Introduction

Healthcare workers (HCW) and other hospital occupants are at increased risk of tuberculosis (TB) infection in high TB burden countries [1]. In Bangladesh, TB specialized hospitals admit and treat patients with drug-susceptible and drug-resistant TB as inpatients. Prior studies show that TB specialized hospitals implement TB IPC guidelines regularly [2]. On the other hand, tertiary care general hospitals admit presumptive TB patients for diagnosis and occasionally provide treatment to TB patients with comorbidities. Prior studies that pulmonary TB patients stay on average 5.5 days. There was the minimal implementation of TB IPC measures in public tertiary care general hospitals [3]. In addition, pulmonary TB patients in inpatients wards with limited or no IPC measures have been shown to increase HCW exposure to TB, resulting in 42% of latent TB infections among HCW [4].

TB infection prevention and control (IPC) is one of the World Health Organization’s (WHO) 12 collaborative activities to prevent TB [5]. In low-and middle-income countries (LMIC), TB IPC is considered a critical strategy for TB prevention and control in health settings [6]. International TB IPC guidelines that target health settings can be traced back to the 1980s, and the most recent TB IPC guidelines was updated and published in 2019 by the WHO [7, 8]. As part of the WHO’s TB control strategy, many LMIC have developed national TB IPC guidelines; however, the implementation of these guidelines remains inconsistent due to multifaced factors prevailing in diverse health settings [8]. Bangladesh developed its national TB IPC guidelines in 2011, which includes a three-level hierarchy of control measures: administrative, environmental, and personal respiratory protection [9]. In 2019, Bangladesh ranked seventh among the 30 high TB burden countries globally [10]. The estimated incidence rate of all forms of TB per 100,000 population was 221 [10].

Bangladesh’s National TB control program (NTP) is working to strengthen TB IPC in health settings. However, there has been no published literature to date that assessed the preparedness of hospitals to comply with the recommendations. To support NTP, it is crucial to understand the local context and broader health system factors, including behavioural and organizational factors [11]. An assessment of target facilities to understand the barriers, leadership presence, attitudes of frontline HCW, and local leaders in TB IPC programs can support the tailoring and implementing guidelines [12, 13]. We adopted the WHO IPC Assessment Framework (IPCAF) tool to conduct the hospital preparedness assessments for TB IPC [14]. The assessment aimed to examine HCW’ knowledge and attitudes towards TB IPC guidelines and the preparedness of two public tertiary care hospitals in Bangladesh to implement the guidelines.

Materials and methods

Study sites and participants

Between January and December 2019, a qualitative study was undertaken involving two public tertiary care hospitals: Hospital A and Hospital B. Key characteristics of the hospitals have been presented in Table 1. The rationale for selecting these hospitals was that they admitted approximately 400 adult pulmonary TB patients annually. Staff members involved in TB patient management were recruited from: medicine wards, directly observed therapy short-course (DOTS) clinics, radiology department, and administrative units. To recruit participants for interviews and focus group discussions (FGD), the team made a list of HCW who were either members of hospital committees, or those who were deemed to be influential in decision-making, knowledgeable, and experienced in dealing with infectious disease patients, or those staff who were seen as influence other’s practices. The team then selected those with the highest duration working in tertiary care hospitals.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 1. Key characteristics of the study hospitals, 2019.

https://doi.org/10.1371/journal.pone.0263115.t001

Results

The team conducted 16 KIIs and four FGDs. The field team approached 18 HCWs for KIIs and 38 HCWs for FGDs, where 16 HCW consented to participate as key informants and 35 consented as FGD participants. Among all participants, 29 were doctors, 20 were nurses, and two were ancillary workers. The participants’ mean age was 39 years (standard deviation (SD)-9.6), and 32 were male. The key-informants had an average of 20.6 years (SD:10.8) of work experience, and the FGD participants had an average of 6.8 years (SD:6.3) of work experience in public tertiary care hospitals. We report policy level, hospital level, and behavioural level factors that may impact the implementation of TB IPC measures.

The reported role that healthcare workers can play in implementing TB IPC

The participants pointed out several factors that can be followed to better implement the TB IPC guidelines (Table 2). The nurses said that the doctors take a patient history and may be aware of presumptive TB patients. The doctors may inform the nurses about presumptive TB patients, and the nurses can take necessary TB IPC measures. The nurses also mentioned that they could put a sign near a TB patient bed so that HCW can take standard precautions while visiting a TB patient. The nurses also added that they could use curtains around presumptive TB patients’ beds as they have space limitations for isolation. One of the key-informant said-

The patients enter the hospital through outdoors and emergency departments. We want TB IPC implementation, including cough screening at these two places. We also need to introduce the TB IPC control measures among the senior and junior level HCW. There should be frequent training so that HCW can attain the training by rotation. We also need infrastructure for TB screening and isolation. There is a need to collaborate with the government, hospital, and private sectors to reduce TB infection in the hospital.

