Being safe, feeling safe, and stigmatizing attitude among primary health care staff in providing multidrug-resistant tuberculosis care in Bantul District, Yogyakarta Province, Indonesia

Introduction Patient-centered care approach in multidrug-resistant tuberculosis care requires health worker safety that covers both being safe and feeling safe to conduct the services. Stigma has been argued as a barrier to patient-centered care. However, there has been relatively little research addressing the issues of safety and stigma among health staff. This paper explored the issue of being safe, feeling safe, and stigmatizing attitude among health staff working with multidrug-resistant tuberculosis cases in primary health care facilities in Indonesia. Methods Using a mixed methods research design, data was collected with structured questionnaires among 123 staff, observations of infection control in 17 primary health care facilities, and in-depth interviews among 22 staff. Results The findings showed suboptimal infection control infrastructures for the primary health care facilities. The knowledge and motivation to follow multidrug-resistant tuberculosis care protocols are suboptimal. Feeling unsafe is related to stigmatizing attitude in providing multidrug-resistant tuberculosis care. Conclusion Being safe, feeling unsafe, and stigmatizing attitude are challenges in providing patient-centered multidrug-resistant tuberculosis care in primary health care facilities in Indonesia. Serious efforts are needed on all levels to ensure safety and prevent irrational stigma.


Introduction
A patient-centered care approach has been considered as an essential pillar for tackling tuberculosis (TB) global epidemic [1]. Improving a patient-centered care approach, however, cannot be implemented effectively without simultaneously managing the safety of health workers, and it has been recognized that the safety of health workers is an imperative that should receive more serious attention [2].
Health worker safety, however, is not just a matter of being safe. Health workers must also feel safe. In the absence of this broader health worker safety, there is a danger that patient-centered care will be jeopardized because of stigmatizing attitudes toward TB patients by unsafe health workers. Indeed, there is a significant body of literature looking at stigmatizing attitudes among health staff working in TB care, and the adverse consequences on care [3][4][5].
Indonesia is one of 27 countries with a high burden of multidrug-resistant tuberculosis (MDR-TB) globally [6]. Notwithstanding the burden of MDR-TB in Indonesia and the known impact of stigma on the provision of patient-centered care, there has been a limited number of research addressing the issue. This is particularly unfortunate because it is well known that levels of stigma, as a social construct, vary with context [5,7].
The Indonesian Ministry of Health has promoted direct observed treatment (DOT) [8,9] as the strategy in the management of TB. Since 2014, there has been a concerted effort to identify the primary health care facility (PHC) closest to MDR-TB patient's address as the center of care. A primary health care facility should provide treatment management for MDR-TB patients at the continuation phase.
Bantul district, a district in Yogyakarta Province of Indonesia, adopted the policy of DOT for MDR-TB at PHC settings in 2014. However, in 2016, there was the rejection of staff at PHCs to conduct DOT service for MDR-TB. This paper explored the issue of safety, feeling safe, and stigmatizing attitude among health staff working with MDR-TB cases in PHCs in Bantul district, Yogyakarta Province, Indonesia.

Research settings
Bantul district is one of five districts in Yogyakarta Province of Indonesia, with 971,000 inhabitants. The population density is nearly 2000 per square kilometer. There are 27 PHCs located in 17 sub-districts of Bantul [10]. Seven of 27 PHCs have implemented DOT. The number of presumptive MDR-TB cases has increased in the last 3 years, i.e., 31 cases in 2013, 41 cases in 2014, and 45 cases in 2015 (District Health Office of Bantul, unpublished data).

Study design
We applied a concurrent nested mixed methods design [11] that combined a health care facility-based cross-sectional study and a qualitative study. The cross-sectional study was conducted to assess the health staff's knowledge about infection control and their attitude to conduct proper care protocols for MDR-TB. Through the survey, we would like to describe whether health staff 's knowledge about infection control and the attitude to conduct DOT for MDR-TB were lacking. The cross-sectional study is nested in the qualitative observations of infection control infrastructure and staff behavior. The observation aimed to assess the infrastructure and activities related to infection control. We also conducted in-depth interviews to explain the findings from the cross-sectional study and observation, and to explore about reasons behind irrational stigma behavior.

