- Email: [email protected]
Tuberculosis infection control in rural South Africa: survey of knowledge, attitude and practice in hospital staff
Journal of Hospital Infection 79 (2011) 333e338
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Tuberculosis infection control in rural South Africa: survey of knowledge, attitude and practice in hospital staff Z. Kanjeea, d, *, K. Catterickb, d, A.P. Mollb, d, K.R. Amicoc, d, G.H. Friedlanda, d a
Yale University School of Medicine, New Haven, CT, USA Church of Scotland Hospital, Tugela Ferry, KwaZulu-Natal, South Africa c Center for Health, Intervention & Prevention, University of Connecticut, Storrs, CT, USA d Tugela Ferry Care and Research Collaboration, Tugela Ferry, KwaZulu-Natal, South Africa b
a r t i c l e i n f o
s u m m a r y
Article history: Received 12 December 2010 Accepted 7 June 2011 by J.A. Child Available online 5 October 2011
A baseline assessment of tuberculosis infection control (TB IC) knowledge, attitude and practice (KAP) was conducted among staff in a resource-limited rural South African hospital where nosocomially transmitted multi- and extensively drug-resistant (M/XDR) TB had been reported. Assessment consisted of anonymous questionnaires and direct observation during JulyeSeptember 2007, soon after the report of M/XDR-TB. Data were obtained from 57 questionnaires and 10 h of direct observation. While knowledge and attitudes were generally supportive of TB IC implementation, 49.1% of staff felt that the hospital did not care about them and/or was not working to prevent staff TB infections, and 42.9% were less willing to continue as a healthcare worker because of staff TB/MDR-TB/XDR-TB deaths. Practices were variable. The recent appointment of an IC officer and implementation of natural ventilation were strengths, but the facility lacked a TB IC policy, the patient TB screening process was inadequate, and 41.5% of respondents were unaware of their personal human immunodeficiency virus (HIV) status. Respondents reported a number of barriers to TB IC implementation such as concerns about the confidentiality of staff health information, the stigma of TB and HIV, inadequate resources, and patient non-compliance. Assessment of staff KAP provided useful data regarding deficits and barriers to TB IC, and helped to focus subsequent IC strategies. Given the critical importance of reducing nosocomial TB transmission, it is recommended that facilities should conduct simplified TB IC assessment, ensure the confidentiality of staff health information, address the stigma of TB/HIV, and implement multi-faceted TB IC facility and behavioural change interventions. Behavioural science methods have the potential to improve TB IC research and implementation. Ó 2011 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Infection control Knowledge/attitude/practice MDR/XDR-TB South Africa Tuberculosis
Introduction Tuberculosis (TB) is a major threat to global health. In 2008, there were 9.4 million new cases and 1.3 million deaths attributable to this infection.1 The emerging epidemics of multi- and extensively drug-resistant (M/XDR) TB further imperil patients and public health.2 With the advent of human immunodeficiency virus (HIV), TB is highly transmissible in resource-limited healthcare facilities.3 * Corresponding author. Address: Internal Medicine Residency Program, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA. Tel.: þ1 617 732 5775; fax: þ1 617 264 6346. E-mail address: [email protected] (Z. Kanjee).
Healthcare workers (HCWs) are at increased risk for acquiring TB in such settings.4 A recent retrospective chart review in South Africa found far higher rates of M/XDR-TB (diagnosis and admittance) among HCWs compared with the general population, which can lead to severe medical and psychosocial consequences.5,6 Experiences at the Church of Scotland Hospital (COSH), a district hospital in rural South Africa, further underscore the dangers of nosocomial transmission to patients and HCWs. A 2006 report from this facility described 53 patients, including four HCWs, infected with rapidly and almost uniformly fatal XDR-TB.7 Evidence indicated that this infection was likely to have been transmitted nosocomially.8 Global TB and HIV experts advocate infection control (IC) as a key TB control strategy.9 Although TB IC guidelines are available
0195-6701/$ e see front matter Ó 2011 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2011.06.017
334
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
for resource-limited settings, their implementation is believed to be inadequate.10,11 Mathematical modelling studies at COSH have demonstrated that simple and feasible TB IC and other steps in the facility could nearly halve the anticipated number of XDR-TB cases.12 This assessment was undertaken to characterize TB IC implementation in a resource-limited setting with reported nosocomial M/XDR-TB transmission.
Methods This cross-sectional baseline audit of staff TB IC knowledge, attitude and practice (KAP) was completed between July and September 2007 at COSH, a 350-bed government district hospital serving an impoverished population in rural KwaZulu-Natal, South Africa. Staff had received no formal training in TB IC, but had been informally instructed by medical staff regarding the use of open windows and respirators (i.e. N95 face masks). The TB case rate in the area is >1000/ 100,000/year, with 70% HIV co-infection. Regional antenatal HIV prevalence is >30%. An outpatient TB programme cares for >1200 cases/year, and COSH sees a large caseload of M/XDR-TB.
Data collection and analysis An anonymous, 100-item questionnaire was administered to HCWs. This instrument assessed TB IC knowledge (TB symptoms/ transmission and the use of surgical masks/respirators), attitude (opinions regarding various TB IC tasks and topics) and practice (frequency of, and barriers to, TB IC implementation). Unannounced direct observations were conducted each weekday for two weeks. Staff were not generally informed of the purpose of the direct observations. Implementation of natural and mechanical ventilation, use of masks by TB cases, and use of respirators by staff were directly observed and recorded in general medical and TB wards, and radiology and outpatient departments. The presence/ absence of TB IC policies and personnel, as well as routine TB
screening/triage and isolation procedures and facilities, were also assessed by observation. Data were organized into the three realms of TB IC, namely administrative (appoint a TB IC officer, develop a TB IC policy with monitoring and evaluation, separate cases/suspects, screen patients and separate/expedite suspect evaluation, cough hygiene, staff awareness of personal HIV status, redeploy HIV-positive staff from high-risk areas, staff seek TB diagnosis if symptomatic), environmental (natural/mechanical ventilation) and personal protective equipment (respirators for staff).10 Blank or invalid responses to the questionnaire were not included in the analysis. Questionnaire knowledge items were scored as correct/incorrect, attitude items were collapsed from five- to three-point rating scales, and the distributions of reported practice implementation frequency and barriers to greater implementation were recorded. Direct observations were compiled and compliance rates were calculated by hospital department. Data were used to describe TB IC KAP; knowledge and attitude were characterized by questionnaire, and practice was characterized by questionnaire and direct observation. Results Questionnaires were completed by 57 available HCWs. Of the respondents, 75.4% were female and 43.8% were professional or enrolled nurses. TB and medical ward workers comprised 40.4% and 36.8% of respondents, respectively. Career lengths ranged from less than six months (10.5%) to more than 20 years (5.3%), with most HCWs having worked for one to five years (35.1%). Ten hours of direct observation were conducted. Knowledge Respondents’ knowledge of TB symptoms, transmission and respirators is presented in Table I. Other than ‘fever’, >90% of staff were able to identify classic pulmonary TB symptoms. Transmission-related knowledge was high. Respondents were
Table I Knowledge of survey respondents Knowledge element Symptom identification
Transmission
Respirators
Item Night sweats (yes) Weight loss (yes) Cough for more than two weeks (yes) Coughing up blood (yes) Memory loss (no) Tiredness/malaise (yes) Pain with urination (no) Ear pain (no) Blurry vision (no) Watery eyes (no) Dizziness (no) Fever (yes) Headache (no) Many bacterial infections (no) Patients with active TB disease can infect people by coughing (true) TB is often spread from person to person through the air (true) TB is often spread from person to person by sexual contact (false) Patients with active TB disease can infect people by spitting (true) HIV-positive patients are more vulnerable to catching TB than HIV-negative patients (true) TB is often spread from person to person by blood (false) Patients with active TB disease can infect people by sharing food or drinks (false) Patients with active TB disease are more likely to infect others if they have a cough that produces a lot of sputum (true) Patients with active TB disease can infect people by talking (true) A wet or dirty N95 can still be used (false) Is this an N95 or a surgical mask? (surgical mask) Is this an N95 or a surgical mask? (N95) N95s protect healthcare workers and visitors by stopping TB particles from being breathed in (true) How often should you check if your N95 is airtight and does not allow any air to pass/leak? (every time you put on an N95)
TB, tuberculosis; HIV, human immunodeficiency virus.
