You are here

Improving patient retention through ConnecTB

Improving patient retention through ConnecTB
Publication date: 
Sep 2017


The Global Tuberculosis report of 2016 reports that there were an estimated 10.4 million new (incident) TB cases worldwide in 2015. It also reports that there were an estimated 480 000 new cases of multidrug-resistant TB (MDR-TB) and an additional 100 000 people with rifampicin-resistant TB (RR-TB) who were also newly eligible for MDR-TB treatment. However only 25 000 (20%) were enrolled into care, with a global treatment success rate of 52%. 

South Africa reported 294 203 (64.9%) of the estimated 454 000 incident cases. 13 257 MDR cases were initiated on treatment in 2015 which was 64.2% of the 20 637 diagnosed cases. The 2013 treatment success rate for MDR-TB cases was 48% and that for XDR-TB cases was 24%. 

Of note is the high number of patients lost to follow up in the MDR-TB programme. Of the 10 880 MDR-TB patients initiated on treatment in 2013, 2411 (22.2%) were lost to follow up from TB treatment. Provincial analysis show loss to follow up (LTFU) rates above 30% in some provinces. Patients diagnosed with TB who do not complete their treatment represent a significant failure in the provision of care. These patients may continue to spread TB in the community thus retaining patients in care can reduce tuberculosis transmission to others. 

Health facilities do not always have adequate resources readily available for visiting TB patients in their respective communities. For this reason, national community outreach programs have been implemented to assist patients during their treatment journey, also ensuring that all their close contacts are being screened and tested for TB regularly.  There is, however, a great need to find innovative ways to locate discharged TB patients at their homes in order to support them and their close contacts more effectively. In light of the high loss to follow up, a mobile health solution was developed in an effort to improve retention of patients receiving MDR-TB treatment. 

Mobile Health Solution: ConnecTB

Mobile Health (mHealth) is the use of mobile and wireless technologies to support the achievement of health objectives. The technology takes advantage of the growth in the number of mobile users and rapid expansion of mobile networks. 

To address the problem of high loss to follow up rates in South Africa, the project developed a mHealth application, ConnecTB. This was based on the University Research Company (URC) experience in successfully implementing mHealth packages in Bangladesh to improve patient retention. mHealth uses technologies that are accessible and cost effective to provide invaluable real time insight into the quality of program performance, such as frequency of DOT visits; side effect profile of drug regimens and contact tracing coverage. 

The project contracts local NGOs to implement the direct observation of treatment using community health workers (CHWs). There was need to standardise the support visits to include patient counselling on their medication, prompt for side effects and providing health education to the family. ConnecTB was developed to:

  • ensure DOT visits are conducted
  • record the side effects experienced by the patients
  • ensure contact tracing is done
  • asses home infection control measures and to
  • track retention in care. 

ConnecTB is composed of a web portal and two complimentary Android applications. The web portal is used by administrators to register patients, supervisors and directly observed therapy (DOT) providers, and by supervisors to review submitted DOT surveys.  The Patient Management (PM) Android application is used by DOT providers to submit DOT visit surveys and geo-tags. 

Once TB patients are registered on the platform by the administrator, they are allocated to a DOT provider near the patient who will provide comprehensive support to the patient. This includes education on the medication, side effects assessments, contact tracing and health education to the family. In addition to adherence support, the system can geotag the patient location allowing for identification of hot spots. Supervisors can remotely verify completed visits and follow up with the DOT provider where there are gaps in supporting the patient.


ConnecTB has been implemented in the supported districts from June 2015 to date. For implementation, the project prioritised the NGOs who were funded by the project and they were caring for MDR TB patients. 

Selection of Implementers:

  • Three NGOs were initially selected for implementation.
  • St. Francis Hospice was selected in Eastern Cape. The NGO was working in Nelson Mandela Bay Metro conducting DOT support. St. Francis Hospice employed 5 nurses and 22 Community Care Givers (CCGs) trained on DOT provision. 
  • Letsema Circle Trust was selected in Eastern Cape working in Buffalo City district. Letsema employed 2 Professional nurses and 8 CCGs.
  • Isiphephelo was selected in Mpumalanga. The NGO was providing DOT support in Gert Sibande district using 4 CCGs and 2 Professional nurses. 
  • Each CCG supported 10 patients while each nurse supported 15 patients. Each CCG was allocated a smart phone to conduct DOT and the supervisors were provided access to server and automated reports.

Following the selection of the NGOs, the names of the patients in the catchment areas of the NGOs were obtained from facilities and captured on ConnecTB. Patients were allocated to the DOT supporters who provided comprehensive support to the patients.  


The ConnecTB application has been implemented from June 2015 to date. As of September 2017. ConnecTB has been expanded to 11 NGOs. The cumulative number of MDR-TB patients enrolled on Connect TB has increased from the initial 106 in June 2015 to 924. In addition, ConnecTB has been extended to monitor DS TB patients. 

