Recruitment capability
The review of the DHS data(25, 26) in Cameroon identified 16.9% children <5 years per household and a mean of 5 household members, which would represent 0.85 children <5 years per household. In Uganda there are 18% children <5 years per household, with a mean of 4.5 household members, which corresponds to 0.81 children <5 years per household. Yuen et al. has looked into the household composition of TB patients for all countries and estimated 0.83 (95% confidence interval (CI) 0.80-0.86) children <5 years per household in Cameroon and 0.93 (0.89-0.96) in Uganda(27). Therefore, the assessment of the cluster facilities’ capacity to enroll child contacts <5 was made using data on the number of index cases from TB registers, assuming there was one index case per household and one child contact <5 years per index case.
Figure 1. Retrospective data of tuberculosis bacteriologicaly confirmed cases in Cameroon and Uganda.
From Figure 1 we observe that the clusters facilities in Cameroon had enough TB patients to meet the minimum number of 50 TB B+ patients per cluster per year to reach the study sample size within 12 months, except for one cluster (cluster 10). In Uganda there were 3 clusters that did not meet the minimum of 50 bacteriologically positive TB patients (cluster 1, cluster 3 and cluster 7). Despite some fluctuations, data were consistent between the 2017 NTP reports and data collected retrospectively from TB registers between April 2018 and March 2019.
Acceptability
The team conducted 11 FGD with 42 men and 32 women. The mean FGD duration was 105 minutes for women and 128.5 minutes for men. One FGD with women in the Littoral region of Cameroon was not done as the required minimum number of 6 participants was not reached. Twenty-four IDI were conducted with providers and community leaders. (Additional Table 3). We discussed contextual and perceived barriers to facility-based TB child contact management, perceived benefits of a community intervention, and prerequisites for its implementation.
Barriers to facility-based TB screening
All TFP in both countries stated they ask TB patients to bring back their children to the health facility for TB screening as requested by national TB guidelines. In their overall assessment based on experience, patients’ adherence to such a request is poor. During the FGDs, some TB patients confirmed they were indeed requested to bring their children to the facility for TB screening, among them, few did comply. Others declared that they were never asked to bring their children to the health facility for screening, stating they might have done so if requested.
The main reasons cited or anticipated by TB patients for not bringing their children to the health facility were financial, sociocultural or stigma-related. As illustrated below, facility-based child contact TB screening carries financial costs that make child-contact screening less financially and geographically accessible, especially transport costs for those living far away from the facility and those who have many children <5 years to bring for screening and more generally, the poor. This is corroborated by health personnel who did understand parents may face challenges when they cannot afford the transport cost and other costs such as buying food for children while waiting for the screening procedures.
“I need to get transport to transport about 8 people to come back and forth. For starters coming here and going back home alone, they take me for 3000 shillings. So I have to spend 3000 for each of the eight people to come here at the health facility and also spend 3000 shillings to transport them back, so transport would strain me” – Male participant, Uganda
“When you ask a parent to transport his 5 kids and bring them to the health center, he’s going to ask: “Are you paying for transport?”” – CHW, Cameroon
While these practical and objective facts played a role in parents’ non adherence to healthcare providers’ recommendations, some FGD participants believed non-adherence was rather a form of child neglect and pointed at the sociocultural aspects such as gender roles in childcare religion that prevent men from attending health centers when a child is sick. Traditional beliefs about diseases and TB causation (witchcraft) associated with mistrust in “modern medicine” or confidence in traditional medicine or religious prayers, negatively and strongly influence help- and health-seeking behaviors and trajectories, as well as treatment itineraries.
“You know, African families really like going to healers in the neighborhood. He [the TB patient] will look for herbs before going to the hospital. We, African families, love the healers.” – male participant - Cameroon
“There are some people who think that they have been bewitched especially when that person has a persistent cough. The person may even start to blame the neighbor whom they had a quarrel with for being responsible for the cough” – Community leader, Uganda
Whether associated to HIV or a stand-alone determinant of TB non-disclosure, TB stigma was ever-present in all FGDs and emerged as a barrier to TB health facility-based screening. Overt stigma was of particular concern and played an important role in how patients live with the disease.
