of directness for details. bb. ab. Small effect seen, and in the opposite direction to expected. Confidence interval is narrow, but crosses 1. cc. ac. PLEASE NOTE: The total number of studies measuring the effect of isolation on the incidence of active TB disease in secondary/tertiary care settings was three. One study was excluded from the summary analysis (certainty estimates and crude summaries of findings [meta-analysis was NOT conducted]) because it did not report results in a format suitable for aggregation. This was (first author, year published): 1) O'Hara, 2017. Please see the separate footnote that summarises the results of this study. References 1. Jones, SG. Evaluation of a human immunodeficiency virus rule out tuberculosis critical pathway as an intervention to decrease nosocomial transmission of tuberculosis in the inpatient setting. AIDS Patient Care Stds; 2002. 2. Jarvis, WR. Nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis. Am J Infect Control; 1995. 3. Roth VR, Garrett DO,Laserson KF,Starling CE,Kritski AL,Medeiros EAS,Binkin N,Jarvis WR. A multicenter evaluation of tuberculin skin test positivity and conversion among health care workers in Brazilian hospitals.. Int J Tuberc Lung Dis; 2005. 4. Wenger PN, Otten J,Breeden A,Orfas D,Beck-Sague CM,Jarvis WR. Control of nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis among healthcare workers and HIV-infected patients. Lancet; 1995. 5. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 5WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 8 Prevention and Control 2019 UPDATE 6. Uyamadu N, Ahkee S,Carrico R,Tolentino A,Wojda B,Ramirez J. Reduction in tuberculin skin-test conversion rate after improved adherence to tuberculosis isolation. Infect Control Hosp Epidemiol; 1997. 7. Maloney SA, Pearson ML,Gordon MT,Del Castillo R,Boyle JF,Jarvis WR. Efficacy of control measures in preventing nosocomial transmission of multidrug-resistant tuberculosis to patients and health care workers. Ann Intern Med; 1995. 8. Fridkin SK, Manangan L,Bolyard E,Jarvis WR. SHEA-CDC TB survey, Part II: Efficacy of TB infection control programs at member hospitals, 1992. Society for Healthcare Epidemiology of America. Infect Control Hosp Epidemiol; 1995. 9. Blumberg HM, Watkins DL,Berschling JD,Antle A,Moore P,White N,Hunter M,Green B,Ray SM,McGowan Jr. J E. Preventing the nosocomial transmission of tuberculosis. Ann Intern Med; 1995. 10. Behrman AJ, Shofer FS. Tuberculosis exposure and control in an urban emergency department. Ann Emerg Med; 1998. 11. Bangsberg DR, Crowley K,Moss A,Dobkin JF,McGregor C,Neu HC. Reduction in tuberculin skin-test conversions among medical house staff associated with improved tuberculosis infection control practices. Infect Control Hosp Epidemiol; 1997. 12. Holzman, RS. A comprehensive control program reduces transmission of tuberculosis to hospital staff. Clin Infect Dis; 1995. 13. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 14. Harries AD, Hargreaves NJ,Gausi F,Kwanjana JH,Salaniponi FM. Preventing tuberculosis among health workers in Malawi. Bull WHO; 2002. 15. Claassens M, van Schalkwyk C,du Toit E,Roest E,Lombard CJ,Enarson DA,Beyers N,Borgdorff MW. Tuberculosis in Healthcare Workers and Infection Control Measures at Primary Healthcare Facilities in South Africa. PLoS One; 2013. 6WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 9 Prevention and Control 2019 UPDATE Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can respiratory isolation / separation of people with presumed or demonstrated infectious TB reduce TB transmission to other persons attending healthcare settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Respiratory isolation No respiratory isolation Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settings (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in all settings (n = 2 studies; n = 543 individuals at risk) 2 1,2,a observational studies serious b not serious very seriousc seriousd none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in low TB burden settings (n = 2 studies; n = 543 individuals at risk) 2 1,2,a observational studies serious b not serious very seriousc seriousd none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in high TB burden settings (n = 0 studies; n = 0 individuals at risk) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in primary care (n = 0 studies; n = 0 individuals at risk) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in secondary/tertiary care (n = 2 studies; n = 543 individuals at risk) 2 1,2,a observational studies serious b not serious very seriousc seriousd none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in HIV-negative individuals (n = 0 studies; n = 0 individuals at risk) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in HIV-positive individuals (n = 2 studies; n = 543 individuals at risk) 2 1,2,a observational studies serious b not serious very seriousc seriousd none 5/237 (2.1%) 45/306 (14.7%) RR 0.143 (-- to --) 126 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. Please note that meta-analysis was *not* conducted - all summary estimates and measures of effect are crude estimates. b. Serious risk of bias, probable to alter the results: exposure is different for each study between before and after groups; also isolation measures were in effect before and then more so after. Not a clear differentiation of intervention vs. no intervention. c. Multiple interventions were introduced at the same time. d. Both studies had small sample sizes. The total at-risk population was 543; a total 50 events were included. References 1. Moro ML, Errante I Infuso A Sodano L Gori A Orcese CA Salamina G D'Amico C Besozii G Caggese L. Effectiveness of infection control measures in controlling a nosocomial outbreak of multidrug-resistant tuberculosis among HIV patients in Italy.. Int J Tuberc Lung Dis; 2000. 