Mobile Health Technologies could enhance Public Private Mix for Tuberculosis care in Rural Southwestern Uganda: Qualitative Findings

This was a qualitative study design that involved in-depth interviews with 13 key informants (private healthcare workers) purposively selected between June and July 2020 due to their active involvement in Tuberculosis care from four private hospitals in Mbarara City. The interviews were transcribed and coded to identify key themes for analysis using content analysis.


Objective
To explore the potentials of mobile health technologies in fostering public-private mix for Tuberculosis care in Uganda.

Methods
This was a qualitative study design that involved in-depth interviews with 13 key informants (private healthcare workers) purposively selected between June and July 2020 due to their active involvement in Tuberculosis care from four private hospitals in Mbarara City. The interviews were transcribed and coded to identify key themes for analysis using content analysis.

Results
Mobile Health technologies (such as mobile apps, text messages) have the potential to map and link patients from private hospitals to the referral units, support patient medication adherence, notify and report Tuberculosis cases to the Ugandan Ministry of Health, and enhance patient care and monitoring.

Conclusion
Mobile Health technologies have the potential to revolutionize Tuberculosis care by establishing a centralized pathway for linking the referred patients from private hospitals to public hospitals. Future research should focus on assessing the utilization of mobile health technologies in enhancing access to referral units by presumptive Tuberculosis patients referred from private hospitals in low-resource settings.

Background
The public-private mix for tuberculosis care Tuberculosis (TB) remains one of the top causes of death globally despite its being curable and preventable [1]. Globally, About 10 million people developed TB disease in 2019 of which 1.4 million died [2]. The burden of TB disease in Uganda remains high with an incidence rate of 200 (117-303) per 100,000 [3]. Despite the Human Immunode ciency Virus (HIV)-Positive TB incidence rate of 78 per 100,000, Uganda is still ranked among the 30 high TB/HIV burden countries in the world [2]. Developing countries are continuing to struggle towards achieving the 2035 global targets of the End-TB strategy.
These include achieving a 95% decline in deaths due to tuberculosis compared with 2015, reaching an equivalent 90% reduction in tuberculosis incidence rate from a projected 110 cases/100 000 in 2015 to 10 cases/100 000 or less by 2035 and that by 2020, no tuberculosis-affected person or family should face catastrophic costs due to tuberculosis care [4]. Intensi ed research and innovation is one of the pillars for achieving these targets, under which, research to optimize implementation and impact, and promote innovations is a major action. It has been reported that indeed most innovations cannot be translated into effective local action without careful planning and adaptation, and partnership with stakeholders [4]. The private healthcare sector is a key stakeholder in TB care whose engagement needs to be more enhanced by Government-based National Tuberculosis Programs (NTPs). Although previously considered to be better avenues for providing promotive, preventive, curative components of general healthcare [5], government health facilities are continually becoming more overwhelmed by the increased TB disease burden, hence needing supplementation by private health facilities. The public-private mix (PPM) was introduced in 2003 by World Health Organization (WHO) to involve all the relevant healthcare providers (private health facilities) in the provision of TB care and control [6]. There is some evidence that shows that PPM is cost-effective for TB care [7], and increases case detection [8]. From 2010-2019, there has been a signi cant contribution of PPM which resulted in the increase in the annual number of case noti cations (1.8 million cases in 2019) among the high TB burden countries (i.e. Bangladesh, India, Indonesia, Myanmar, Nigeria, Pakistan, and the Philippines) [2]. Despite some global progress in the implementation and expansion of PPM, many private healthcare providers remain unengaged [9]. In Uganda, the engagement of private health providers in TB care and prevention by the Ugandan Ministry of Health/ National Tuberculosis and Leprosy Programme (MOH/NTLP) is minimal despite being a signi cant contributor of health services, with 47% of all hospitals in Uganda being private [10]. Although many patients in Uganda access healthcare through private service providers, most private healthcare providers (such as private clinics and hospitals) remain incapacitated to provide TB treatment and care [11]. They mainly screen patients for symptoms and refer patients for better management in government health facilities. This could result in delayed case detection and noti cation rates.
To address the current gap in PPM for TB care, guidelines for enabling the standardization of PPM for TB were developed in 2018 by Uganda stop TB partnership and the Ugandan Ministry of Health/NTLP [10].
According to this report, the implementation of PPM in Uganda is conducted via a phased approach to enable the engagement of private facilities in TB service provision through training and mentorship, to build the capacity of the participating private health facilities. Progress indicates that a handful of private health practitioners including doctors, nurses, and laboratory personnel from six districts (Mbale, Tororo, Jinja, Iganga, Gulu, and Kitgum) in eastern and Northern Uganda were trained in the implementation of PPM activities. In addition to this capacity building, the MOH in partnership with United States Agency for International Development (USAID) Defeat TB project trained 20 private health practitioners to improve quality TB care to enable private health facilities to improve TB case noti cation, detection, and treatment in Kampala, Wakiso and Mukono Districts in Central region, Uganda. Although this continuous mentorship and coaching resulted in the engagement of over 150 health facilities in the country [10], facilities and healthcare workers from the southwestern part of the country were left out.