“People fear to disclose that they have TB because they might lose their jobs, they might lose their relationships; it may cause people to be isolated” – male participant, Uganda
“They think this [TB] is a shameful disease and that people will mock them” – male participant, Cameroon
“After I was diagnosed with TB, my husband threatened me “I do not want to hear anyone discussing this with anyone.” And I have never told anyone […] and it hurts me inside” – female participant, Uganda
Healthcare providers shared similar views regarding financial shortcomings, stigma and sociocultural norms. In addition, TFP highlighted the centrality of the initial encounter (or counseling) with the index case in helping patients understand the importance of TB prevention, including the need for child-contact TB screening and TPT. Indeed, provision of TB literacy and the rapport that is built during this initial visit are essential for a good follow-up during treatment. When such initial visit is rushed due to a high workload and patients queuing at the health center, the information is not passed to the patient and in consequence, this important step is overlooked. As explained in one instance by a doctor and facility manager, health worker’s priority is to cure patients who present themselves at the facility, not to manage the contacts.
“They [TB patients] are not coming back with the children not because they don’t want to, but because maybe they did not understand an important part [of the health education]” – CHW, Cameroon
“The health facility is overwhelmed by the services here. You get here and you find many people here and that means that on that day you will not be attended to and might need to return the following day. That means that you will have lost two days of work, the child has also missed school. And you know many of us earn an income by the day” – male participant, Uganda
Conditions for acceptability of a community intervention
From the patients’ perspectives the proposed intervention (19) was acceptable and made sense in both countries, as it will be helpful to many in overcoming the main barriers to facility-based child-contact TB screening, and in particular, transport costs that many TB patients cannot afford.
Besides removing distance and related transport costs, patients noted further benefits of the household visit, including the confirmation of the child’s good health (not TB infected) and ensuring through TPT that a parent’s TB infection will not be passed to the children.
“I would accept because I had it [TB]… and I need to make sure my children are healthy” – female patient, Cameroon
Additionally, parents get the opportunity to address other health or environmental problems with the CHW that might come up during the home visit discussions.
“Many people cannot afford to go to the health facility. When you go to the home, you can teach many things and they are able to know and understand better rather than spend the whole day at the health facility with a child in the back” – CHW, Uganda
All participants welcomed the community intervention. Only one male TB patient participant in Cameroon stated he preferred taking his children to the health center for TB screening because all investigations are available there as opposed to the limited knowledge or diagnostic means in the community intervention model. Indeed, the discourses of some other participants showed the obstacles they went through from district hospitals to the capital city on their diagnosis itinerary.
Elsewhere, many participants raised concerns about unintended disclosure and subsequent stigma from the other community members following home visits. The proposed approach to home visit did not include enquiries and screening of other children <5 years living in the same compound, participants were ambivalent when asked about the opportunity to also screen such children as the risk of stigma increases. Those who would agree to the screening of children playing together with their own children, preferred to inform their parents themselves based on past positive relationship and their living together experience.
“They [the family of the index case] need to check if the neighbors are not sick. It will be difficult, but if the family has accepted their fate [having TB], they can help the others [the neighbors] accept as well” – CHW, Cameroon
“Some clients do not want their neighbors to know because they know that the moment these know that they are having TB, they will either be chased from their place where they are renting or sometimes they may be isolated” – TB focal person, Uganda
From the providers’ points of view, the intervention was coherent and welcome though they questioned its sustainability.
One CHW even highlighted the fact that many research projects test interventions in the communities and when they finish the project and remove the means, there is no benefit left for the community:
“You [implementing organizations] come, you tell us what has to be done, you teach us what to do, it [the project] starts well and after a certain time it stops. And we don’t understand why it stopped.” – CHW, Cameroon.
Prerequisites of feasibility of community TB screening and TPT management
Both patients, community leaders and health staff agreed that the cornerstone of this community intervention is the explanation given and the counseling offered by the TFP at the first visit with the index case. During this visit, TB education should be done, rapport should be created through demonstrating empathy, providing options, and ensuring confidentiality.
“During the first visit is when the rapport is created. Once the patient gets to know that you are friendly and you will keep their information, you will not release it to any other person; through my experience, these clients are willing to welcome you to their homes” – TB focal person, Uganda
This was echoed by healthcare personnel, especially in Uganda, where TFP and CHWs stressed successful implementation depends on thorough information given to the index case. Of paramount importance is the quality of the initial counseling, and therefore on CHWs’ training, experience and acquired legitimacy.
“I think that if we explain very well to the patient, it will be acceptable. […] when it is well explained, the patient understands the benefit [of the community intervention] from the explanation” – health facility manager, Cameroon
“At the beginning I told you, if you give them information on the initiation day, if you give them the information, they can give the medicine [to their children].” – TB focal person, Uganda
The opinions of the patients converged towards their need to be informed well in advance of the timing of the home visit so that they can ensure that their children are at home at the time of the visit or that they have informed their partner -if they have not yet done so- or relevant people based on need-to-know. These conditions provided, the team is welcome to come and perform the screening and even take time to discuss other family health needs or concerns.
