Effect of mobile health interventions on enhancing completion of the diagnostic process among peoplr with presumed tuberculosis in Central Uganda

Abstract

Background: In 2025, tuberculosis (TB) remained the leading infectious killer despite being preventable. The “WHO End TB” strategy calls for a reduction in TB incidence by 90% and deaths by 95% by 2035. However, progress towards achieving this strategy has been partly hampered by the continued loss to follow-up (LTFU) of patients along the TB care cascade. In the TB care cascade, patients can be lost either before they are screened or before they are diagnosed when presumed with TB or before they initiate treatment after testing positive for the infection or before completing treatment. At the health facility level, the biggest LTFU of patients happens between one being presumed to have TB and getting a diagnosis leading to loss to follow-up (LTFU) before completing the diagnostic process despite having reached the facility. Continuous LTFU of people with presumed TB before they are diagnosed propagates community transmission of mycobacterium tuberculosis (Mtb) and poor patient outcomes. There is therefore an urgent need to come up with interventions to reduce this LTFU of people with presumed TB before they are tested. Objectives: The main objective of this thesis was to assess the magnitude of pre-diagnosis LTFU and the effect of mHealth interventions to enhance completion of the diagnostic process among people with presumed TB in Central Uganda so as to inform the design of interventions. Specific objectives were: i) to determine the proportion of pre-diagnosis LTFU and factors associated in Central Uganda (sub-study 1); ii) to explore the facilitators and barriers to completion of the diagnostic process among people with presumed TB in Central Uganda (sub-study 2); iii) to synthesize evidence on mHealth interventions used to reduce LTFU among people with presumed TB and those diagnosed (sub-study 3); iv) to assess the effect of the SMS or phone call intervention to enhance completion of the diagnostic process among people with presumed TB (sub-study 4) and v) to explore factors that influence use of the SMS or phone call intervention to enhance completion of the diagnostic process among people with presumed TB (sub-study 4). Methods: The study was implemented under four sub-studies based on the Socio-ecological (SEM) model using both quantitative and qualitative methods of data collection. In sub-study 1, a retrospective cohort analysis of secondary data was conducted among people with presumed TB who presented to the study facilities from 1st January 2019 to 31st December 2022 to determine the proportion of pre-diagnosis LTFU and factors associated. Modified Poisson regression modelling was used to estimate adjusted prevalence ratios (adj.PR) and factors associated with TB pre-diagnosis LTFU. In sub-study 2, twenty-five In-depth interviews (IDIs) with purposively selected people with presumed TB who either completed or dropped out of the diagnostic process were conducted to explore facilitators and barriers to completion of the diagnostic process using a qualitative descriptive design. In addition, six focus group discussions (FGDs) with patients who had recently initiated TB treatment and 20 key informant interviews (KIIs) with health workers were also conducted. Thematic analysis was used for analysis. In sub-study 3, a systematic review was conducted to synthesize evidence on what mHealth interventions have been previously used to reduce LTFU in the TB presumption phase. In sub-study 4, a mixed methods study was embedded in an individual randomised controlled trial (RCT) to assess the effect of Short Message Service (SMS) and phone call intervention on completing the TB diagnostic process and explore factors that influence their use. Results: From sub-study 1, overall 28.3% (3,699/13.064) experienced pre-diagnosis LTFU within 30 days from the day they were presumed with TB. Pre-diagnosis LTFU was more likely to occur among patients aged 0-14 years (adj.PR=1.1, 95% CI: 1.06, 1.24), females (adj.PR=1.06, 95% CI: 1.01, 1.12), those with no record of place of residence (adj. PR=2.7, 95% CI: 2.54, 2.93), and those with no record of phone contact (Paper 1). In sub-study 2: the key facilitators for completion of the TB diagnostic process included; persistent symptoms, obtaining same day results, prior TB knowledge, social support and caring health workers. Barriers were; TB and HIV stigma, long turnaround time of results, stock out of supplies, unclear appointment for collection of results, lack of transport to return to the health facility, and poor health worker attitude (Paper 2). From the systematic review in sub-study 3, SMS, phone calls, WhatsApp messages and mobile applications (Apps) were the mHealth interventions utilised in the TB presumption phase. Phone calls and SMS were however the commonly used interventions mainly used as reminders to collect or deliver TB result, and for linkages and referrals to diagnostic centres. The evaluation of the SMS and phone call interventions revealed that completion of the TB diagnostic process was highest among participants randomised to the phone call arm compared to those in the standard of care 74.3% (434/584). There was no statistically significant effect for those in the SMS arm (adj RR 0.99, 95% CI: 0.91,1.08). Individual factors, social influences, technological factors, intervention design factors, health system factors and participant experiences influenced the use of the SMS and phone call intervention (sub-study 4). Conclusion: From the study findings in this thesis, more than a quarter of the people with presumed TB got lost before getting tested or before receiving their test results. Completion of the TB diagnostic process was facilitated by obtaining same day results, prior TB knowledge, social support, while TB and HIV stigma, long turnaround time of results and lack of transport to return to the health facility were barriers. SMS and phone calls were the commonly used mHealth interventions in the TB presumption phase. From the evaluation of the intervention, completion of the TB diagnostic process was highest among participants in the phone call reminders arm compared to the standard of care. Implementation of SMS or phone call interventions is influenced by social influences, technological factors and intervention design factors. To enhance completion of the TB diagnostic process, there is need for empowering laboratories with resources to avail same day results, short of which mHealth interventions that incorporate phone calls should be considered to communicate readiness of results to the patients. Design of SMS and phone call interventions should take into consideration individual factors, social, technological factors, user experiences and localised design strategies. Social support and stigma reduction interventions should be embraced both in the community and at the health facility to support people with presumed TB to complete the diagnostic process.