Research participants and sampling
For the survey, we selected staff that met the inclusion criteria in all 27 PHCs: health staff involved in implementing protocols for MDR-TB care (registration staff, medical doctors, nurses, lab staff ) and infection control (staff responsible for health promotion and environmental health). Sample size calculation was done based on the formula for a descriptive cross-sectional study [12]. By estimating that 50% of staff among total 180 population of relevant health staff at PHCs (Bantul district health office 2015, unpublished data) were knowledgeable about the MDR-TB care procedure and infection control, 95% confidence level, and precision 10%, we calculated a minimum sample size of 123 individuals for our study. Data on eligible staff were obtained from the District Health Office. We selected the staff by quota sampling until we reached the minimum sample size.
For the in-depth interviews, we selected health staff at PHCs relevant to MDR-TB care by purposive sampling (medical doctor, nurses, lab staff, and TB program staff ) for in-depth interviews. We also conducted observations of infection control facilities with a TB infection control consultant to assess the adequacy of infrastructure and procedures of MDR-TB care in PHCs. Both in-depth interviews and observations were conducted until data saturation was achieved.

Data collection
Two trained surveyors collected data from surveys among PHC staff under the supervision of the last author. The first and second authors mainly conducted both in-depth interviews and observations of staff behavior in conducting DOT for MDR-TB patients. An infection control expert conducted the observation of infection control infrastructure and proper MDR-TB protocols.
Structured questionnaires about knowledge on infection control and observation guidelines were developed based on the national guideline for infection control related to TB control programs at the health care facility level [13]. The questionnaires and the observation guidelines were piloted and validated. The content validity of the quantitative questionnaires and guideline of in-depth interviews were assessed through the panel of experts during the proposal finalization workshop. We piloted the quantitative questionnaires to 15 primary health care staff to assess their clarity to each question. The observation checklist of infection control was validated by the expert from Universitas Gadjah Mada, Indonesia.

Data analysis
Qualitative data were analyzed using the content analysis technique [14,15]. From the text of transcripts, we identified phrases/sentences containing specific meanings (meaning units). The manifest meanings and latent meanings were generated from the meaning units, which were assigned as codes. The codes with similarity were grouped into categories. Open Code software [16] was used for facilitating the data analysis. To improve the trustworthiness of the qualitative data in the data analysis, we conducted peer-debriefing in the researcher team.
Quantitative data were used for triangulation of the qualitative data in the mixed methods design. Quantitative data were analyzed by comparing the knowledge and attitude of staff by their characteristics.

Research ethics
Ethical approval was obtained from the Ethics Committee, Faculty of Medicine, Universitas Sebelas Maret at Surakarta Indonesia. We obtained written informed consent from all study participants. The researchers conducted the informed consent for the qualitative data collection, while two enumerators conducted the informed consent for the quantitative data collection. The enumerator received training from co-PIs to conduct the informed consent process and collecting data.

Characteristics of study participants
A total of 22 in-depth interviews were conducted in 17 PHCs, with 19 in-depth interviews with clinical staff (7 physicians, 1 dentist, 9 nurses, and 2 lab staff) in 17 PHCs, and key informant interviews among three TB program managers at the district/provincial level. Ten out of 19 clinical staff were also acted as TB program staff.
A total of 123 health staff in all 27 PHCs were included in the cross-sectional study ( Table 1). The majority of the study informants were clinical staff (medical doctors, nurses, lab staff ). On average, the study informants were in the productive age (39 ± 12 years old).
Most of them were female (68%), with the level of education of 3 years Diploma. About 90% of health staff has been working for more than 5 years.