% Correct 100.0 100.0 96.5 91.2 62.7 91.1 88.0 85.2 74.5 71.7 60.4 58.9 54.9 25.9 98.2 96.5 94.7 93.0 89.5 83.9 78.6 64.3 19.6 94.7 89.3 89.3 85.7 78.6
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
generally well informed regarding key routes of TB transmission. Knowledge of key respirator facts varied from 78.6% to 94.7%. Attitude Questionnaire data suggested that staff had several concerns about TB IC at the facility. Almost half (49.1%) of those surveyed felt that the hospital administration did not care about them and was not working hard to prevent nosocomial TB infection in staff. Three-quarters (75.4%) of respondents found the lack of a dedicated female MDR-TB ward ‘very distressing’. As several staff members had died from confirmed TB/MDR-TB/XDR-TB, 82.1% and 42.9% of respondents were less willing to work in high-risk areas of the hospital or to work as an HCW, respectively. Due to the lack of a formal policy, 37.5% of respondents were less likely to follow TB IC recommendations, and 23.2% saw TB IC as less important than they would have done if the facility had such a policy in place. All respondents agreed that ‘It is very important to prevent the spread of TB in this hospital’. The great majority (80.7e100.0%) of those surveyed agreed that most TB IC tasks (relocation of HIVpositive staff, surgical masks for TB cases/suspects, opening doors and windows, and cough hygiene instruction) were ‘important’ or ‘extremely important’, and were comfortable undertaking them (Figure 1), although 14.0% believed that the redeployment of HIVpositive staff from a high-risk area is ‘not at all important’. Many (40.4%) said that they would not have felt comfortable undergoing HIV testing at COSH, seeking a personal TB diagnosis at COSH if symptomatic (28.1%), or (hypothetically) requesting redeployment out of a high-risk area if they were HIV positive (19.3%). Practice Administrative practices were variable. An IC officer was appointed in early 2007 but no TB IC policy or monitoring was in
335
place. At the outpatient/admissions department, TB cases or suspects were not routinely identified or expedited through services, and there was no separate waiting area for these patients. Although TB wards existed, only the male TB ward had an additional isolation unit for confirmed MDR-TB cases. Additional direct observation data are presented in Table II. Cough hygiene was inconsistent. Most respondents (77.4%) reported ‘always’ informing coughing patients about cough hygiene, while 13.2% reported doing so ‘often’. Many respondents (38.0%) reported ‘always’ offering surgical masks to TB cases on the ward. Direct observation indicated that 0e32.1% of admitted TB cases wore masks. Many surveyed staff (41.5%) reported that they were unaware of their personal HIV status. Most respondents (65.5% and 69.1%) reported that doors and windows, respectively, were ‘always’ open in their work area. Direct observation (during winter days) showed large interdepartmental differences in natural ventilation, with open windows ranging from 35.3% in the outpatient TB office to 99.0% in the radiology department. Many (43.6%) respondents claimed that they ‘always’ check for a tight facial seal when using a respirator, while 19.6% reported that they did not know how to do so. More than half (54.7%) of those surveyed reported that they ‘always’ use a respirator when in a room with a TB case/suspect, while 37.7% claimed that they ‘often’ do so. Direct observation showed compliance ranging from 24.1% to 98.5% by department. Survey respondents identified various barriers to the implementation of TB IC activities (see Table III for barriers endorsed by >20% of respondents). Respondents cited patient non-compliance as a barrier to cough hygiene; patient complaints and occasional cold weather as barriers to natural ventilation; confidentiality and stigma concerns as barriers to staff testing themselves for TB/HIV, as well as redeployment of HIV-positive HCWs; and insufficient supplies, discomfort and appearance as impediments to wider use of respirators by staff.
Own HIV diagnosis
Own TB diagnosis if symptomatic
Importance: % saying task is ‘important’ or ‘extremely important’
Relocation if hypothetically HIV+ Making TB suspects/cases wear surgical masks when in the hospital Leaving doors open to prevent the spread of TB
Comfort: % who ‘agree’ or ‘strongly agree’ they are comfortable undertaking task
Leaving windows open to prevent the spread of TB Telling coughing patients to follow cough hygiene procedures 0.0
25.0
50.0
75.0
100.0
Figure 1. Importance of performing and comfort undertaking various tuberculosis (TB) infection control activities. Note that ‘Importance’ questions related to staff human immunodeficiency virus (HIV) and TB diagnoses refer generally to the importance of diagnosis, whereas the ‘Comfort’ questions refer specifically to comfort undergoing diagnosis at the Church of Scotland Hospital.