The Potentials Of Mhealth
Mobile health (mHealth) refers to the use of portable devices with software applications to provide services and manage patient information [12]. These technologies include mobile phones, tablets, and personal digital assistants. The widespread usage of mobile phones globally with the number of subscribers growing exponentially and exceeding even the population size in some countries [13,14] provides opportunities for their use in the health sector. mHealth interventions are less expensive, with the potential to reach a wide client base and the hard-to-reach groups [15,16]. The adoption and integration of mHealth in the routine operations of the health sector have been slow compared to other sectors like banking, airline, and insurance [17]. There is some evidence that mHealth technologies can enhance PPM through TB screening, case noti cation, treatment adherence, data collection and management, patient referral, and follow-up [18] and promote the adoption of health behaviors including HIV/AIDS prevention [19,20] and prenatal service utilization [21][22][23].
The use of mHealth technologies (e.g. smartphone apps, text messages, wearable devices) in supporting TB care is highly recommended by the WHO as emerging opportunities that can close the gaps in TB care [24]. These technologies can potentially enhance the identi cation of lost follow-ups [25] among TB patients, support TB case identi cation, diagnosis, noti cation, reporting, and monitor drug adherence [26]. Although the application of mHealth in TB care and management particularly in developing countries is still in its infancy, the modest existing studies show some potential. For instance, mHealth technologies (e.g., SMS texts and real-time medication adherence monitors) can improve TB medication adherence through reminders [27], foster the provision of social support (e.g. money sent on mobile phones for transport, phone calls for emotional support) to TB patients [28]. mHealth technologies are low-cost applications that are sensitive to both the private sector ways of working and the expectations of the NTPs [9] with the ability to produce tremendous results in facilitating the engagement of the private hospitals. However, there is a dearth of research about their application in fostering PPM. This knowledge would practically inform the future development of mHealth interventions for engaging private health facilities in TB care. Yet the use of such technologies can potentially address the current highlighted by Khan and colleagues [29]-weak patient referral systems, lack of appropriate patient tracking and support systems, and the lack of the ability to cross-link and identify patients who move between private and public sectors, which continue to cripple the performance of TB control programs. The purpose of this study was to explore the potentials of mHealth in fostering PPM in Uganda.

Study design and setting
We conducted a cross-sectional study with a qualitative research approach that involved in-depth semistructured interviews with healthcare workers (medical doctors and nurses). The methodology for this study has been described previously [30]. Brie y, the participants were recruited from three private forpro t (PFP) and one private not-for-pro t (PNFP) hospital in Mbarara City, in South-Western Uganda. These facilities were purposively selected for having relatively large medical establishments, a large number of health care workers, being involved in TB case detection, and being at the center of handling a large number of outpatients in Mbarara City, which increases the likelihood of detection of presumptive TB cases. After performing symptom-based screening of patients for TB, all hospitals refer the presumptive TB patients to Mbarara Regional Referral hospital which is the only NTP facility that manages Tuberculosis in Western Uganda.

Selection of study participants
Key informants from four private hospitals in Mbarara City were purposively selected based on their current or previous involvement in TB-related activities such as case identi cation, referral, and treatment, as well as their willingness to participate in the study. A total of 13 healthcare workers (11 medical doctors and two nurses) were recruited (Table 1 below). Speci cally, participants were included based on the following inclusion criteria: a) employed medical doctor/nurse in a private hospital setting, b) have experience in TB-related activities, c) willing and able to give Informed consent.

Data collection
Between June and July 2020, a semi-structured interview guide was developed in English, pre-tested to ensure data reliability, validity and accuracy. The researcher (WT) who has experience and training in qualitative research and research ethics, conducted semi-structured in-depth interviews with the medical doctors and nurses from the identi ed study sites. The interviews involved engaging the health workers to tell their stories as prompted by the investigator. Each interview lasted between 30 and 40 minutes and was carried out in a private convenient place preferred by the respondents. Interviews were recorded digitally with participants' consent and transcribed thereafter. Questions about the perceived bene ts and barriers of mHealth in fostering PPM were elicited. Following each conducted interview, the transcripts were reviewed for quality assurance, clarity, and detail by authors AM, AD and FK. A short survey was then administered to the medical doctors and nurses to gather their social demographic details.

