Operational Challenges in Conducting a Subnational TB Prevalence Survey in India: Lessons Learned for Resource-Limited, High-Burden Settings

Ethical Approval

The study protocol (017/NIRT-IEC/2021) was approved by the Institutional Human Ethics Committee of the National Institute for Research in Tuberculosis.

Data Collection

The experiences and challenges in implementing the TB prevalence survey were collated into a report that gleaned information from periodic meetings between the central team and site-level staff, letters of communication between various stakeholders, and informal discussions with site staff and collaborators. In addition, details from different departments related to procurements made, services requested, and services monitored were collected. The information was reviewed and broadly organized based on common themes. We describe the challenges in detail and the strategies used to overcome them in the following section.

Human Resources Training and Management

The survey involved around 150 staff for this project. Medical officers and pulmonologists were identified and trained at the district level for X-ray reading. There was a high turnover of trained staff as the survey required staff to continuously travel to various places throughout the state. Similarly, many medical officers who were trained were transferred to other places as part of the routine department procedures. Hence, in addition to the refresher training given to the teams at regular intervals, we had to repeatedly provide training whenever a new person joined the team.

To address these challenges, refresher trainings were done in person and through video conferencing at regular intervals to ensure good performance in survey conduction. Assessment modules were used to measure the effectiveness of the repetitive trainings. All staff were aware of all procedures so that even during the COVID-19 crisis, when many staff had fallen sick, the team was competent to manage the survey activities and perform the additional responsibilities for their sick colleagues.

COVID-19-Related Disruptions

The survey was halted for 2–3 months due to the COVID-19 lockdown. Many staff became sick from COVID-19 infection, and due to the risk of illness, a few were hesitant to rejoin and left the job. Coordinating with the public health department helped mobilize the required personal protective equipment for staff. The national COVID-19 vaccine campaign gave high priority to health care workers, and all survey staff were immediately vaccinated when the vaccine was launched.

Due to the COVID-19 restrictions, there were challenges in the cluster coverage, especially in urban clusters. Many of the households were locked, and many refused to include their household in the study. Participants were hesitant to reveal any symptoms due to fear of being quarantined due to COVID-19. Local community leaders were engaged to combat participants’ fears and concerns, and staff were trained to speak sensitively regarding COVID-19 issues.

Survey Participation and Community Engagement

In urban areas, there was low participation because many people had to travel to various places for work and were not available during the survey. Some of them had even migrated to their native villages during the COVID-19 pandemic, as they had an option to work remotely. In rural and remote areas, participants had difficulties traveling to the main survey site from their houses. Survey participation increased by extending the hours of activities, providing transport for those participants living far from the survey site, providing refreshments for participants, and offering other health services (e.g., free blood sugar check-ups).

In some clusters, there were issues in which participants from 1 street refused to go to another street where the survey camp was placed. This resulted in changing the survey site many times to enable the participation of all community members. In certain clusters, participants, especially women, were reluctant to participate in the survey without getting concurrence from their husbands and religious heads.

To help increase community engagement, at each cluster, the district TB officer and local community leaders held an inauguration for the survey. In certain clusters, religious leaders were approached to motivate the community to participate. The slow rate of cluster coverage was tackled by having strong team motivation, constant supervision and monitoring at the cluster site by the central team, and repeated trainings on communication and community engagement strategies.

X-Ray Machine Breakdowns

Having an operational X-ray vehicle was crucial and a rate-limiting factor that required immediate attention. To stay on track with the survey timelines, we had to take a minimum of 100 X-rays each day per team. Due to the long hours needed to conduct the survey, we faced several breakdowns of X-ray machines and generators that were used for the survey. Thus, the survey had to be suspended temporarily until the X-ray unit was repaired. Because the outdoor temperatures were 37–40 degrees Celsius, X-ray machines and generators often overheated despite having an air conditioner in the X-ray van. Backup machines were available when breakdowns occurred. Having a comprehensive maintenance contract with the supplier with regular maintenance reduced the time that equipment could not be used.

Laboratory Issues

The sputum samples had to be transported and tested through the routine TB program in the state. We faced delays in samples being delivered to the reference laboratories by parcel services, sometimes more than 48 hours. This was a major challenge, especially in the summer, because it was difficult for the cold chain to be maintained, resulting in sample liquefaction and a high risk of contamination. When unavoidable delays in sample transport were anticipated, care was taken to arrange for the storage of samples in nearby hospitals with the cooperation of the local public health department.

In certain clusters, the number of participants eligible for sputum testing by X-ray was more than expected, posing a challenge because of the limited availability of liquid culture tubes in the middle of the survey. Until the supply of tubes was restored, we switched to solid culture for a few weeks to avoid a sample backlog for fear of contamination.

When there was a shortage of consumables, strong leadership among all the stakeholders helped mobilize resources from various sources without affecting the flow of the survey.

Delays in X-Ray Reading

During the COVID-19 pandemic, district TB officers were assigned COVID-19-related activities, and pulmonologists/radiologists were assigned COVID-19 duties as well as outpatient duties, so it was challenging for them to report X-rays on the same day or the next day. Despite these challenges, we successfully completed all X-ray readings within 48 to 72 hours. Due to the delay in X-ray reporting by a few days, the survey team faced many challenges with sputum collection because some participants had traveled out of town.

To address this challenge, digital X-rays were uploaded to the central server where staff could read the images from any place at any time.

Data Management

There were network issues during natural calamities, and in some rural areas, the network signal was very poor. The Internet connection with the devices used by enumerators had issues, as the routers in the van had poor signal strength. Hence, staff came to the survey site often to sync data, resulting in delays, especially in rural areas where households were situated far apart. In some rural areas of the state, some team members traveled to nearby places that had good network coverage so that they could sync the data to the central server daily. Daily data cleaning and checking and daily communication with the team ensured high-quality data collection in a timely manner.