336
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
Table II Direct observation of mask use by tuberculosis (TB) patients, natural ventilation and staff use of respirators in key wards/departments Male TB ward Mask use No. of audits Total TB cases Total cases wearing masks Compliance (%) Windows No. of audits Total windows Total open windows Compliance (%) Doors No. of audits Total doors Total open doors Compliance (%) Mechanical ventilation units (MVUs) No. of audits Total MVUs Total MVUs turned on Compliance (%) Respirators No. of audits Total staff Total staff using respirators Compliance (%)
Female TB ward
Male MDR-TB ward
Outpatient TB office
Radiology
Outpatient department
26 543 0 0
25 690 0 0
25 88 0 0
25 53 17 32.1
e e e e
e e e e
26 1,144 971 84.9
25 850 636 74.8
25 250 194 77.6
25 275 97 35.3
25 200 198 99.0
20 420 168 40.0
26 26 26 100.0
25 50 44 88.0
25 25 25 100.0
25 25 18 72.0
25 25 25 100.0
20 20 19 95.0
26 26 26 100.0
25 25 25 100.0
e e e e
e e e e
e e e e
e e e e
26 66 65 98.5
25 76 73 96.1
25 21 19 90.5
20 79 19 24.1
e e e e
e e e e
MDR, multi-drug-resistant.
Discussion Anonymous questionnaires were conducted and direct observation of staff was undertaken to assess TB IC KAP in a resourcelimited district hospital in rural South Africa in 2007. The facility has a high prevalence of TB and HIV, and is the site of nosocomial transmission of M/XDR-TB. Relevant TB IC knowledge was generally high and attitudes were positive in most areas. Respondents considered many TB IC tasks to be important and felt comfortable undertaking them, but personal TB and HIV diagnosis, and redeployment of HIV-positive staff, were important exceptions. Measures of TB IC practices were variable, with good natural ventilation but inconsistent use of respirators by staff. Other deficits included the lack of a formal TB IC policy, a poor
TB patient screening and separation process, and staff being unaware of their personal HIV status. Respondents endorsed a number of barriers to better implementation of TB IC activities, several of which lend themselves to possible intervention. This assessment informed the design and implementation of a feasible and successful facility TB IC programme. These findings provide more specific details regarding TB IC implementation in resource-poor settings than previous assessments, several of which have only provided vague data by relying on answers to broad and/or binary questions.13,14 In terms of implementation, COSH compared well with other facilities in the developing world. This audit found HCWs with higher levels of knowledge than found at a South African tertiary hospital, higher self-reported frequency of respirator use than found at a Thai facility, more
Table III Barriers endorsed by >20% of respondents as preventing greater implementation of various tuberculosis infection control (TB IC) activities TB IC activity Cough hygiene Natural ventilation
Obtaining confidential HIV testing for self at facility
HIV-positive staff requesting redeployment from high-risk areas Requesting TB diagnosis for self if symptomatic
Respirator use in high-risk situations
HCW, healthcare worker.
Barriers
% Endorsing
‘Patients won’t listen’ ‘Tissues and surgical masks are not available’ ‘Patients complain when doors and windows are open’ ‘It is sometimes too cold’ ‘No one is in charge of this, so no one makes sure it gets done’ ‘HCWs think [Occupational Health] would not maintain confidentiality’ ‘It is frightening to get tested for HIV’ ‘HCWs do not want to know their status’ ‘HCWs think other staff would stigmatize them for getting a test’ ‘HCWs do not want to admit to their supervisors that they are HIV positive’ ‘If HCWs tell their supervisors, then their co-workers will find out they are HIV positive’ ‘Other staff members will think they are just pretending to be sick to get away from work’ ‘HCWs worry that other staff will think they have HIV if they have TB’ ‘HCWs fear potential rejection and stigma from staff if they have TB’ ‘HCWs do not trust the doctors to maintain confidentiality’ ‘HCWs think it is shameful to admit they might have TB’ ‘HCWs do not trust the health services to take care of them if they have TB’ ‘It is frightening to test for TB’ ‘Other staff members will think they are just pretending to be sick to get away from work’ ‘Sometimes there are not enough N95s’ ‘N95s are uncomfortable’ ‘HCWs are confused about how long they should wear N95s, so they do not wear them’ ‘N95s destroy HCWs’ hairstyles’ ‘No one has given HCWs good training on this’
35.8 26.4 73.6 66.0 22.6 65.5 54.5 43.6 41.8 60.4 43.4 35.8 66.7 51.9 48.1 42.6 33.3 24.1 22.2 66.7 42.6 38.9 31.5 22.2
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
consistent directly observed respirator use than found at a Brazilian TB reference hospital and 40 Chinese TB centres, and better cough hygiene procedures than found in a Peruvian emergency department.13,15e18 Like the situation in the Peruvian emergency ward, the patient TB screening and triage process at COSH was weak.18 This research may have demonstrated a discrepancy between self-report and direct observation. While >30% of respondents claimed that they ‘always’ offered surgical masks to TB cases/ suspects, 0e32.1% of TB cases were observed actually wearing masks. While some patients may have refused, it is likely that this difference is due to an overestimation in self-report. This discrepancy is consistent with a review of assessments of adherence to healthcare guidelines, which argued that self-report often overestimates compliance due to respondents’ tendency to over-report socially desirable (i.e. adherent) behaviour.19 Given that several of the TB IC assessments conducted to date rely heavily on self-report, some of the literature may overestimate TB IC implementation in resource-limited settings.13,14,18 TB IC self-report questions must be refined and phrased to provide specific opportunities for individuals to report non-adoption in a manner that recognizes a pull towards adherent self-presentation. Whenever possible, self-report should be supplemented by direct observation data. A number of survey respondents cited stigma or confidentiality concerns as barriers to staff TB/HIV testing and redeployment of HIV-positive HCWs. These findings are in line with a review finding that the stigma of HIV prevents HCWs from undertaking prevention activities and accessing care.20 Efforts to reduce the stigma of HIV among HCWs can be effective in increasing staff testing.21 Improvements in confidentiality are correlated with marked rises in HCW uptake of HIV services.22 Stigma education and improvements in confidentiality could make HIV-positive staff more likely to disclose their status to superiors in order to request redeployment. It is possible that reducing the stigma of TB and improving confidentiality could have similar benefits in terms of staff TB testing. Although the results of this survey suggest a critical role for facility policies, more effort is needed to ensure HCW implementation. Facilities must enable implementation by providing an adequate supply of necessary resources, such as respirators and cough hygiene materials. Additionally, hand hygiene research has shown that single training sessions do not create lasting change in staff practice, and sustained, multi-pronged interventions are most conducive to lasting behaviour change.23 Effective additional steps that facilities can consider include strategically placed reminders, routine observation and feedback, written materials, and promotion of knowledge of self-efficacy.23,24 Moreover, the application of sophisticated behavioural science methods to TB IC research could ensure better implementation of policies.25 This audit has several limitations. First, the survey was conducted in English, which was a second or third language for most participants. Second, although the questionnaire was anonymous, respondents may not have been truthful about their attitudes and practices, fearing that their supervisors would be informed of their responses. Third, the questionnaire was completed by a sample of the HCWs who were available to participate, and some departments were disproportionately over-represented. Finally, the research was conducted at a relatively atypical facility in that there had been substantial research attention on nosocomial transmission before this assessment; as such, the results may not be generalizable to all resource-limited hospitals. Nevertheless, this report provides new information that could be applied to similar institutional settings. The assessment of staff TB IC KAP provides useful baseline information regarding deficits and barriers to effective IC strategies. This approach can demonstrate a facility’s strengths and weaknesses, highlight pertinent HCW concerns, and determine where to concentrate efforts.