Safety
The observation in 17 PHCs showed suboptimal infrastructure and fidelity of activities to infection control protocols as per the national guideline (Table 2). Proper layout and zoning were only available in eight out of the 17 PHCs; unfortunately, only one PHC provided DOT for MDR-TB had optimal zoning. In addition to zoning, the suboptimal infrastructure of infection control was shown for other types of infrastructures including exhaust fan, infectious disposal system, and sputum collection area. On the other hand, the availability of infectious waste disposal and surgical masks for patients with a cough were adequate.
The observations also revealed unsafe practices such as health staff's unawareness about the direction of airflow while they were providing DOT. The health staff even stood in the opposite direction of airflow. The observation also found the health staff unknowingly allowed the MDR-TB patients to be without surgery masks when the patients were having cough during the DOT.
The cross-sectional study revealed that the knowledge of health staff about infection control protocols was inadequate in some domains (Table 3). Health staff had relatively sufficient knowledge about the disposal of tissues used to close mouth when coughing (84%).
Length of work, n (%) However, their knowledge about other domains of infection control was still suboptimal. Knowledge about timing to provide cough etiquette were higher among clinical and TB program staff compared to other types of staff (registration, environmental health, health promotion staff ). The same pattern was found in terms of setting the infrastructure such as a fan for infection control purpose. Additionally, the quantitative data also showed a low proportion of staff with adequate knowledge of infection Air ventilation in laboratory 9 4 Specific sputum collection area 6 5 Available hand scrub liquid 10 5 Available and utilized surgical masks for patients 15 6 Optimal activities Triage 14 4 Education 8 3

Separation 10 3
PHCs primary health care facilities, DOT direct observed treatment, MDR-TB multidrug-resistant tuberculosis  control in two domains: cough screening among visitors (38%) and use of a respiratory mask (48%) ( Table 3). Knowledge about cough screening was lower among staff who were not directly involved in the process of DOT for MDR-TB patients (e.g., registration staff, health promotion, or environmental health staff ) compared to clinical or TB program staff (p value = 0.008). There was no statistically significant difference between staff working at PHCs with DOT for MDR-TB service and those working in PHCs without DOT for MDR-TB service in terms of knowledge of staff about cough (p value = 0.120) and the use of a respiratory mask (p value = 0.705). We also found a similar pattern when we compared between staff who received training on DOT for MDR-TB and infection control with those who did not.

Feeling safe
Our quantitative data showed that approximately half of the health staff stated about feeling stressed and fearful in conducting MDR-TB care (Table 4). Staff at a PHC providing DOT for MDR-TB service perceived less burden of workload compared to the ones at a PHC without the DOT service (56% vs. 32%, p value = 0.011). However, there was a relatively similar proportion of staff with perceived fear to be infected or feeling afraid to talk with MDR-TB patients by the type of staff, working/not working at PHCs with DOT, and history of trainings on DOT or infection control.
The findings from the cross-sectional survey also confirmed a lack of confidence and perceived capacity to provide MDR-TB care in the PHCs (Table 5). Only 65% of health staff felt confident in conducting DOT for MDR-TB patients. Less than 60% of health staff reported their understanding of DOT procedures. Similarly, only 53% of total staff stated that infection control is properly managed in the PHCs. Clinical and TB program staff showed lower confidence in the infection control in PHCs compared to other staff (p value = 0.020).
Our study also showed the lack of staff motivation to conduct MDR-TB care. Only 67% showed their commitment to providing DOT in accordance to proper protocols for MDR-TB patients. Notably, 26% perceived their preference not to provide proper care, when it was considered a choice ( Table 5).
The observation showed inefficiency in the use of the infection control infrastructure of infection control in our observations. For illustration, in a DOT room outside the PHC building that had much airflow, there was an ordinary fan and exhaust fan had been set up. Both fans were used when DOT was provided.
The feeling of being unsafe was also reflected by unnecessary practices by health staff such as wearing double hand gloves, waterproofed gown, plastic boots, and double masks. Those practices reflected efforts to have maximum protection. As an illustration, a nurse at the PHC stated, "I am afraid of being infected… [I know that] the PHC staff who conducts the DOT wear Table 4 Distribution of primary health care staff who stated perceived stress and fear to conduct multidrug-resistant tuberculosis care  Feeling safe or unsafe to conduct DOT for MDR-TB patients was an interaction between four domains: fear, perceived risk, security, and adequate training (Table 6). Fear was perceived as a health staff 's natural expression when taking care of patients with MDR-TB. However, the perceived risk of being infected varied: We are at high risk of getting infected, however, there is no certainty on our insurance if we are infected (Health staff 9, a nurse) Masks have been provided at the registration desk. There were TB patients here, but health staffs did not wear it. Moreover, there will be an MDR TB patient here (Health staff 13, a nurse who was also a TB program staff ) The awareness about the risk of being infected increased when the PHC received an MDR-TB patient for DOT, as expressed by an informant, "They (health staffs) were shocked when they found out there was an MDR-TB patient here" (Health staff 16, a nurse who was also a TB program staff ).
Perceived risk and lacking security among the staff prompted the domination of fear in their response in regard to DOT for MDR-TB patients. While training was considered as an intervention to reduce the fear, infection control capacity in the PHC was essential. Training only was perceived could not overcome fear.
…Our intention is to help patients, and we have no choice. We are sure we will be safe as long as our infection control is adequate. (Health staff 11, a nurse) We provided training to our staff, but it could not overcome fear. (Health staff 6, a physician) Information from the in-depth interviews showed that the coverage and quality of training were perceived inadequate by some health staff at PHCs. The survey data showed that only 29% of health staff received the training on DOT for the MDR-TB procedure. Our confirmation with the TB program managers revealed that the training was conducted in a working day at the PHC as a preparation to receive a back referred patient from the MDR-TB center hospital.   Feeling safe/unsafe as an interaction between fear, perceived risk and security, as well as level of knowledge.
"Health staffs were scared to get infected. Fear is human, is not it?" (Health staff 1)