To date, 924 multidrug-resistant (MDR-TB) patients have been linked to the ConnecTB platform at various stages of MDR-TB treatment. A total of 133 patients completed treatment, while 42 died. Further, 249 patients either developed complications requiring referral to hospital, or moved away from the district. Eight patients were lost to follow-up.  

A further 1,125 drug-susceptible TB (DS-TB) patients from 11 districts were recently enrolled on the platform; their monitoring is ongoing.

Cumulatively the system has recorded over 174 000 successful DOT visits and 4342 side effects of which 593 were severe and required referral. 3146 contacts were identified and screened, 283 had symptoms of TB and were referred to health facilities for testing. The system reports DOT completion rates of up to 95% for supported TB patients.

The additional benefit of ConnecTB is geo-mapping. As TB patients are located in their communities, the geo locations are mapped on the mobile application and a real-time map is updated in the background providing an overview of where patients reside. This information highlights TB hot spot areas in need of intervention, specifically contact management to determine the transmission of the disease to those in close proximity to TB patients. To date, five districts have been mapped and hot spots identified in each areas as shown in the figures below. 


The use of the ConnecTB application enables the project to present real time interactions between the CCGs and TB patients. It also enables the project to pinpoint the location of this interaction, the duration of this interaction and real time reports of adverse events. Thus the tool enables standardisation of patient support visits while at the same time acting as a supervision tool for CHW.

While the supervisory aspect of ConnecTB is important for programmatic monitoring and evaluation of success or failure of intervention, standardisation of patient management is especially important as health care delivery in the community is frequently implemented by cadres with minimal medical training. The high number of clients referred to hospitals for side effect evaluation requires more rigorous investigation as it offers an important clue to the high lost to follow up rates for MDR-TB clients on treatment, receiving adherence support through the government funded community based organisations.  The clustering of MDR-TB patients evidenced by the geo location of supported clients also points to high transmission zones in the community. These require more community wide interventions, including expanded contact management and targeted community activations (health education, and community screening). Collaboration with the national laboratory services is also a consideration to genotyping of TB disease in these high transmission zones.

Currently, ConnecTB records patients who are active, those who have completed treatment, those who died and those referred to the hospital, The current version of the tool does not record interactions for patients admitted to the hospital. These patients are recorded as transferred to hospital until they are discharged back into the community. To address this challenge, the newer version of the tool under development will include a hospital module to allow for continued monitoring of patients while in hospital. The next version will also report on cohort analysis in the same manner that the National TB program reports on outcomes. It will also have reminders for sputum collection and also capture results of tests done with algorithms to work out outcomes based on the results. The hospitals and clinics will also have the application so that patients who move to clinics and hospitals are captured while there and not reported as having missed their doses. 

To further enhance health education through the application, the project is in the process of incorporating videos and flips charts for education to the tool. This will enhance counselling of patients and provision of health education to the patient and the family. 

Is ConnecTB cost effective?

The cost of implementing ConnecTB includes the cost of setting up the application, purchase of smartphones, monthly data bundles for CHW and the cost of NGOs to support the TB patients. The project conducted an interim cost effectiveness analysis of using ConnecTB. This was done by comparing the retention rates from the National Department of Health (NDOH) 2013 cohort, who were followed up at local facilities, with retention rates for patients followed up by community health workers using ConnecTB.   The mean additional total cost per patient followed up using mHealth is $1100.00 for the duration of the treatment. The cost of losing a patient was the conservative estimate of treating one patient for MDR-TB (USD$ 6772). (If a patient is lost, they can infect 15 people per annum of which 10% may go on to develop TB disease) 

In the 2013 cohort, 22.8 % patients were lost to follow up, compared to only < 5% for those supported through ConnecTB. The interim cost analysis shows an average cost saving $ 376 per patient retained in care using ConnecTB. In addition to the cost savings, patients on ConnecTB received comprehensive follow up care which included identification of side effects and referral to care. On average, three contacts were screened per patient, leading to early diagnosis and linkage to TB care.  

Based on the preliminary results, the project recommends that mHealth approaches should be scaled up to other NGOs managing DRTB patients as it is cost-effective.  An additional analysis will be conducted to assess outcomes of patients on ConnecTB compared to those on routine care. The current model of implementing ConnecTB in through funded NGOs. The National department of health has employed community health workers who are part of the Ward Based Outreach Teams (WBOTs). The project is in the process of piloting ConnecTB using WBOTs. This will ensure sustainability beyond the life span of the project. This will further reduce the cost as the WBOTs are already funded through the department of health.  


The use of the ConnecTB application has resulted in impressive high adherence rates. The use of the application will improve the outcomes of treatment for XDR-TB and MDR-TB patients.