337
The following recommendations are made for resource-limited hospitals: e Prior to implementing TB IC interventions and training, facilities should conduct a simplified and targeted baseline TB IC assessment, including sensitive self-report items and direct observation. e Facilities should ensure confidentiality of staff health information and reduce the stigma of TB/HIV in order to improve HCW uptake of personal diagnosis and other risk-reduction strategies. e Rather than single training sessions, multiple behavioural change interventions should be used with ongoing monitoring and evaluation of effectiveness to ensure successful implementation. e Researchers and facility administrators should pursue the application of behavioural science methods to TB IC to strengthen the field and improve implementation.
Conflict of interest statement None declared. Funding sources Support from the Yale School of Medicine Office of Student Research, the Wilbur J. Downs International Health Travel Fellowship, the Gilead Foundation, the Doris Duke Charitable Foundation, the Irene Diamond Fund and the President’s Emergency Plan for AIDS Relief. Acknowledgements The authors thank R. Schwarz for helpful comments on this paper. References 1. World Health Organization. Global tuberculosis control: a short update to the 2009 report. Geneva: WHO; 2009. 2. Gandhi N, Nunn P, Dheda K, et al. Multidrug-resistant and extensively drugresistant tuberculosis: a threat to global control of tuberculosis. Lancet 2010;375:1830e1843. 3. Bock NN, Jensen PA, Miller B, Nardell E. Tuberculosis infection control in resource-limited settings in the era of expanding HIV care and treatment. J Infect Dis 2007;196:S108eS113. 4. Menzies D, Joshi R, Pai M. Risk of tuberculosis infection and disease associated with work in health care settings. Int J Tuberc Lung Dis 2007;11:593e605. 5. O’Donnell MR, Jarand J, Loveday M, et al. High incidence of hospital admissions with multidrug-resistant and extensively drug-resistant tuberculosis among South African health care workers. Ann Intern Med 2010;153:516e522. 6. Padayatchi N, Daftari A, Moodley T, Madansein R, Ramjee A. Case series of the long-term psychosocial impact of drug-resistant tuberculosis in HIV-negative medical doctors. Int J Tuberc Lung Dis 2010;8:960e966. 7. Gandhi NR, Moll A, Sturm AW, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 2006;368:1575e1580. 8. Andrews JR, Gandhi NR, Moodley P, et al. Exogenous reinfection as a cause of multidrug-resistant and extensively drug-resistant tuberculosis in rural South Africa. J Infect Dis 2008;198:1582e1589. 9. World Health Organization. WHO three I’s meeting: intensified case finding (ICF), isoniazid preventive therapy (IPT) and TB infection control. Report of a joint WHO HIV/AIDS and TB department meeting. Geneva: WHO; 2008. 10. Scano F. WHO policy on TB infection control in health-care facilities, congregate settings and households. Geneva: WHO; 2009. 11. Shenoi S, Escombe AR, Friedland G. Transmission of drug-susceptible and drugresistant tuberculosis and the critical importance of airborne infection control in the era of HIV infection and highly active antiretroviral therapy rollouts. Clin Infect Dis 2010;50:S231eS237. 12. Basu S, Andrews JR, Poolman EM, et al. Prevention of nosocomial transmission of extensively drug-resistant tuberculosis in rural South African district hospitals: an epidemiological modelling study. Lancet 2007;370:1500e1507. 13. Luksamijarulkul P, Supapvanit C, Loosereewanich P, Aiumlaor P. Risk assessment towards tuberculosis among hospital personnel: administrative control,
338
14.
15.
16.
17.
18.
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338 risk exposure, use of protective barriers and microbial air quality. Southeast Asian J Trop Med Public Health 2004;35:1005e1011. Harries AD, Hargreaves NJ, Gausi F, Kwanjana JH, Salaniponi FM. Preventing tuberculosis among health workers in Malawi. Bull World Health Organ 2002;80:526e531. Sissolak D, Bamford C, Mehtar S. The potential to transmit Mycobacterium tuberculosis at a South African tertiary teaching hospital. Int J Infect Dis 2010;14:e423ee428. Biscotto C, Pedroso E, Starling C, Roth V. Evaluation of N95 respirator use as a tuberculosis control measure in a resource-limited setting. Int J Tuberc Lung Dis 2005;9:545e549. He GX, van den Hof S, van der Werf MJ, et al. Infection control and the burden of tuberculosis infection and disease in health care workers in China: a crosssectional study. BMC Infect Dis 2010;10:313. Escombe A, Huaroto L, Ticona E, et al. Tuberculosis transmission risk and infection control in a hospital emergency department in Lima, Peru. Int J Tuberc Lung Dis 2010;14:1120e1126.
19. Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence of self-report bias in assessing adherence to guidelines. Int J Qual Health Care 1999;11:187e192. 20. Nyblade L, Stangl A, Weiss E, Ashburn K. Combating HIV stigma in health care settings: what works? J Int AIDS Soc 2009;12:15. 21. Uys L, Chirwa M, Kohi T, et al. Evaluation of a health setting-based stigma intervention in five African countries. AIDS Patient Care STDs 2009;23:1059e1066. 22. Uebel K, Friedland G, Pawinski R, Holst H. HAART for hospital health care workers: an innovative programme. S Afr Med J 2004;94:423e427. 23. Naikoba S, Hayward A. The effectiveness of interventions aimed at increasing handwashing in healthcare workers e a systematic review. J Hosp Infect 2001;47:173e180. 24. Tavolacci MP, Merle V, Pitrou I, Thillard D, Serra V, Czernichow P. Alcoholbased hand rub: influence of healthcare workers’ knowledge and perception on declared use. J Hosp Infect 2006;64:149e155. 25. Pittet D. The Lowbury lecture: behaviour in infection control. J Hosp Infect 2004;58:1e13.