Analysis
Through the initial review of transcripts, researcher WT developed an initial coding scheme which was reviewed and discussed by authors AM and FK. The inductive content analysis approach [31] was used to generate themes and subthemes from the codebook. Categories about the potentials of mHealth (e.g. Referral, supporting medication adherence, TB case noti cation) in fostering PPM in TB care were extracted from the transcripts. The codebook was reviewed by AM to ensure that the relevant content was extracted. WT assembled the codebook following an iterative approach that included category construction of codes (namely: Mapping and linking referred patients to the referral points, supporting patient care and medication adherence, and Tuberculosis cases noti cation and reporting) to represent the content, writing the operational de nitions, elaborations and illustrative quotes. Intercooled STATA (Statacorp 2013 Stata Statistical Software Release 13; StataCorp, College Station, TX, USA) was used to describe study participants' characteristics.

Results
Thirteen healthcare workers (key informants) composed of 11 medical doctors (85%) and two nurses (15%) from four private hospitals in Mbarara City participated in this study. The majority of the participants 8 (62%) were male, with a median age of 29 years, the majority were recruited from private for-pro t private hospitals (85%) and with 5 or fewer years of medical practice; as indicated in Table 1 below Concerns: Lack of knowledge to use the technology Some participants expressed concerns about the lack of knowledge among the health workers to use the developed mobile applications which might hinder the operation and implementation of the technology due to abandonment by the intended user. One medical doctor from Hospital 2 reported that "Some health workers may not know how to use the app and in the long run may abandon it".

Concerns: Poor internet connection
Participants also raised the concern of poor internet connection (e.g., slow connections or lack of internet) which might affect the usability of mHealth interventions that rely entirely on the internet. One participant from Hospital 3 "If the app uses the internet, in some places, there is poor network connections, which might affect the operation of the intervention".

Discussion
Our study explored the potentials of mHealth in fostering PPM in Uganda. Overall, key informants discussed the potentials of mHealth technologies in fostering PPM in TB care. mHealth technologies (such as mobile apps, text messages) have the potential to map and link patients from private hospitals to the referral units, support patient care and medication adherence, as well as notify and report TB cases to the Ugandan Ministry of Health.

Meaning Of The Study And Comparison With Other Studies
The potential of mHealth to map and link referred patients to the referral points could eliminate the risk of loss of patients along the pathway of care but could also result in enhanced patient care and monitoring. Through such mHealth interventions, the proportion of patients that reach their expected referral points and who get started on medication on time could increase, with subsequent reductions in diagnostic and drug initiation delays. Most of the private hospitals in Uganda are not accredited to provide TB services, this limits their efforts and engagement in TB care despite being at the center of receiving a majority (48.4%) of the population who seek healthcare compared to the government (public) health facilities (34%) [32]. Therefore, as a common practice, most of these private hospitals refer presumptive TB patients to public hospitals [33]. However, the lack of mechanisms for following up these patients in private hospitals [30] still sti es TB control efforts. In this regard, the adoption of the current ICT technologies like smartphones, mobile apps (mHealth) as a mechanism for following up these patients after referral could ensure easier patient tracking and successful referral processes from private to public health facilities. For example, the use of telephone calls as a measure for following up the referred TB patients from private facilities to the peripheral units has been reported in Cambodia and resulted in increased retention of the referred TB patients [34]. In India the utilization of a smartphone application by rural private healthcare practitioners to refer patients to TB microscopy Centre located at public health facilities resulted in an increased number of referrals, reduced time taken for diagnosis, and treatment initiation [35]. However, on the other hand, the implementation and utilization of mHealth in fostering PPM in Africa remain unknown, yet most countries in Africa still grapple with a high number of TB cases (25% of the global TB cases come from Africa) [2]. Understanding how mHealth is utilized to engage the private sector in TB care especially in African settings has the potential to ensure better TB disease outcomes in Africa.
The potential of mHealth in supporting patient care and medication adherence among TB patients could enhance timely in-take of medication and the attendance of hospital appointments for pill re lls. Particularly in our study, participants reported that mHealth technology would enable healthcare workers to monitor patients thus enabling adherence to medication and attendance of hospital appointments.
mHealth provides quick access to information for both patients and healthcare workers and offers a qualitative and quantitative interaction between the healthcare providers and the patients [23]. Several studies utilizing mHealth technologies for supporting medication adherence among both TB and HIV patients have been conducted in Uganda [19,27] and were found to be acceptable and feasible. A randomized control trial [36] that was conducted in Karachi Pakistan utilized two-way short message service (SMS) reminders to encourage newly diagnosed patients from public and private health facilities to take their medication. However, the utilization of mHealth technologies between the public and private health facilities in Uganda for TB care is unknown. This leaves a research gap and calls for a special investigation on how mHealth technologies can be utilized for enhancing PPM in TB care.
Our study ndings report that mHealth interventions could also enhance quick and timely noti cation of TB cases from the private health facilities to national TB programs to ensure better planning of TB management services. Notifying cases is key in enhancing the tracking and management of the patients. . Although mHealth has been utilized in the noti cation of TB cases elsewhere, its experience among healthcare workers in settings like Uganda where healthcare workers might be overwhelmed with a large number of clients (high patient-doctor ratio) and fail to nd time for real-time use of the application to notify cases; this is being explored in our ongoing research investigation.
In this study, participants raised concerns about the lack of knowledge to use the developed mobile applications. User involvement during intervention development may play a role in acquainting system end-users with the intervention which in the long run will facilitate the acceptability of the developed application among users [39]. This approach helps the developers to understand user behaviors, requirements, and skills at each phase of the design [40]. Training users on how to use the application before implementation may improve usability and address this concern. Users who are not well trained to use nd it hard to use the technology [41]. Therefore, system implementers should ensure conducting initial training sessions before rolling out the interventions. In addition, developers should develop interventions that are easy to use [18] which are intended to ease the work of healthcare workers.
The reported poor internet connectivity or lack of internet access especially for interventions that entirely run on the internet or require the internet to function can hinder the usability of the application and result in abandonment of the interventions. mHealth interventions that entirely rely on the internet to operate, require stable internet connections to ensure thorough operability. The jurisdiction of the cost burden of internet packages for running these internet-based interventions must be clearly de ned lest the cost burden on the users can't be sustained in the long run. mHealth interventions should at all costs remove the cost burden from the users if the successful implementation is to be ensured. Considering the development of standalone interventions or applications that do not require the internet to operate can address this concern.
The general implementation and coverage of PPM