Fear is human Fear is a natural expression
Health staff were scared to get infected -Perceived risk "They (health staffs) were shocked when they found out there was an MDR-TB patient here" (Health staff 16) The health staff were shocked when they found out there was an MDR-TB patient at the PHC -"Masks have been provided at the registration desk. There were TB patients here, but health staffs did not wear it.
Moreover, there will be MDR-TB patient here" (Health staff 13) The health staff (at the registration desk) did not wear mask, while there will be MDR-TB patient (at the PHC) Awareness about risk of being infected is low "We are at high risk of getting infected, however there is no certainty on our insurance if we are infected" (health staff 9) No certainty on our insurance if we are infected Lack security for staff in doing DOT for MDR-TB patients.
Perceived security "They are scared to death. There is no insurance or guarantee of safety from the management" (Health staff 5) There is no insurance or guarantee of safety from the management "…Our intention is to help patient, and we have no choice. We are sure we will be safe as long as our infection control is adequate" (Health staff 11) We will be safe as long as our infection control is adequate The infection control is perceived inadequate "We will not get infected if the Infection Control is adequate" (Health staff 11) The health staff will not get infected if the infection control is adequate Infection control application is perceived essential to prevent MDR-TB transmission "First of all, they were afraid to get infected and they are questioning whether the Infection Control has been adequate yet" (Health staff 10) Health staff are questioning whether the infection control has been adequate yet The infection control is perceived inadequate "It will be useless to give training to the health staff. If they are scared, it's personal…" (Health staff 2) It will be useless to give training to give training

Stigmatic attitude
The stigmatic attitude was also shown in our observation. Some health staff kept unnecessary distances from the patients when they conducted DOT for MDR-TB care. Even a health staff was talking with the patient from a small window while the patient was in the DOT room and also locked the door of the DOT room. The in-depth interview data revealed that fear was a driver to a stigmatic attitude to MDR-TB patients.