Available online at www.sciencedirect.com
Journal of Hospital Infection journal homepage: www.elsevierhealth.com/journals/jhin
Tuberculosis infection control in rural South Africa: survey of knowledge, attitude and practice in hospital staff Z. Kanjeea, d, *, K. Catterickb, d, A.P. Mollb, d, K.R. Amicoc, d, G.H. Friedlanda, d a
Yale University School of Medicine, New Haven, CT, USA Church of Scotland Hospital, Tugela Ferry, KwaZulu-Natal, South Africa c Center for Health, Intervention & Prevention, University of Connecticut, Storrs, CT, USA d Tugela Ferry Care and Research Collaboration, Tugela Ferry, KwaZulu-Natal, South Africa b
a r t i c l e i n f o
s u m m a r y
Article history: Received 12 December 2010 Accepted 7 June 2011 by J.A. Child Available online 5 October 2011
A baseline assessment of tuberculosis infection control (TB IC) knowledge, attitude and practice (KAP) was conducted among staff in a resource-limited rural South African hospital where nosocomially transmitted multi- and extensively drug-resistant (M/XDR) TB had been reported. Assessment consisted of anonymous questionnaires and direct observation during JulyeSeptember 2007, soon after the report of M/XDR-TB. Data were obtained from 57 questionnaires and 10 h of direct observation. While knowledge and attitudes were generally supportive of TB IC implementation, 49.1% of staff felt that the hospital did not care about them and/or was not working to prevent staff TB infections, and 42.9% were less willing to continue as a healthcare worker because of staff TB/MDR-TB/XDR-TB deaths. Practices were variable. The recent appointment of an IC officer and implementation of natural ventilation were strengths, but the facility lacked a TB IC policy, the patient TB screening process was inadequate, and 41.5% of respondents were unaware of their personal human immunodeficiency virus (HIV) status. Respondents reported a number of barriers to TB IC implementation such as concerns about the confidentiality of staff health information, the stigma of TB and HIV, inadequate resources, and patient non-compliance. Assessment of staff KAP provided useful data regarding deficits and barriers to TB IC, and helped to focus subsequent IC strategies. Given the critical importance of reducing nosocomial TB transmission, it is recommended that facilities should conduct simplified TB IC assessment, ensure the confidentiality of staff health information, address the stigma of TB/HIV, and implement multi-faceted TB IC facility and behavioural change interventions. Behavioural science methods have the potential to improve TB IC research and implementation. Ó 2011 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved.
Keywords: Infection control Knowledge/attitude/practice MDR/XDR-TB South Africa Tuberculosis
Introduction Tuberculosis (TB) is a major threat to global health. In 2008, there were 9.4 million new cases and 1.3 million deaths attributable to this infection.1 The emerging epidemics of multi- and extensively drug-resistant (M/XDR) TB further imperil patients and public health.2 With the advent of human immunodeficiency virus (HIV), TB is highly transmissible in resource-limited healthcare facilities.3 * Corresponding author. Address: Internal Medicine Residency Program, Brigham and Women's Hospital, 75 Francis Street, Boston, MA 02115, USA. Tel.: þ1 617 732 5775; fax: þ1 617 264 6346. E-mail address: [email protected] (Z. Kanjee).
Healthcare workers (HCWs) are at increased risk for acquiring TB in such settings.4 A recent retrospective chart review in South Africa found far higher rates of M/XDR-TB (diagnosis and admittance) among HCWs compared with the general population, which can lead to severe medical and psychosocial consequences.5,6 Experiences at the Church of Scotland Hospital (COSH), a district hospital in rural South Africa, further underscore the dangers of nosocomial transmission to patients and HCWs. A 2006 report from this facility described 53 patients, including four HCWs, infected with rapidly and almost uniformly fatal XDR-TB.7 Evidence indicated that this infection was likely to have been transmitted nosocomially.8 Global TB and HIV experts advocate infection control (IC) as a key TB control strategy.9 Although TB IC guidelines are available
0195-6701/$ e see front matter Ó 2011 The Healthcare Infection Society. Published by Elsevier Ltd. All rights reserved. doi:10.1016/j.jhin.2011.06.017
334
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
for resource-limited settings, their implementation is believed to be inadequate.10,11 Mathematical modelling studies at COSH have demonstrated that simple and feasible TB IC and other steps in the facility could nearly halve the anticipated number of XDR-TB cases.12 This assessment was undertaken to characterize TB IC implementation in a resource-limited setting with reported nosocomial M/XDR-TB transmission.
Methods This cross-sectional baseline audit of staff TB IC knowledge, attitude and practice (KAP) was completed between July and September 2007 at COSH, a 350-bed government district hospital serving an impoverished population in rural KwaZulu-Natal, South Africa. Staff had received no formal training in TB IC, but had been informally instructed by medical staff regarding the use of open windows and respirators (i.e. N95 face masks). The TB case rate in the area is >1000/ 100,000/year, with 70% HIV co-infection. Regional antenatal HIV prevalence is >30%. An outpatient TB programme cares for >1200 cases/year, and COSH sees a large caseload of M/XDR-TB.
Data collection and analysis An anonymous, 100-item questionnaire was administered to HCWs. This instrument assessed TB IC knowledge (TB symptoms/ transmission and the use of surgical masks/respirators), attitude (opinions regarding various TB IC tasks and topics) and practice (frequency of, and barriers to, TB IC implementation). Unannounced direct observations were conducted each weekday for two weeks. Staff were not generally informed of the purpose of the direct observations. Implementation of natural and mechanical ventilation, use of masks by TB cases, and use of respirators by staff were directly observed and recorded in general medical and TB wards, and radiology and outpatient departments. The presence/ absence of TB IC policies and personnel, as well as routine TB
screening/triage and isolation procedures and facilities, were also assessed by observation. Data were organized into the three realms of TB IC, namely administrative (appoint a TB IC officer, develop a TB IC policy with monitoring and evaluation, separate cases/suspects, screen patients and separate/expedite suspect evaluation, cough hygiene, staff awareness of personal HIV status, redeploy HIV-positive staff from high-risk areas, staff seek TB diagnosis if symptomatic), environmental (natural/mechanical ventilation) and personal protective equipment (respirators for staff).10 Blank or invalid responses to the questionnaire were not included in the analysis. Questionnaire knowledge items were scored as correct/incorrect, attitude items were collapsed from five- to three-point rating scales, and the distributions of reported practice implementation frequency and barriers to greater implementation were recorded. Direct observations were compiled and compliance rates were calculated by hospital department. Data were used to describe TB IC KAP; knowledge and attitude were characterized by questionnaire, and practice was characterized by questionnaire and direct observation. Results Questionnaires were completed by 57 available HCWs. Of the respondents, 75.4% were female and 43.8% were professional or enrolled nurses. TB and medical ward workers comprised 40.4% and 36.8% of respondents, respectively. Career lengths ranged from less than six months (10.5%) to more than 20 years (5.3%), with most HCWs having worked for one to five years (35.1%). Ten hours of direct observation were conducted. Knowledge Respondents’ knowledge of TB symptoms, transmission and respirators is presented in Table I. Other than ‘fever’, >90% of staff were able to identify classic pulmonary TB symptoms. Transmission-related knowledge was high. Respondents were
Table I Knowledge of survey respondents Knowledge element Symptom identification
Transmission
Respirators
Item Night sweats (yes) Weight loss (yes) Cough for more than two weeks (yes) Coughing up blood (yes) Memory loss (no) Tiredness/malaise (yes) Pain with urination (no) Ear pain (no) Blurry vision (no) Watery eyes (no) Dizziness (no) Fever (yes) Headache (no) Many bacterial infections (no) Patients with active TB disease can infect people by coughing (true) TB is often spread from person to person through the air (true) TB is often spread from person to person by sexual contact (false) Patients with active TB disease can infect people by spitting (true) HIV-positive patients are more vulnerable to catching TB than HIV-negative patients (true) TB is often spread from person to person by blood (false) Patients with active TB disease can infect people by sharing food or drinks (false) Patients with active TB disease are more likely to infect others if they have a cough that produces a lot of sputum (true) Patients with active TB disease can infect people by talking (true) A wet or dirty N95 can still be used (false) Is this an N95 or a surgical mask? (surgical mask) Is this an N95 or a surgical mask? (N95) N95s protect healthcare workers and visitors by stopping TB particles from being breathed in (true) How often should you check if your N95 is airtight and does not allow any air to pass/leak? (every time you put on an N95)
TB, tuberculosis; HIV, human immunodeficiency virus.