2. Stroud LA, Tokars JI Grieco MH Crawford JT Culver DH Edlin BR Sordillo EM Woodley CL Gilligan ME Schnieder N Williams J Jarvis WR. Evaluation of infection control measures in preventing the nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis in a New York city hospital. Infect Control Hosp Epidemiol; 1995. 7WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 10 Prevention and Control 2019 UPDATE Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can effective treatment of patients with TB disease reduce TB transmission to HCWs (including community HCWs) when compared to transmission to the same populations in settings where treatment is not yet administered? Setting: International Certainty assessment № of patients Effect Certainty Importance № of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Effective treatment Treatment – [delayed or] not DST-based Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settings 4 1,2,3,4,a,b observational studies very serious c serious d very serious e very serious f none 42/3081 (1.4%) 155/3260 (4.8%) RR 0.29 (-- to --) 34 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in low TB burden settings 4 1,2,3,4,a,b observational studies very serious c serious d very serious e very serious f none 42/3081 (1.4%) 155/3260 (4.8%) RR 0.29 (-- to --) 34 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in LTBI incidence/prevalence in high TB burden settings - not measured - - - - - - - - Reduction in LTBI incidence/prevalence in primary care - not measured - - - - - - - - Reduction in LTBI incidence/prevalence in secondary/tertiary care 4 1,2,3,4,a,b observational studies very serious c serious d very serious e very serious f none 42/3081 (1.4%) 155/3260 (4.8%) RR 0.29 (-- to --) 34 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in all settings - not measured - - - - - - - - CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. Please note that the study included by Welbel et al. does not describe, specifically, the implementation of treatment based on drug susceptibility, but only describes the introduction of drug susceptibility testing. We have assumed that the results of testing were then used to inform treatment. b. Please note that meta-analysis was *not* conducted - pooled estimates and measures of effect are crude estimates. c. There are design specific issues to these studies. Mainly, it is not possible to ascertain the effect of the intervention in question as the intervention is grouped with other interventions, which presents a serious risk of bias. There is also a serious design issue with the study by Wenger et al., as the intervention only differs slightly between before and after (3 agents vs. 4 agents). Though studies were not designed specifically to answer our question, the way they are designed does not give us confidence in the results of interest. d. Some inconsistency exists. In the study by Jarvis, in particular, certain results are reported as unavailable, but the site of origin of these results is not specified, so this cannot be accounted for in analysis. In addition, in the study by Welbel et al., overall denominators for at-risk individuals are provided, but not the time period for which these individuals were at risk, reducing confidence in the estimates of risk. e. Indirectness is severe and from many sources: population, intervention, and comparators (please see assessment of directness for details). f. Serious imprecision exists. For a dichotomous outcome all studies have fewer than 110 cases (range 10–104). Samples sizes are also low in three studies (range 65–650; the exception is Welbel et al, with a sample size of 4,329). References 1. Jarvis, WR. Nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis. Am J Infect Control; 1995. 2. Wenger PN, Otten J,Breeden A,Orfas D,Beck-Sague CM,Jarvis WR. Control of nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis among healthcare workers and HIV-infected patients. Lancet; 1995. 3. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 4. Maloney SA, Pearson ML,Gordon MT,Del Castillo R,Boyle JF,Jarvis WR. Efficacy of control measures in preventing nosocomial transmission of multidrug-resistant tuberculosis to patients and health care workers. Ann Intern Med; 1995. 8WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 11 Prevention and Control 2019 UPDATE Author(s): TB Centre, London School of Hygiene & Tropical Medicine Date: 27-29 March 2018 Question: Can effective treatment of patients with TB disease reduce TB transmission to other persons attending healthcare settings when compared to transmission to the same populations in settings where treatment is not yet administered? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Effective treatment Treatment – [delayed or] not DST-based Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence in all settings (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in all settings (n = 1 study) 1 1,a observational studies serious b not seriousc very seriousd seriouse none 5/193 (2.6%) 19/216 (8.8%) RR 0.295 (-- to --) 62 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in low TB burden settings (n = 1 study) 1 1,a observational studies serious b not seriousc very seriousd seriouse none 5/193 (2.6%) 19/216 (8.8%) RR 0.295 (-- to --) 62 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in high TB burden settings (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in primary care (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in secondary/tertiary care (n = 1 study) 1 1,a observational studies serious b not seriousc very seriousd seriouse none 5/193 (2.6%) 19/216 (8.8%) RR 0.295 (-- to --) 62 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL Reduction in active TB incidence/prevalence in HIV-negative individuals (n = 0 studies) - not measured - - - - - - - - Reduction in active TB incidence/prevalence in HIV-positive individuals (n = 1 study) 1 1,a observational studies serious b not seriousc very seriousd seriouse none 5/193 (2.6%) 19/216 (8.8%) RR 0.295 (-- to --) 62 fewer per 1,000 (from -- to --) ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. Please note that meta-analysis was *not* conducted - all summary estimates and measures of effect are crude estimates. b. No significant difference in the treatment in the before and after groups (1.5 vs. 2.0 drugs given before vs. after; range 0-4 in both periods; p = 0.2). Exposure is also different for between before and after groups. c. As there is only one study included we cannot comment on heterogeneity of results between studies. d. Authors describe "expanded use of antituberculous drugs" in 'after' period, but no description of time to treatment; therefore unable to assess for difference compared with delayed treatment administration. e. Small numbers of cases in both arms. Overall number of exposed individuals = 409 (n = 216 before; n = 193 after) References 1. Stroud LA, Tokars JI Grieco MH Crawford JT Culver DH Edlin BR Sordillo EM Woodley CL Gilligan ME Schnieder N Williams J Jarvis WR. Evaluation of infection control measures in preventing the nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis in a New York city hospital. Infect Control Hosp Epidemiol; 1995. 9WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 12 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can respiratory hygiene (or cough etiquette) in people with presumed or confirmed TB reduce TB transmission to healthcare workers in healthcare or other congregate settings to reduce TB transmission when compared to settings where these interventions are not implemented? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence - all settings (n=2) 2 1,2 observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Two studies were included. Heterogeneity in the interventions precluded meta-analysis. The two studies both found a reduction in TST conversions in the intervention compared to control group. In Roth (n=7735), a composite intervention including surgical mask use by patients (comparing two hospitals in the intervention arm to two in the control arm) reduced TST conversions by between 4.1 and 12.4 conversions per 1,000 person months. In Yanai 2003, a composite intervention including patient masks was associated with a decrease in TST conversions from 13/77 (16.9%) to 2/96 (2.1%) – a decrease of 14.8%. 1,2,c ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n=2) 2 2,3 observational studies serious a not serious seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Two studies were included. Heterogeneity in the interventions precluded meta-analysis. In these two studies, surgical mask use by patients was a part of a composite intervention. They both found a reduction in TB in the intervention compared to control group. In Harries 2002, the use of surgical masks by patients as a part of a composite intervention of 13 components reduced the TB notification rate from 100/2697 (3.7%) to 96/2979 (3.2%). In Yanai 2003, a composite intervention including patient masks was associated with a decrease in TB cases from 30/4357 (0.7%) to 19/4780 (0.4%), a reduction in 0.29 cases/100 person years. Therefore, both studies were associated with a decrease in TB cases. 2,3,c ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. The one included study had a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and interventions are poorly described. The intervention is a composite intervention including engineering, respiratory protection and administrative controls, of which cough hygiene is one component (downgraded by one level). c. No single effect estimate/meta-analysis was possible due to heterogeneity of outcomes. References 1. Roth VR, Garrett DO,Laserson KF,Starling CE,Kritski AL,Medeiros EAS,Binkin N,Jarvis WR. A multicenter evaluation of tuberculin skin test positivity and conversion among health care workers in Brazilian hospitals.. Int J Tuberc Lung Dis; 2005. 2. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 3. Harries AD, Hargreaves NJ,Gausi F,Kwanjana JH,Salaniponi FM. Preventing tuberculosis among health workers in Malawi. Bull WHO; 2002. 10WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 13 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can respiratory hygiene (or cough etiquette) in people with presumed or confirmed TB reduce TB transmission to other persons attending healthcare settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Respiratory hygiene No respiratory hygiene Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence (n=1) (Animal study, surgical mask use by patient with TB) 1 1,a observational studies not serious b not serious seriousc not serious strong association 36/90 (40.0%) 69/90 (76.7%) not pooled see comment ⨁⨁◯◯ LOW CRITICAL Reduction in TB incidence/prevalence (n=1) 1 2,d observational studies serious e not serious seriousf not serious strong association all plausible residual confounding would suggest spurious effect, while no effect was observed 0/44 (0.0%) 26/90 (28.9%) not pooled see comment ⨁⨁◯◯ LOW CRITICAL Reduction in TB incidence/prevalence in people living with HIV (n=1) 1 2,d observational studies serious e not serious seriousf not serious strong association all plausible residual confounding would suggest spurious effect, while no effect was observed 0/44 (0.0%) 26/90 (28.9%) not estimable ⨁⨁◯◯ LOW CRITICAL CI: Confidence interval; RR: Risk ratio Explanations a. Dharmadhikari 2012 measured the effect of surgical mask use by MDR-TB patients upon TST conversion in guinea pigs. The mask use was associated with a substantial reduction in infection 69/90 (76.6%) to 36/90 (40.0%), a reduction by 36.6% in guinea pigs. The reviewers assessed that indirectness was an important concern, given differences between humans and guinea pigs. This led to downgrading the quality of evidence by one point. A steady rise in infection risk over the study period, indicating a dose-response relationship with the duration of exposure. This led to upgrading the quality assessment by one. Therefore, this was rated as low quality evidence. b. The blinding of the individuals reporting the outcomes was not stated. c. The biology of latent TB infection in guinea pigs is different than that in humans. Therefore there is a serious concern of indirectness (Downgraded by one level). d. Moro 2000 (n= 134) study evaluated the effect of surgical mask use for prevention of transmission of MDR-TB, with the outcome of MDR-TB. In this study, surgical mask use by patients was a part of a composite intervention. There was a reduction of 29% in the incidence of TB between the intervention group (0/44 (0%)) and the control group (26/90 (29%)). e. The included study has a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). f. The comparator and interventions are poorly described. The interventions comprise multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). References 1. Dharmadhikari, . Surgical Face Masks Worn by Patients with Multidrug-Resistant Tuberculosis. Am J Respir Crit Care Med; 2012. 2. Moro ML, Errante I Infuso A Sodano L Gori A Orcese CA Salamina G D'Amico C Besozii G Caggese L. Effectiveness of infection control measures in controlling a nosocomial outbreak of multidrug-resistant tuberculosis among HIV patients in Italy.. Int J Tuberc Lung Dis; 2000. 11WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 14 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can upper room GUV reduce TB transmission in healthcare workers in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence (n=3) 3 1,2,3 observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Three studies in humans evaluated this outcome. In Fella, a composite outcome including UVGI was associated with a reduction in TST conversion from 41/303 (13.5%) in the intervention group to 21/446 (4.7%) in the control group – a reduction of 8.8%. In Yanai 2003, a composite intervention including patient masks was associated with a decrease in TST conversions from 13/77 (16.9%) to 2/96 (2.1%) – a decrease of 14.8%. Therefore, both studies demonstrated a reduction in TST conversions. Welbel 1995 showed that mechanical ventilation, in combination with other engineering measures, was associated with a reduction in TST conversions from 98/2,221 (4.4%) to 6/2108 (0.28%), a reduction of 4.1%. Heterogeneity in the interventions precluded meta-analysis. ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n= 1) Upper room UVGI No upper room UVGI Relative (95% CI) Absolute (95% CI) 1 2,c observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 19/4780 (0.4%) 30/4357 (0.7%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval Explanations a. The included studies have a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and interventions are poorly described. The interventions comprise multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). c. Only one study evaluated this outcome In Yanai 2003, a composite intervention including patient masks was associated with a decrease in TB cases from 30/4357 (0.7%) to 19/4780 (0.4%), a reduction in 0.29 cases/100 person years. References 1. Fella P, Rivera P, Hale M, Squires K, Sepkowitz K. Dramatic increase in tuberculin skin test conversion rate among employees at a hospital in New York City. Am J Infect Control; 1995. 2. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 3. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 12WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 15 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can upper room GUV reduce TB transmission in persons in TB care or others in high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence (n=0) in humans 0 - CRITICAL Reduction in TB incidence/prevalence (n=0) in humans 0 - CRITICAL Reduction in LTBI incidence/prevalence (animal studies) (n=2) 2 1,2 randomised trials not serious not serious 1,a seriousb not serious none Two animal studies were included, measuring infection in guinea pigs arising from exhausted air from patient wards. Both studies showed a reduction in infection with use of UVGI. The measured absolute reductions were 25.5% (Escombe), 46.7% (Mphaphlele). ⨁⨁⨁◯ MODERATE CRITICAL Reduction in TB incidence/prevalence (animal studies) (n=1) 1 2 randomised trials not serious not serious c seriousd not serious none One animal studies was included. This was conducted in guinea pigs, exposed to air from patients with TB. In this study, UVGI was associated with a reduction in TB on autopsy of 5%. ⨁⨁⨁◯ MODERATE