Creating rapport was also essential for CHW and they should be trained on this subject. The legitimacy and training of a CHW was also an important determinant of feasibility according to patients and stakeholders. Some respondents and interviewees were skeptical regarding the implication of CHWs because of past negative experiences with careless CHWs who would spread information about people they visit, which shows a lack of confidentiality.
Generally, FGD participants preferred trained CHWs who are polite and explain well all activities that will take place. There was no preference for gender, as long as the person is well trained.
“The most important thing is if you send a health worker with the expertise in whatever he or she is going to do. It does not matter if it is a doctor or a nurse, as long as they have the expertise in whatever they are going to do. It does not matter whether the health worker is male or female” – male participant, Uganda
Even with the best intentions in mind, unintentional disclosure cannot be totally ruled out due to the specific local context, and some participants expressed their concern. Others were not at all worried about disclosure and subsequent consequences.
“They [the neighbors] will maybe spread this [the information] everywhere. “Oh, she has this, she has that”. It’s difficult, it’s very difficult [to disclose]” – female participant, Cameroon
“For me it’s not a problem. It’s my house, my family” – female participant, Cameroon
An essential point discussed only by the health staff and community leaders is the CHWs’ motivation. This is a term historically referring to money used to compensate CHWs for their time and transport. Motivation is always requested when doing any kind of research activity in both countries as highlighted by the qualitative assessment:
“If we have enough staff and there are [financial] resources, it [TB screening] can be improved” – community leader, Uganda
“If there is motivation, they [the CHW] will do the work” – community leader, Cameroon
Adaptation and integration
TB services and available tools
In all cluster facilities contact investigation was done by a health care worker (nurse, clinician), for children <5 years in both countries. HIV-positive contact children were also screened in Uganda. Both countries had registers recording child contacts initiated on TPT and thir TPT outcome. At the time of assessment, National TB Programs were about to introduce in both countries a contact screening register to record all household contacts per index cases with the results of their TB screening None of the two countries had tools to monitor TPT adherence and tolerability. In Uganda, TB contact screening could be done at community level by the facility TFP. In practice, this activity was done only with support from implementing partners to cover transport cost. At the time of the site assessment, no registered data were available about the number nor age of household child contacts in both countries.
Six months isoniazid prophylaxis was used in all facilities and was delivered monthly at the facility by the TFP. All study facilities were expecting to introduce the 3 months isoniazid rifampicin (3RH) TPT under the CaP TB Project. TB screening, clinical and microbiological diagnosis for children with presumptive TB and drugs and treatment monitoring were free of charge. Families had to pay for further TB investigations like chest X-ray. Drug-resistant cases and complicated cases were referred to higher-level health facilities.
HIV testing was provided at the health facility in all study sites, in close collaboration but in separate units of the same department in Cameroon with the exception of two clusters facilities where TB and HIV services were fully integrated, and integrated in the same department in Uganda.
In all study sites TB drugs were stored at the TB clinic at room temperature in a locked cabinet and in two health facilities at the facility pharmacy.
A reference and counter-reference system between the CHW and the PHC staff or higher-level health facilities was set in the Ugandan clusters but almost inexistant or not functional in the Cameroon clusters.
Table 2 below summarizes practices and available tools under the standard of care in the two countries.