% Correct 100.0 100.0 96.5 91.2 62.7 91.1 88.0 85.2 74.5 71.7 60.4 58.9 54.9 25.9 98.2 96.5 94.7 93.0 89.5 83.9 78.6 64.3 19.6 94.7 89.3 89.3 85.7 78.6
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
generally well informed regarding key routes of TB transmission. Knowledge of key respirator facts varied from 78.6% to 94.7%. Attitude Questionnaire data suggested that staff had several concerns about TB IC at the facility. Almost half (49.1%) of those surveyed felt that the hospital administration did not care about them and was not working hard to prevent nosocomial TB infection in staff. Three-quarters (75.4%) of respondents found the lack of a dedicated female MDR-TB ward ‘very distressing’. As several staff members had died from confirmed TB/MDR-TB/XDR-TB, 82.1% and 42.9% of respondents were less willing to work in high-risk areas of the hospital or to work as an HCW, respectively. Due to the lack of a formal policy, 37.5% of respondents were less likely to follow TB IC recommendations, and 23.2% saw TB IC as less important than they would have done if the facility had such a policy in place. All respondents agreed that ‘It is very important to prevent the spread of TB in this hospital’. The great majority (80.7e100.0%) of those surveyed agreed that most TB IC tasks (relocation of HIVpositive staff, surgical masks for TB cases/suspects, opening doors and windows, and cough hygiene instruction) were ‘important’ or ‘extremely important’, and were comfortable undertaking them (Figure 1), although 14.0% believed that the redeployment of HIVpositive staff from a high-risk area is ‘not at all important’. Many (40.4%) said that they would not have felt comfortable undergoing HIV testing at COSH, seeking a personal TB diagnosis at COSH if symptomatic (28.1%), or (hypothetically) requesting redeployment out of a high-risk area if they were HIV positive (19.3%). Practice Administrative practices were variable. An IC officer was appointed in early 2007 but no TB IC policy or monitoring was in
335
place. At the outpatient/admissions department, TB cases or suspects were not routinely identified or expedited through services, and there was no separate waiting area for these patients. Although TB wards existed, only the male TB ward had an additional isolation unit for confirmed MDR-TB cases. Additional direct observation data are presented in Table II. Cough hygiene was inconsistent. Most respondents (77.4%) reported ‘always’ informing coughing patients about cough hygiene, while 13.2% reported doing so ‘often’. Many respondents (38.0%) reported ‘always’ offering surgical masks to TB cases on the ward. Direct observation indicated that 0e32.1% of admitted TB cases wore masks. Many surveyed staff (41.5%) reported that they were unaware of their personal HIV status. Most respondents (65.5% and 69.1%) reported that doors and windows, respectively, were ‘always’ open in their work area. Direct observation (during winter days) showed large interdepartmental differences in natural ventilation, with open windows ranging from 35.3% in the outpatient TB office to 99.0% in the radiology department. Many (43.6%) respondents claimed that they ‘always’ check for a tight facial seal when using a respirator, while 19.6% reported that they did not know how to do so. More than half (54.7%) of those surveyed reported that they ‘always’ use a respirator when in a room with a TB case/suspect, while 37.7% claimed that they ‘often’ do so. Direct observation showed compliance ranging from 24.1% to 98.5% by department. Survey respondents identified various barriers to the implementation of TB IC activities (see Table III for barriers endorsed by >20% of respondents). Respondents cited patient non-compliance as a barrier to cough hygiene; patient complaints and occasional cold weather as barriers to natural ventilation; confidentiality and stigma concerns as barriers to staff testing themselves for TB/HIV, as well as redeployment of HIV-positive HCWs; and insufficient supplies, discomfort and appearance as impediments to wider use of respirators by staff.
Own HIV diagnosis
Own TB diagnosis if symptomatic
Importance: % saying task is ‘important’ or ‘extremely important’
Relocation if hypothetically HIV+ Making TB suspects/cases wear surgical masks when in the hospital Leaving doors open to prevent the spread of TB
Comfort: % who ‘agree’ or ‘strongly agree’ they are comfortable undertaking task
Leaving windows open to prevent the spread of TB Telling coughing patients to follow cough hygiene procedures 0.0
25.0
50.0
75.0
100.0
Figure 1. Importance of performing and comfort undertaking various tuberculosis (TB) infection control activities. Note that ‘Importance’ questions related to staff human immunodeficiency virus (HIV) and TB diagnoses refer generally to the importance of diagnosis, whereas the ‘Comfort’ questions refer specifically to comfort undergoing diagnosis at the Church of Scotland Hospital.