CRITICAL CI: Confidence interval Explanations a. The direction and magnitude of the effect was consistent across the studies. One study (Mphaphlele) involved two study periods, where the rate of infectiousness differed based upon the location of the exhaust outlet in the room. The data were pooled in the final analysis. The direction of the effect was the same in both time periods. b. These three studies evaluated tuberculin skin test conversion among guinea pigs exposed to air removed from tuberculosis wards. Differences in the nature of transmission to guinea pigs, compared to humans, are likely to be significant (Downgraded one level). c. The direction and magnitude of the effect was consistent across the studies. d. These studies were conducted among guinea pigs (3 studies) and rabbits (1 study). Tuberculosis was diagnosed by autopy. Differences in the nature of transmission to animals and the measurement of the outcome (autopsy diagnosed disease) compared to humans are likely to be significant (Downgraded one level). References 1. Mphaphlele M, Dharmadhikari AS,Jensen PA,Rudnick SN,van Reenen TH,Pagano MA,Leuschner W,Sears TA,Milonova SP,van der Walt M,Stoltz AC,Weyer K,Nardell EA. Institutional Tuberculosis Transmission Controlled Trial of Upper Room Ultraviolet Air Disinfection: A Basis for New Dosing Guidelines. Am J Respir Crit Care Med; 2015. 2. Escombe AR, Moore DAJ,Gilman RH,Navicopa M,Ticona E,Mitchell B,Noakes C,Martinez C,Sheen P,Ramirez R,Quino W,Gonzalez A,Friedland JS,Evans CA. Upper-Room Ultraviolet Light and Negative Air Ionization to Prevent Tuberculosis Transmission. Plos Medicine; 2009. 13WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 16 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can mechanical ventilation reduce TB transmission in healthcare workers in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence (n= 7) 7 1,2,3,4,5,6,7 observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Seven studies evaluated the effect of mechanical ventilation upon TST conversion, each as a part of a composite intervention. Heterogeneity in the interventions precludes meta-analysis. Blumberg 1995 showed that the composite intervention, including 90 negative pressure rooms with fans, was associated with a reduction in TST conversions from 118/3579 (3.3%) to 23/5,153 (0.4%) – a reduction of 2.9%. Welbel 1995 showed that mechanical ventilation, in combination with other engineering measures, was associated with a reduction in TST conversions from 98/2,221 (4.4%) to 6/2108 (0.28%), a reduction of 4.1%. Wenger 1995 found that mechanical ventilation, including installation of 23 isolation rooms, was associated with a reduction in TST conversion from 7/25 (28%) to 3/17 (18%), a reduction of 10%. Maloney 1995 found that mechanical ventilation, in combination with other measures, was associated with a reduction in TST conversions from 15/90 (16.7%) to 4/78 (5.1%), a reduction by 11.5%. Roth 1995 showed that mechanical ventilation was associated with a similar TST conversion rate (7.4 / 1,000 person years without the measures, and 8.1 per 1,000 person years with the measures). Menzies 2002 was conducted among HCWs in microbiology and pathology laboratories. Ventilation was lower among those with TST conversion than among those without TST conversion (p<0.001). The adjusted odds ratio for those with half of the recommended ventilation versus the recommended ventilation was 1.3 (95% CI 0.9-1.9). Finally, in Fella 1995, a composite outcome including UVGI was associated with a reduction in TST conversion from 41/303 (13.5%) in the intervention group to 21/446 (4.7%) in the control group – a reduction of 8.8%. In summary, six of the seven studies showed a reduction in the incidence of TST over the study period. ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n=0 ) 0 not pooled see comment - CRITICAL Reduction in LTBI incidence/prevalence in TB laboratory workers (n=1) Use of ventilation systems (mechanical) No use of ventilation systems (mechanical) Relative (95% CI) Absolute (95% CI) 1 7,c observational studies serious a not serious seriousd not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 14 97 - see comment ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval 14WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 17 Prevention and Control 2019 UPDATE Explanations a. The included studies have a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and interventions are poorly described. The interventions are largely comprised of multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). c. This study conducted among HCWs in microbiology and pathology laboratories in 17 Canadian hospitals. The study measured mechanical ventilation within the laboratory facilities, and assesed the number of health workers with TST conversions during the study period. The study found that among 14 HCWs with TST conversions, the mean mechanical ventilation was 16.7 (SD 2.4) air changes per hour (ACH) . Among 97 staff without TST conversions, the mean mechanical ventilation was 32.5 (SD 22.7) ACH. Therefore, ventilation was lower among those with TST conversion than among those without TST conversion (p<0.001). The adjusted odds ratio for those with half of the recommended ventilation versus the recommended ventilation was 1.3 (95% CI 0.9-1.9). d. Differences in intervention (applicability). The comparator and intervention is poorly described. The intervention comprises multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). References 1. Fella P, Rivera P, Hale M, Squires K, Sepkowitz K. Dramatic increase in tuberculin skin test conversion rate among employees at a hospital in New York City. Am J Infect Control; 1995. 