Table 2
Practices and tools in the routine system
Activity | Cameroon | Uganda |
Index case identification | By the TFP at the health facility using the TB register | By the TFP at the health facility using the TB register |
Contact investigation | At the health facility. Contact tracing register about to be introduced | Possibility of household contact investigation by the TFP Contact register about to be introduced |
Symptom screening | At the health facility, no tool | Possibility of household screening, intensified case finding tool (checklist) |
HIV testing of child contacts | Only medical personnel at the health facility | Possibility of HIV testing by CHWs or healthcare staff |
TPT initiation | 6H, at the health facility, recorded in the TPT register by the TFP | 6H, at the health facility, recorded in the TPT register by the TFP |
TPT follow-up: adherence and tolerability | Adherence and tolerability not assessed. No tool for TPT adherence. TPT register used for follow-up at the health facility | Adherence and tolerability not assessed. No tool for TPT adherence. TPT register used for follow-up at the health facility |
Safety management | At facility. No tool for safety evaluation | At facility. No tool for safety evaluation |
PT outcome assessment | According to national TB guideline: completed, death, lost to follow-up. At the health facility by the TFP | According to national TB guideline: completed, death, lost to follow-up. At he health facility by the TFP |
TB diagnosis | TB investigations at the health facility or referral at a higher-level facility Available tools: CXR, sputum collection, NPA, XpertMTB/RIF testing Laboratory results in the lab register | TB investigations at the health facility or referral at a higher-level facility Available tools: CXR, sputum collection, XpertMTB/RIF testing. Laboratory data collected in the lab register |
TFP = tuberculosis focal person, CXR = chest radiography, NPA = nasopharyngeal aspirate, H = isoniazid; TB = tuberculosis |
Checking data quality
A total of 1091 TB patients, out of which 708 were B+, have been registered between May 1st, 2018, and Oct 31st, 2018, in the TB registers of the cluster sites. The overall median rate of missing data was 0.3% (interquartile range (IQR) [0%-3%]) in Cameroon, ranging from 0 to 8.6% and 0.4% (IQR [0%-0.6%]) in Uganda, ranging from 0 to 1.4%. The median error rate was 1.1% (IQR [0.6%-1.4%]) in Cameroon, ranging from 0.3–3.6% and 0.0% (IQR [0%-0%]) in Uganda. The biggest rate of missing data was for the registration date, with a maximum of 8.2% in cluster 6. The biggest rate of errors was for the type of TB, with a maximum of 2.1% in cluster 5. (see Additional Tables 1 and 2).
Resources and procedures for the community intervention
The type of human resources at facility level was similar in the two countries. The community intervention involved mainly TFP and in addition, one clinician was identified as a safety monitor for referred children with TPT tolerability concerns and was trained for safety assessment. Regarding CHWs, in Uganda, village health team were already involved in TB activities at facility level within the CaP TB project (called Linkage facilitators). It was proposed to identify CHW for the community intervention among the linkage facilitators. In Cameroon, since there was no CHW involved in TB activities, they were identified among existing CHW involved in other community health activities (COSA - Comité de Santé (health committee)). Based on literature review of community interventions, findings of the acceptability survey and discussion with stakeholders in both countries, a procedure for selection of the CHW was proposed including the following criteria: having experience with community work, living in the same community, medium level of education, time to perform the tasks, accepted and respected by the communities. In Cameroon there were 3 CHW per intervention site, with a total of 15 CHW and in Uganda there were 2 CHW per intervention site for HC IV and one CHW for HC III, with a total of 12 CHW. In both countries, it was proposed that CHW will report to the facility TFP.
Taking into consideration the absence of research experience of CHW, to guaranty good quality of data and to ensure that a clear distinction could be made between activities related to the intervention and activities related to research, it was proposed that RAs will accompany CHWs to households and will be in charge of the informed consent procedure for contacts and data entry in the electronic case report form (eCRF) from source documents filled by the CHW.
Transport cost for the community activities was identified as a barrier by both TB patients and providers in the acceptability survey and by stakeholder during study preparation. It was proposed that the study will cover the transport cost but that existing public transport will be used as much as possible to ensure the sustainability of the intervention and avoid stigmatization. Good communication between facility TFP and CHW was also identified as a very important factor justifying the allowance of small budget for communication (airtime). Therefore, to ensure sustainability and to comply with existing practices, it was proposed that CHWs will not receive a salary, but will be compensated for their time and transport.
Finally, working with CHW on a new intervention implied to develop simple tools and check lists for TB symptom screening, adherence and tolerability assessment. These tools were developed in coordination with country TB stakeholders (see Additional Table 4 for symptom screening checklist). These tools were incorporated in simple standard operating procedures used for the training of the CHWs. CHWs were also trained to recognize potential severe symptoms or signs that would justify urgent reference of the child to the facility that could be related to other disease than TB. Although the initial aim of the study was that CHW will initiate child contact on TPT in the household, both National TB Programs of Cameroon and Uganda requested that initiation be done by a nurse in the household and that the CHW will be in charge of the follow-up on his own. They also requested a more frequent follow-up by the CHW, one and two weeks after initiation instead of 4 weeks as done at facility by TFP.
In Cameroon, due to national guidelines, HIV testing could only be performed by a nurse, therefore HIV testing in the community was done by a nurse. Cascade training was organized by country research team in each cluster facility sites followed by supervision by the RAs.