336
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
Table II Direct observation of mask use by tuberculosis (TB) patients, natural ventilation and staff use of respirators in key wards/departments Male TB ward Mask use No. of audits Total TB cases Total cases wearing masks Compliance (%) Windows No. of audits Total windows Total open windows Compliance (%) Doors No. of audits Total doors Total open doors Compliance (%) Mechanical ventilation units (MVUs) No. of audits Total MVUs Total MVUs turned on Compliance (%) Respirators No. of audits Total staff Total staff using respirators Compliance (%)
Female TB ward
Male MDR-TB ward
Outpatient TB office
Radiology
Outpatient department
26 543 0 0
25 690 0 0
25 88 0 0
25 53 17 32.1
e e e e
e e e e
26 1,144 971 84.9
25 850 636 74.8
25 250 194 77.6
25 275 97 35.3
25 200 198 99.0
20 420 168 40.0
26 26 26 100.0
25 50 44 88.0
25 25 25 100.0
25 25 18 72.0
25 25 25 100.0
20 20 19 95.0
26 26 26 100.0
25 25 25 100.0
e e e e
e e e e
e e e e
e e e e
26 66 65 98.5
25 76 73 96.1
25 21 19 90.5
20 79 19 24.1
e e e e
e e e e
MDR, multi-drug-resistant.
Discussion Anonymous questionnaires were conducted and direct observation of staff was undertaken to assess TB IC KAP in a resourcelimited district hospital in rural South Africa in 2007. The facility has a high prevalence of TB and HIV, and is the site of nosocomial transmission of M/XDR-TB. Relevant TB IC knowledge was generally high and attitudes were positive in most areas. Respondents considered many TB IC tasks to be important and felt comfortable undertaking them, but personal TB and HIV diagnosis, and redeployment of HIV-positive staff, were important exceptions. Measures of TB IC practices were variable, with good natural ventilation but inconsistent use of respirators by staff. Other deficits included the lack of a formal TB IC policy, a poor
TB patient screening and separation process, and staff being unaware of their personal HIV status. Respondents endorsed a number of barriers to better implementation of TB IC activities, several of which lend themselves to possible intervention. This assessment informed the design and implementation of a feasible and successful facility TB IC programme. These findings provide more specific details regarding TB IC implementation in resource-poor settings than previous assessments, several of which have only provided vague data by relying on answers to broad and/or binary questions.13,14 In terms of implementation, COSH compared well with other facilities in the developing world. This audit found HCWs with higher levels of knowledge than found at a South African tertiary hospital, higher self-reported frequency of respirator use than found at a Thai facility, more
Table III Barriers endorsed by >20% of respondents as preventing greater implementation of various tuberculosis infection control (TB IC) activities TB IC activity Cough hygiene Natural ventilation
Obtaining confidential HIV testing for self at facility
HIV-positive staff requesting redeployment from high-risk areas Requesting TB diagnosis for self if symptomatic
Respirator use in high-risk situations
HCW, healthcare worker.
Barriers
% Endorsing
‘Patients won’t listen’ ‘Tissues and surgical masks are not available’ ‘Patients complain when doors and windows are open’ ‘It is sometimes too cold’ ‘No one is in charge of this, so no one makes sure it gets done’ ‘HCWs think [Occupational Health] would not maintain confidentiality’ ‘It is frightening to get tested for HIV’ ‘HCWs do not want to know their status’ ‘HCWs think other staff would stigmatize them for getting a test’ ‘HCWs do not want to admit to their supervisors that they are HIV positive’ ‘If HCWs tell their supervisors, then their co-workers will find out they are HIV positive’ ‘Other staff members will think they are just pretending to be sick to get away from work’ ‘HCWs worry that other staff will think they have HIV if they have TB’ ‘HCWs fear potential rejection and stigma from staff if they have TB’ ‘HCWs do not trust the doctors to maintain confidentiality’ ‘HCWs think it is shameful to admit they might have TB’ ‘HCWs do not trust the health services to take care of them if they have TB’ ‘It is frightening to test for TB’ ‘Other staff members will think they are just pretending to be sick to get away from work’ ‘Sometimes there are not enough N95s’ ‘N95s are uncomfortable’ ‘HCWs are confused about how long they should wear N95s, so they do not wear them’ ‘N95s destroy HCWs’ hairstyles’ ‘No one has given HCWs good training on this’
35.8 26.4 73.6 66.0 22.6 65.5 54.5 43.6 41.8 60.4 43.4 35.8 66.7 51.9 48.1 42.6 33.3 24.1 22.2 66.7 42.6 38.9 31.5 22.2
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338
consistent directly observed respirator use than found at a Brazilian TB reference hospital and 40 Chinese TB centres, and better cough hygiene procedures than found in a Peruvian emergency department.13,15e18 Like the situation in the Peruvian emergency ward, the patient TB screening and triage process at COSH was weak.18 This research may have demonstrated a discrepancy between self-report and direct observation. While >30% of respondents claimed that they ‘always’ offered surgical masks to TB cases/ suspects, 0e32.1% of TB cases were observed actually wearing masks. While some patients may have refused, it is likely that this difference is due to an overestimation in self-report. This discrepancy is consistent with a review of assessments of adherence to healthcare guidelines, which argued that self-report often overestimates compliance due to respondents’ tendency to over-report socially desirable (i.e. adherent) behaviour.19 Given that several of the TB IC assessments conducted to date rely heavily on self-report, some of the literature may overestimate TB IC implementation in resource-limited settings.13,14,18 TB IC self-report questions must be refined and phrased to provide specific opportunities for individuals to report non-adoption in a manner that recognizes a pull towards adherent self-presentation. Whenever possible, self-report should be supplemented by direct observation data. A number of survey respondents cited stigma or confidentiality concerns as barriers to staff TB/HIV testing and redeployment of HIV-positive HCWs. These findings are in line with a review finding that the stigma of HIV prevents HCWs from undertaking prevention activities and accessing care.20 Efforts to reduce the stigma of HIV among HCWs can be effective in increasing staff testing.21 Improvements in confidentiality are correlated with marked rises in HCW uptake of HIV services.22 Stigma education and improvements in confidentiality could make HIV-positive staff more likely to disclose their status to superiors in order to request redeployment. It is possible that reducing the stigma of TB and improving confidentiality could have similar benefits in terms of staff TB testing. Although the results of this survey suggest a critical role for facility policies, more effort is needed to ensure HCW implementation. Facilities must enable implementation by providing an adequate supply of necessary resources, such as respirators and cough hygiene materials. Additionally, hand hygiene research has shown that single training sessions do not create lasting change in staff practice, and sustained, multi-pronged interventions are most conducive to lasting behaviour change.23 Effective additional steps that facilities can consider include strategically placed reminders, routine observation and feedback, written materials, and promotion of knowledge of self-efficacy.23,24 Moreover, the application of sophisticated behavioural science methods to TB IC research could ensure better implementation of policies.25 This audit has several limitations. First, the survey was conducted in English, which was a second or third language for most participants. Second, although the questionnaire was anonymous, respondents may not have been truthful about their attitudes and practices, fearing that their supervisors would be informed of their responses. Third, the questionnaire was completed by a sample of the HCWs who were available to participate, and some departments were disproportionately over-represented. Finally, the research was conducted at a relatively atypical facility in that there had been substantial research attention on nosocomial transmission before this assessment; as such, the results may not be generalizable to all resource-limited hospitals. Nevertheless, this report provides new information that could be applied to similar institutional settings. The assessment of staff TB IC KAP provides useful baseline information regarding deficits and barriers to effective IC strategies. This approach can demonstrate a facility’s strengths and weaknesses, highlight pertinent HCW concerns, and determine where to concentrate efforts.