2. Wenger PN, Otten J,Breeden A,Orfas D,Beck-Sague CM,Jarvis WR. Control of nosocomial transmission of multidrug-resistant Mycobacterium tuberculosis among healthcare workers and HIV-infected patients. Lancet; 1995. 3. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 4. Maloney SA, Pearson ML,Gordon MT,Del Castillo R,Boyle JF,Jarvis WR. Efficacy of control measures in preventing nosocomial transmission of multidrug-resistant tuberculosis to patients and health care workers. Ann Intern Med; 1995. 5. Blumberg HM, Watkins DL,Berschling JD,Antle A,Moore P,White N,Hunter M,Green B,Ray SM,McGowan Jr. J E. Preventing the nosocomial transmission of tuberculosis. Ann Intern Med; 1995. 6. Roth VR, Garrett DO,Laserson KF,Starling CE,Kritski AL,Medeiros EAS,Binkin N,Jarvis WR. A multicenter evaluation of tuberculin skin test positivity and conversion among health care workers in Brazilian hospitals.. Int J Tuberc Lung Dis; 2005. 7. Menzies D, Fanning A,Yuan L,FitzGerald JM. Factors associated with tuberculin conversion in Canadian microbiology and pathology workers. Am J Respir Crit Care Med; 2003. 15WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 18 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can mechanical ventilation reduce TB transmission in persons in TB care or others in high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Use of ventilation systems (mechanical) No use of ventilation systems (mechanical) Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence (n= 1) 1 1,a observational studies serious b not serious very seriousc not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 73/189 (38.6%) 75/297 (25.3%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n=0) 0 - CRITICAL CI: Confidence interval Explanations a. Muecke 2006 found rooms with mechanical ventilation were associated with an increase in TST conversions from 75/297 (25%) to 73/189 (39%). Risk difference was +14% with ventilation in rooms compared to no ventilation. Confounding factors are likely, with temporal factors likely playing an important role. b. Temporal factors may have explained difference, shown by the increased infectivity in the second semester. The opening of windows in ventilated and non-ventilated rooms was not reported. c. Transmission in rooms with mechanical ventilation was compared to transmission in rooms without mechanical ventilation. The duration of exposure varied between rooms, and seasonal variation means that other forms of ventilation (e.g. open windows) cannot be excluded. References 1. Muecke C, Isler M,Menzies D,Allard R,Tannenbaum TN,Brassard R. The use of environmental factors as adjuncts to traditional tuberculosis contact investigation. Int J Tuberc Lung Dis; 2006. 16WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 19 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can mixed mode ventilation reduce TB transmission in healthcare workers in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence (n= 2) 2 1,2 observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Two studies addressed this question. Heterogeneity in the interventions precludes meta-analysis. In Yanai 2003, a composite intervention including mixed mode ventilation was associated with a decrease in TST conversions from 13/77 (16.9%) to 2/96 (2.1%) – a decrease of 14.8%. Behrman 1998 evaluated mixed mode ventilation, and other interventions including respiratory protection. TST conversions decreased from 6/50 (12%) to 0/64 (0%) over the study period. Therefore, both studies showed a reduction in TST conversions. Heterogeneity in the interventions precluded meta- analysis. ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n= 1) Use of ventilation systems (mixed) No use of ventilation systems (mixed) Relative (95% CI) Absolute (95% CI) 1 1,c observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 19/4780 (0.4%) 30/4357 (0.7%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval Explanations a. The included study has a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and intervention is poorly described. The intervention comprises multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). c. The one included study, Yanai 2003, demonstrated that the composite intervention, including mixed mode ventilation, was associated with a decrease in TB cases from 30/4357 (0.7%) to 19/4780 (0.4%), a reduction of 0.29 cases/100 person years. References 1. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 2. Behrman AJ, Shofer FS. Tuberculosis exposure and control in an urban emergency department. Ann Emerg Med; 1998. 17WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 20 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can the use of particulate respirators reduce TB transmission in healthcare workers in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment Impact Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Reduction in LTBI incidence/prevalence (n=9) 9 1,2,3,4,5,6,7,8,9 observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Nine studies examined the effect of particulate respirators upon TST conversion. These studies produced effects in the same direction (reducing infection), however the magnitude of the effect varied considerably between settings. Concerns around confounding due to multiple interventions, and heterogeneity of the interventions, means that the findings were not meta-analyzed. Bangsberg 1997 compared the effect of respiratory masks and fit testing for staff