Limitations of the study
Our study is not without limitations that should be considered during the interpretation of our research ndings. First, being a formative study, the investigator solicited the perceptions of private healthcare workers about using mHealth in TB care before they could use the intervention in real life. The results, therefore, do not document the practical real-life experiences of utilizing the mHealth intervention in TB care. Second, our study employed a purposive sampling approach for recruiting key informants, which has the potential for selection bias. Lastly, this study was conducted in private hospitals in an urban setting; this could limit the generalizability of our ndings in rural settings, where private hospital operations and practices might be different.

Implications for policy and practice makers
The utilization of mHealth in TB care among private hospitals in Uganda is lacking. This limits the realization of mHealth bene ts in the achievement of better TB treatment outcomes and the realization of PPM. Our ndings imply that mHealth has the potential to revolutionize TB care by establishing a centralized pathway for linking the referred patients from private to public hospitals and ensuring that they are not lost along the TB care pathway, have been diagnosed, and started on medication. Our study implies that mHealth can be a cost-effective solution for enhancing the full integration and engagement of the private healthcare facilities in TB care thus the achievement of PPM.
The implementation of the mHealth interventions requires training of users to equip them with skills to enable intervention acceptability and usability. mHealth interventions that entirely rely on the internet to operate, require stable internet connections to ensure thorough operability. Standalone applications that do not require the internet to operate can offer better alternatives. mHealth interventions should at all costs remove the cost burden from the users if the successful implementation is to be ensured.
Our ndings recognize that there is a need for future researchers to assess the utilization of mHealth for following up patients that have been referred from the private hospitals in African settings to ascertain its effectiveness on TB care outcomes and its contribution to the PPM effort for TB care.

Unanswered questions and future research
Exploring the real-life experiences of how mHealth can be used to foster the establishment of PPM in rural Uganda is key in informing policy formulation to address TB care outcomes.

Conclusion
This study showed that mHealth could provide acceptable and feasible approaches to map and link TB patients to the referral units, enhance patient care and medication adherence and notify TB cases.
Understanding how mHealth technologies are perceived by private healthcare workers is key in informing the development of interventions that are acceptable and feasible and effective in enhancing the engagement of private hospitals in TB care. Findings from this formative study are informing the development of an m-Health intervention to enhance the engagement of private hospitals in TB care, which the authors are currently working on. The utilization of mHealth could offer a cost-effective solution in enhancing the implementation of the PPM. The Private hospitals, policymakers, researchers, and developers need to bear in mind that the utilization of mobile health technologies not only improves the quality of care among the presumptive TB patients but also could enhance better treatment outcomes like early diagnosis and drug initiation thus limiting the risk of infection at both the community and household level.

Consent for publication
Not applicable Availability of data and materials The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request

Competing interests
The authors declare that they have no competing interests