Download:

• PPT
  PowerPoint slide
• PNG
  larger image
• TIFF
  original image

Table 2. Participants’ recommendations for TB IPC implementation in health settings.

https://doi.org/10.1371/journal.pone.0263115.t002

Lastly, it was suggested that local hospital HCW should be invited to the national TB IPC guidelines development committee. He said,

Because HCW work at the local level and implement guidelines, their opinions should be reflected in guidelines. The central policymakers develop guidelines based on their perceptions and theoretical knowledge. Due to the lack of local level HCW views in the policy, the hospital contexts may not be reflected in the policy. Therefore, the policy may fail during the implementation stage.

Discussion

This study identified gaps in the TB IPC policies, hospital infrastructure, IPC supplies and equipment, which impact the TB IPC implementation. The findings showed that due to low levels of awareness about the national TB IPC guidelines, both the hospitals were not adhering to TB IPC and surveillance program, which is consistent with a nationally representative survey that found facilities lacked 85% of standard precaution components required for TB management and IPC in Bangladesh [20]. In the study facilities, identifying presumptive TB patients on admission was hindered by a lack of cough screening and a case definition. Separation and isolation of presumptive TB patients were affected by limited space, high patient load, limited supply of consumable IPC resources such as masks and respirators, HCW’ IPC training gaps, and absence of isolation guidelines.

This study identified gaps in national TB IPC guideline recommendations and the hospital context. The guideline recommendations were based on tenuous theoretical principles informed by evidence from high-income low TB burden countries and underestimated resources required for implementation. Due to a lack of dedicated TB wards and clinics, TB patients were in all the adult medicine wards in the study hospitals. As outlined in the WHO’s guidelines on Core Components of IPC, national IPC programs must ensure that infrastructure and IPC resources are available to implement IPC measures successfully [16]. The challenge is translating these recommendations to hospitals that have high bed-occupancy rate of more than 150%, overcrowding on admission days, and understaffing. Since these hospitals receive funding according to the bed capacity, it is likely that they continue to exhaust their resources and so cannot meet the TB IPC needs of staff when caring for the extra patients.

To assist TB IPC interventions and to detect TB outbreaks among patients and HCWs, WHO recommends healthcare associated TB infection surveillance in health settings [7]. While both study hospitals have laboratory capacities to support healthcare associated TB surveillance, it is not occurring. While this was beyond the scope of the project to explore the perspectives of those in government, it may be speculated that awareness about the burden of nosocomial TB in tertiary care hospitals may be low, leading to a failure to prioritize the issue. If surveillance data was available, it could be shared with NTP, hospital and department managers to motivate them to allocate resources for TB-specific IPC programs.

For TB care and control, introducing electronic medical records is one of the WHO’s priority that may provide important backing to different components of WHO’s End TB strategy’s pillars 1 and pillars 2 [21–23]. Paper based recording and reporting increase diagnostic delay, which ultimately increases the length of hospital stay for pulmonary TB patients [24]. From our previous work, we identified that it took around five days to diagnose a patient with TB from admission, while the median length of hospital stay for a pulmonary TB patient without TB treatment was 5.4 days [3]. In the two study hospitals, we identified several key factors that are leading to the delay in diagnosing TB including: (1) a high number of patients; (2) patient admission after office hours; (3) opening time of diagnostic facilities (8:30 AM to 3 PM), and (4) the lack of an EMR system. In alignment with previous calls for action, we feel that it is critical that action be taken to reduce the time taken to diagnose untreated pulmonary TB patients [24]. As recommended by the participants, an EMR system could help to minimize the time these patients spend at the patient registration desk and at diagnostic facilities, as well as helping to improve reporting times. The implementation of EMR systems is aligned with the Bangladesh government’s national policy [25].

The WHO Core Components of IPC guidelines recommends that HCWs should receive education and training on IPC guidelines recommendations [7]. Amongst our participants, none reportedly received any education and training on TB IPC. A few HCWs reportedly received training on clinical management of TB patients that included TB IPC. However, the discussion on TB IPC in training might not be sufficient to motivate the HCWs to use their training to implement TB IPC in the hospitals. Due to limited training dedicated to TB IPC along with a lack of hospital-level TB IPC guidelines, there was no or minimal planning or execution of the IPC measures in the study hospitals The lack of training among nurses and ancillary workers was perceived as barriers to TB IPC; several participants called for the TB IPC guidelines and training on recommendations. Organizing training at these hospitals, promoting core and TB IPC and engaging all HCW, including doctors, nurses, and ancillary workers, may build partnership in the IPC program to help to implement the TB IPC measures.