Discussions
Our study highlighted existing issues about being safe, feeling safe, and stigmatic attitudes among health staff in providing MDR-TB care. Some infrastructures of infection control in PHCs are suboptimal. There was also lacking knowledge of infection control protocols. The stigmatic attitudes manifest in various forms: fear of being infected, avoidance to conduct care, and performing unnecessary overprotected practices of infection control. The cause of the stigmatizing attitude was the fear of being infected. The combination of the suboptimal infection control environment, lacking knowledge, and stigma resulted in unsafe and inefficient practice while doing the MDR-TB care, as well as the discrimination of MDR-TB patients. Our findings on the inadequacy knowledge and practices of TB infection control such as patients' screening corroborate previous findings by studies in the hospital setting in South Africa [17,18]. Low level of knowledge was also worse among administrative staff and other staff who do not have direct relevance to MDR-TB care [19]. Other previous studies in India also found problems with the infection control infrastructure related to drug-resistant TB [20,21].
The fear among health staff due to MDR-TB transmission risk, while they provide care, is the cause of stigmatizing attitude, as was also found in the previous research studies [22][23][24]. Stigma among health staff is hypothesized to influence the practice of infection control [23]. Our data revealed empirical evidence that the fear even induced the unsafe and inefficient method of infection control by wearing binary masks (respiratory and surgical masks) and hand gloves. Our study also provided empirical evidence about the influence of fear in the communication between provider and patient that was also proposed by others [22].

Public health implications
The state of being safe is a priority issue that should be addressed by improving the quality of infection control. Our finding showed a need for knowledge and capacity building for infection control. We argue that this intervention will be a foundation to influence the feeling of safety among health workers to conduct MDR-TB care.
Our study added the gap of knowledge about whether the stigma of health staff to conduct TB care is/is not reasonable. The internalized and enacted stigma among health staff is found related to feeling unsafe due to a lacking infection control infrastructure and knowledge. Improved supervision of MDR-TB care should be performed with the continuation of its implementation.
The findings of stigmatic stigma signified the need for improvement of the quality of care provided by health staff at PHCs. Considering that patient-centered care is one of the pillars of the strategy to end TB [1], we agreed with others that the improvement of the quality of MDR-TB care should include empathy and respect for the patients [24].
Interventions to reduce stigma concerning TB among health care providers are limited [25]. Wu et al.'s study put an effort to decrease internalized stigma among health care workers by training to improve knowledge and attitude among health staff; however, it has been shown less effective among health staff conducting DOT [26]. Hence, more research is needed to get pieces of evidence about the effective strategies to reduce stigma among health staff providing MDR-TB care.

Study limitation and strength
Our study aimed to explore the phenomenon related to the rejection of the staff to conduct MDR-TB direct observed treatment cases. The study design could not assess any causal relationship between issues of safety, feeling safe, and stigmatizing attitude. On the other hand, the findings from the qualitative data could enrich the interpretation of the quantitative data. Therefore, we could have a holistic portrait of the phenomenon.
This study explored the phenomena of safety and stigmatic attitude with qualitative study which was limited in the generalization in other settings. In future, stigma should be measured quantitatively using a locally adapted instrument. The measurement will be useful to Fear is a driver to a stigmatic attitude to MDR-TB patients Stigmatic attitude Fear as a driver of stigmatic attitude Fear support the evidence related to the level of stigma. Since we argue with the urged need to reduce stigma, the quantitative measurement of stigma among health staff will support the evaluation of the effect from the stigma reduction initiative.
Our study added the gap of knowledge about whether the stigma of health staff to conduct TB care is/is not reasonable in the Indonesian context. We found that the internalized and enacted stigma among health staff is found related to feeling unsafe due to a lacking infection control infrastructure and knowledge. Improved supervision of MDR-TB care should be performed with the continuation of its implementation. The findings from our study contributed to current efforts to understand the nature of stigma among health staff in conducting MDR-TB care that is argued as an initial step to address and tacking the implementation issue of DOT for MDR-TB patients [27]. The way forward to patient-centered MDR-TB care would be jeopardized unless stigma among health staff is diminished.

Conclusion
Our study concluded with a clear demonstration of lacking safety, feeling unsafe, and stigmatic attitude among health staff to conduct patient-centered MDR-TB care in PHCs in Indonesia. Serious efforts are needed on all levels to ensure safety and prevent irrational stigma.