337
The following recommendations are made for resource-limited hospitals: e Prior to implementing TB IC interventions and training, facilities should conduct a simplified and targeted baseline TB IC assessment, including sensitive self-report items and direct observation. e Facilities should ensure confidentiality of staff health information and reduce the stigma of TB/HIV in order to improve HCW uptake of personal diagnosis and other risk-reduction strategies. e Rather than single training sessions, multiple behavioural change interventions should be used with ongoing monitoring and evaluation of effectiveness to ensure successful implementation. e Researchers and facility administrators should pursue the application of behavioural science methods to TB IC to strengthen the field and improve implementation.
Conflict of interest statement None declared. Funding sources Support from the Yale School of Medicine Office of Student Research, the Wilbur J. Downs International Health Travel Fellowship, the Gilead Foundation, the Doris Duke Charitable Foundation, the Irene Diamond Fund and the President’s Emergency Plan for AIDS Relief. Acknowledgements The authors thank R. Schwarz for helpful comments on this paper. References 1. World Health Organization. Global tuberculosis control: a short update to the 2009 report. Geneva: WHO; 2009. 2. Gandhi N, Nunn P, Dheda K, et al. Multidrug-resistant and extensively drugresistant tuberculosis: a threat to global control of tuberculosis. Lancet 2010;375:1830e1843. 3. Bock NN, Jensen PA, Miller B, Nardell E. Tuberculosis infection control in resource-limited settings in the era of expanding HIV care and treatment. J Infect Dis 2007;196:S108eS113. 4. Menzies D, Joshi R, Pai M. Risk of tuberculosis infection and disease associated with work in health care settings. Int J Tuberc Lung Dis 2007;11:593e605. 5. O’Donnell MR, Jarand J, Loveday M, et al. High incidence of hospital admissions with multidrug-resistant and extensively drug-resistant tuberculosis among South African health care workers. Ann Intern Med 2010;153:516e522. 6. Padayatchi N, Daftari A, Moodley T, Madansein R, Ramjee A. Case series of the long-term psychosocial impact of drug-resistant tuberculosis in HIV-negative medical doctors. Int J Tuberc Lung Dis 2010;8:960e966. 7. Gandhi NR, Moll A, Sturm AW, et al. Extensively drug-resistant tuberculosis as a cause of death in patients co-infected with tuberculosis and HIV in a rural area of South Africa. Lancet 2006;368:1575e1580. 8. Andrews JR, Gandhi NR, Moodley P, et al. Exogenous reinfection as a cause of multidrug-resistant and extensively drug-resistant tuberculosis in rural South Africa. J Infect Dis 2008;198:1582e1589. 9. World Health Organization. WHO three I’s meeting: intensified case finding (ICF), isoniazid preventive therapy (IPT) and TB infection control. Report of a joint WHO HIV/AIDS and TB department meeting. Geneva: WHO; 2008. 10. Scano F. WHO policy on TB infection control in health-care facilities, congregate settings and households. Geneva: WHO; 2009. 11. Shenoi S, Escombe AR, Friedland G. Transmission of drug-susceptible and drugresistant tuberculosis and the critical importance of airborne infection control in the era of HIV infection and highly active antiretroviral therapy rollouts. Clin Infect Dis 2010;50:S231eS237. 12. Basu S, Andrews JR, Poolman EM, et al. Prevention of nosocomial transmission of extensively drug-resistant tuberculosis in rural South African district hospitals: an epidemiological modelling study. Lancet 2007;370:1500e1507. 13. Luksamijarulkul P, Supapvanit C, Loosereewanich P, Aiumlaor P. Risk assessment towards tuberculosis among hospital personnel: administrative control,
338
14.
15.
16.
17.
18.
Z. Kanjee et al. / Journal of Hospital Infection 79 (2011) 333e338 risk exposure, use of protective barriers and microbial air quality. Southeast Asian J Trop Med Public Health 2004;35:1005e1011. Harries AD, Hargreaves NJ, Gausi F, Kwanjana JH, Salaniponi FM. Preventing tuberculosis among health workers in Malawi. Bull World Health Organ 2002;80:526e531. Sissolak D, Bamford C, Mehtar S. The potential to transmit Mycobacterium tuberculosis at a South African tertiary teaching hospital. Int J Infect Dis 2010;14:e423ee428. Biscotto C, Pedroso E, Starling C, Roth V. Evaluation of N95 respirator use as a tuberculosis control measure in a resource-limited setting. Int J Tuberc Lung Dis 2005;9:545e549. He GX, van den Hof S, van der Werf MJ, et al. Infection control and the burden of tuberculosis infection and disease in health care workers in China: a crosssectional study. BMC Infect Dis 2010;10:313. Escombe A, Huaroto L, Ticona E, et al. Tuberculosis transmission risk and infection control in a hospital emergency department in Lima, Peru. Int J Tuberc Lung Dis 2010;14:1120e1126.
19. Adams AS, Soumerai SB, Lomas J, Ross-Degnan D. Evidence of self-report bias in assessing adherence to guidelines. Int J Qual Health Care 1999;11:187e192. 20. Nyblade L, Stangl A, Weiss E, Ashburn K. Combating HIV stigma in health care settings: what works? J Int AIDS Soc 2009;12:15. 21. Uys L, Chirwa M, Kohi T, et al. Evaluation of a health setting-based stigma intervention in five African countries. AIDS Patient Care STDs 2009;23:1059e1066. 22. Uebel K, Friedland G, Pawinski R, Holst H. HAART for hospital health care workers: an innovative programme. S Afr Med J 2004;94:423e427. 23. Naikoba S, Hayward A. The effectiveness of interventions aimed at increasing handwashing in healthcare workers e a systematic review. J Hosp Infect 2001;47:173e180. 24. Tavolacci MP, Merle V, Pitrou I, Thillard D, Serra V, Czernichow P. Alcoholbased hand rub: influence of healthcare workers’ knowledge and perception on declared use. J Hosp Infect 2006;64:149e155. 25. Pittet D. The Lowbury lecture: behaviour in infection control. J Hosp Infect 2004;58:1e13.