against usual care, prior to the introduction of a new infection control policy. Comparing six months before (0/100, 0% in Jun 1993) to six months after (1/107 1% in Dec 1993) there was a 1% increase in conversion. Comparing the same control period (0% in Jun 1993) to the period 6-12 months after (0% in Jun 1993) there was no difference. Given the low event numbers, these findings were not of significance. Second, Baussano found that staff respiratory protection was associated in a reduction in TST conversion from 26.3/1000 person years to 9.4 / 1000 person years – a reduction of 16.9 / 1000 person years. Third, Blumberg 1995 showed a composite intervention with a particulate respiratory was associated in a reduction of TST conversions from 18/3579 (3.3%) to 25/5153 (0.4%), a 2.9% reduction. Fella 1995 showed that particulate respirators were associated with a reduction in TST conversion from 41/303 (13.5%) to 21/446 (4.7%), a reduction of 8.8%. Dust fume respirators had no effect. Maloney 1995 showed a composite intervention including molded surgical masks was associated with a reduction in TST conversion from 15/90 (16.7%) to 4/78 (5.1%), a reduction by 11.5%. In Yanai 2003, a composite intervention including mixed mode ventilation was associated with a decrease in TST conversions from 13/77 (16.9%) to 2/96 (2.1%) – a decrease of 14.8%. Roth 1995 showed a composite intervention including respirators for health workers was associated with a reduction of infection of between 4.1 and 12.4 conversions per 1,000 persons. Welbel 2009 saw a 4.1% reduction in TST conversions, as a part of a composite intervention. Da costa 2009 showed a reduction of 1.9 TST conversions per month, as a part of a composite intervention. ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n=1) Use of particulate respirators No use Relative (95% CI) Absolute (95% CI) 1 4,c observational studies serious a not serious very seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 19/4780 (0.4%) 30/4357 (0.7%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval 18WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 21 Prevention and Control 2019 UPDATE Explanations a. The included studies have a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and interventions are poorly described. The interventions comprise multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). c. Only one study evaluated this outcome. In Yanai 2003, a composite intervention including use of staff particulate respirators was associated with a decrease in TB cases from 30/4357 (0.7%) to 19/4780 (0.4%), a reduction in 0.29 cases/100 person years. References 1. Maloney SA, Pearson ML,Gordon MT,Del Castillo R,Boyle JF, Jarvis WR. Efficacy of control measures in preventing nosocomial transmission of multidrug-resistant tuberculosis to patients and health care workers. Ann Intern Med; 1995. 2. Fella P, Rivera P, Hale M, Squires K, Sepkowitz K. Dramatic increase in tuberculin skin test conversion rate among employees at a hospital in New York City. Am J Infect Control; 1995. 3. Baussano I, Bugiani M,Carosso A,Mairano D,Barocelli AP,Tagna M,Cascio V,Piccioni P,Arossa W. Risk of tuberculin conversion among healthcare workers and the adoption of preventive measures. Occup Environ Med; 2007. 4. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 5. Roth VR, Garrett DO,Laserson KF,Starling CE,Kritski AL,Medeiros EAS,Binkin N,Jarvis WR. A multicenter evaluation of tuberculin skin test positivity and conversion among health care workers in Brazilian hospitals.. Int J Tuberc Lung Dis; 2005. 6. Blumberg HM, Sotir M,Erwin M,Bachman R,Shulman JA. Risk of house staff tuberculin skin test conversion in an area with a high incidence of tuberculosis. Clin Infect Dis; 1998. 7. Bangsberg DR, Crowley K,Moss A,Dobkin JF,McGregor C,Neu HC. Reduction in tuberculin skin-test conversions among medical house staff associated with improved tuberculosis infection control practices. Infect Control Hosp Epidemiol; 1997. 8. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 9. da Costa P, Trajman A ,Mello FC,Goudinho S,Silva MA,Garret D,Ruffino-Netto A,Kritski AL. Administrative measures for preventing Mycobacterium tuberculosis infection among healthcare workers in a teaching hospital in Rio de Janeiro, Brazil. J Hosp Infect; 2009. 19WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 22 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can the use of particulate respirators reduce TB transmission in persons in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Use of particulate respirators No use Relative (95% CI) Absolute (95% CI) Reduction in LTBI incidence/prevalence (n=0) 0 - CRITICAL Reduction in TB incidence/prevalence (n=1) 1 1,a observational studies serious b not serious very seriousc not serious strong association all plausible residual confounding would suggest spurious effect, while no effect was observed 0/44 (0.0%) 26/90 (28.9%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence in people living with HIV (n=1) 1 1,a observational studies serious b not serious very seriousc not serious strong association all plausible residual confounding would suggest spurious effect, while no effect was observed 0/44 (0.0%) 26/90 (28.9%) not estimable ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval Explanations a. Moro 2000 evaluated the effect of mask use by people entering isolation rooms (including visitors). Surgical masks were used. At the same time, high-risk pentamidine use (a risk for increased cough and transmission) was also ceased. The effect of this intervention reflects a combination of multiple components. Incident MDR-TB reduced from 26/90 (29%) to 0/44 (0%) during the period after the intervention began. The reduction in MDR-TB incidence was 10.6 / 1,000 patient days. Confounding factors are likely, and the effect cannot only be attributed to the respiratory protection program. b. The included study has a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). c. The intervention comprises multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). References 1. Moro ML, Errante I Infuso A Sodano L Gori A Orcese CA Salamina G D'Amico C Besozii G Caggese L. Effectiveness of infection control measures in controlling a nosocomial outbreak of multidrug-resistant tuberculosis among HIV patients in Italy.. Int J Tuberc Lung Dis; 2000. 20WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables WHO Guidelines on Tuberculosis Infection 23 Prevention and Control 2019 UPDATE Author(s): University of Sydney Date: 27-29 March 2018 Question: Can the implementation of respiratory protection programs reduce TB transmission in healthcare workers in TB care or other high TB transmission risk settings when compared to transmission to the same populations in settings with no intervention or different interventions? Setting: International Certainty assessment N° of patients Effect Certainty Importance N° of studies Study design Risk of bias Inconsistency Indirectness Imprecision Other considerations Impact Reduction in LTBI incidence/prevalence (n= 4) 4 1,2,3,4 observational studies serious a not serious seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed Four studies reported on the evaluation of fit testing of particulate respirators, as a part of complex composite interventions. In Yanai 2003, a composite intervention including fit testing of HCWs was associated with a decrease in TST conversions from 13/77 (16.9%) to 2/96 (2.1%) – a decrease of 14.8%. Bangsberg 1997 compared the effect of particulate respirators and fit testing for staff against usual care, prior to the introduction of a new infection control policy. Comparing six months before (0/100, 0% in Jun 1993) to six months after (1/107, 1% in Dec 1993) there was a 1% increase in conversion. Comparing the same control period (0% in Jun 1993) to the period 6-12 months after (0% in Jun 1993) there was no difference. Given the low event numbers, these findings were not of significance. Therefore, the two studies show a stable or reduced proportion of HCWs developing LTBI. Welbel found a 4.3% reduction in TST conversions following introduction of particulate respirators and fit testing. Heterogeneity in the interventions precludes meta-analysis. Da Costa 2009 was a before-after study evaluating the effect of a composite administrative, engineering and respiratory protection intervention upon TST conversion among health workers. The respiratory protection component comprised education of health workers to use particulate respirators (N95 masks), including instructions for their use, maintenance and re-use. TST conversion was assessed at the start of the implementation of the intervention, and after it was implemented. The study found TST conversion decreased from 25/4307 person months (5.8 per 1,000 person months) in 1999-2001 to 15/3858 person months (3.9 per 1,000 person months) – a reduction of 1.9 conversions / person-months. ⨁◯◯◯ VERY LOW CRITICAL Reduction in TB incidence/prevalence (n= 1) Respiratory protection programmes No implementation Relative (95% CI) Absolute (95% CI) 1 1,c observational studies serious a not serious seriousb not serious all plausible residual confounding would suggest spurious effect, while no effect was observed 19/4780 (0.4%) 30/4357 (0.7%) not pooled see comment ⨁◯◯◯ VERY LOW CRITICAL CI: Confidence interval Explanations a. The included study has a high risk of bias (confounding relating to secular trends, non-randomised group allocation, lack of allocation concealment, no adjustment for confounding). b. Differences in intervention (applicability). The comparator and interventions are poorly described. The interventions comprise multiple simultaneous components, including engineering, respiratory protection and administrative controls (downgraded by one level). c. One study evaluated this outcome. In Yanai 2003, a composite intervention including fit testing for HCW masks was associated with a decrease in TB cases from 30/4357 (0.7%) to 19/4780 (0.4%), a reduction in 0.29 cases/100 person years. References 1. Yanai H, Limpakarnjanarat K,Uthaivoravit W,Mastro TD,Mori T,Tappero JW. Risk of Mycobacterium tuberculosis infection and disease among health care workers, Chiang Rai, Thailand. Int J Tuberc Lung Dis; 2003. 2. Bangsberg DR, Crowley K,Moss A,Dobkin JF,McGregor C,Neu HC. Reduction in tuberculin skin-test conversions among medical house staff associated with improved tuberculosis infection control practices. Infect Control Hosp Epidemiol; 1997. 3. Welbel SF, French AL,Bush P,DeGuzman D,Weinstein RA. Protecting health care workers from tuberculosis: a 10-year experience. Am J Infect Control; 2009. 4. da Costa P, Trajman A ,Mello FC,Goudinho S,Silva MA,Garret D,Ruffino-Netto A,Kritski AL. Administrative measures for preventing Mycobacterium tuberculosis infection among healthcare workers in a teaching hospital in Rio de Janeiro, Brazil. J Hosp Infect; 2009. 21WHO consolidated guidelines on tuberculosis. Web annex B. GRADE evidence summary tables For further information, please contact: Global TB Programme World Health Organization 20, Avenue Appia CH-1211 Geneva 27 Switzerland Website: www.who.int/tb