In resourced limited settings, the WHO guidelines on TB IPC recommends natural ventilation in health settings [7]. Our findings showed that natural ventilation in the hospitals was affected by renovation and redistribution of internal space, cold winter air, mosquito nuisance, and by patients on the floor near doors and windows. To prevent airborne infection in the inpatients wards, doors and windows should be kept open day and nights and in all seasons [26]. On some occasions, natural ventilation can be more effective than mechanical ventilation in maintaining airflow and reducing TB transmission [7, 27]. We recommend that medicine, gynaecology, and obstetrics wards keep doors, windows, and other openings open until midnight in all seasons [28]. To ensure adequate ventilation, ancillary workers can be engaged to open the doors and windows in the morning, and nurses can be engaged for monitoring [27, 29]. Moreover, stickers/signage depicting ‘keep doors/windows open’ on the doors and windows and health education on the benefit of keeping doors and windows open can be given to patients and their family caregivers. Family caregivers can also be potential partners to implement natural ventilation intervention [30]. Large fans can be used above the doors and windows to maintain cross-ventilation to ensure the minimum air change per hour. Moreover, airbricks, window shutters etc., can stop the rain from coming in and control the cold breeze during the winter season [5]. Hospital stakeholders should address the factors that hindered opening doors and windows to maintain a minimum air change of six to 12 per hour recommended by WHO as airborne precaution [31].

Risk perceptions towards TB disease can also have an impact on acceptance and use of TB IPC measures [32]. Whilst our participants were aware that they are being exposed to pulmonary TB patients and that they were at risk of TB infection; their risk perception did not translate into action. Limited understanding about disease transmission and poor risk perception may result in non-compliance, poor health-seeking behavior, and treatment that may result in TB transmission in the hospital [33]. The perception amongst participants that health workers could be a source of TB transmission in the hospital was poor, with many HCW continuing to work when they were sick or hide their TB disease status while working in the hospital. The reasons for continuing working while they were still sick could be their fear of losing salary when they will be on sick leave; and fear of being re-assigned or even fired as noted by Delft et al [34]. The reason to hide TB status could be due to the fear of being stigmatized by co-workers. In a study among 879 HCW in South Africa, 19% of the respondents agreed that HCWs with presumptive TB are stigmatized in hospitals [35]. The HCW with TB disease can be a source of infection in the inpatient ward and can affect the effectiveness of TB IPC measures if implemented. The WHO 2019 TB IPC guidelines recommend educating HCWs, patients, family caregivers and community members about causes of stigma and discrimination that can be followed in public tertiary care hospitals to reduce TB-related stigma [7].

This study has limitations. We conducted this study in only two hospitals, and therefore, the study findings are very context specific. The findings presented study can be an indicative of hospital preparedness for TB IPC implementation in similar hospitals in Bangladesh, and other low-income high TB burden countries. Moreover, the study findings are consistent with a prior study that assessed the TB IPC in 11 TB specialized hospitals, tertiary care hospitals, and TB clinics [2]. Secondly, we did not interview any stakeholders from the ministry of health or NTP. Therefore, we could not understand the national-level factors that may influence TB IPC implementation in the local level tertiary care hospitals. Finally, this assessment identified strengths and weaknesses in health systems and services to adopt TB IPC, and the context may change over time and impact effectiveness.

In conclusion, this study found that while there was a positive attitude towards TB IPC and staff members are willing to implement strategies, there are key barriers that need to be addressed. Currently the study hospitals will struggle to implement the TB IPC measure due to high patient load, limited space, and lack of TB IPC specific budget and resources. Establishing new hospitals to reduce patient load is time and resource expensive. Therefore, we recommend leveraging existing resources for the implementation of TB IPC measures. We recommend that the space limitation for presumptive TB patient isolation be minimized by setting up an isolation area in the veranda or corridors with a large open area facing outside. If it is not possible during the rainy monsoon season, a curtain can separate patients inside the inpatient ward. Prior studies showed that the use of curtains had ben efficient in reducing the spread of TB bacteria during coughing, sneezing, and other aerosol-generating activities [36]. All coughing patients should wear a mask throughout while in hospital or until TB disease is ruled out as well other contagious respiratory infections. Besides, the ancillary workers can be engaged in IPC programs to keep the doors and windows open to ensure natural ventilation. The national TB program or other national and international organizations working on TB elimination can visit these hospitals, form TB IPC committees and organize TB IPC training in the hospitals so that the HCW fully understand TB IPC guidelines recommendations and build the task into their daily activities. Some nurses or senior nursing students can be trained on TB advocacy and counselling to provide health education, including respiratory hygiene and counselling, to support TB patients and their family caregivers. All HCW should be given information and motivation, including a care package, so they feel encouraged to undergo TB diagnostic investigation if they have signs and symptoms suggestive of TB. Finally, we recommend public tertiary care hospitals receive a budget according to the annual number of patients, not according to the allocated beds for the hospitals in Bangladesh.

Supporting information

Acknowledgments

We want to thank the study hospitals’ directors and all the study participants for their time and respect. icddr,b acknowledges with gratitude the commitment of CDC to its research efforts. icddr,b is also grateful to the Governments of Bangladesh, Canada, Sweden, and the UK for providing core/unrestricted support.