A Mobile App to Improve Symptom Control and Information Exchange Among Specialists and Local Health Workers Treating Tanzanian Cancer Patients: Human-Centered Design Approach

Abstract

Background: Improving access to end-of-life symptom control interventions among cancer patients is a public health priority in Tanzania, and innovative community-based solutions are needed. Mobile health technology holds promise; however, existing resources are limited, and outpatient access to palliative care specialists is poor. A mobile platform that extends palliative care specialist access via shared care with community-based local health workers (LHWs) and provides remote support for pain and other symptom management can address this care gap.

Objective: The aim of this study is to design and develop mobile-Palliative Care Link (mPCL), a web and mobile app to support outpatient symptom assessment and care coordination and control, with a focus on pain.

Methods: A human-centered iterative design framework was used to develop the mPCL prototype for use by Tanzanian palliative care specialists (physicians and nurses trained in palliative care), poor-prognosis cancer patients and their lay caregivers (patients and caregivers), and LHWs. Central to mPCL is the validated African Palliative Care Outcome Scale (POS), which was adapted for automated, twice-weekly collection of quality of life-focused patient and caregiver responses and timely review, reaction, and tracking by specialists and LHWs. Prototype usability testing sessions were conducted in person with 21 key informants representing target end users. Sessions consisted of direct observations and qualitative and quantitative feedback on app ease of use and recommendations for improvement. Results were applied to optimize the prototype for subsequent real-world testing. Early pilot testing was conducted by deploying the app among 10 patients and caregivers, randomized to mPCL use versus phone-contact POS collection, and then gathering specialist and study team feedback to further optimize the prototype for a broader randomized field study to examine the app's effectiveness in symptom control among cancer patients.

Results: mPCL functionalities include the ability to create and update a synoptic clinical record, regular real-time symptom assessment, patient or caregiver and care team communication and care coordination, symptom-focused educational resources, and ready access to emergency phone contact with a care team member. Results from the usability and pilot testing demonstrated that all users were able to successfully navigate the app, and feedback suggests that mPCL has clinical utility. User-informed recommendations included further improvement in app navigation, simplification of patient and caregiver components and language, and delineation of user roles.

Conclusions: We designed, built, and tested a usable, functional mobile app prototype that supports outpatient palliative care for Tanzanian patients with cancer. mPCL is expressly designed to facilitate coordinated care via customized interfaces supporting core users-patients or caregivers, LHWs, and members of the palliative care team-and their respective roles. Future work is needed to demonstrate the effectiveness and sustainability of mPCL to remotely support the symptom control needs of Tanzanian cancer patients, particularly in harder-to-reach areas.

Keywords: cancer; mHealth; mobile health; mobile phone; pain; palliative care; sub-Saharan Africa; user-centered design.

Figure 1

Palliative Care Outcome Scale of the mobile-Palliative Care Link app for a patient and caregiver (for display in Kiswahili). NA indicates that the caregiver is not present.

Figure 2

Design and development timeline of the mobile-Palliative Care Link app. Early app use feedback involved only specialists and study personnel; field study results are reported elsewhere.

Figure 3

Design of the mPCL app focusing on the patient or caregiver and care team communication and care coordination. mPCL: mobile-Palliative Care Link; POS: Palliative Care Outcome Scale.

Figure 4

Four separate interfaces for four different user roles (from left to right): patient or caregiver, nurse, specialist physician, and local health worker.

Figure 5

Palliative care Outcome Scale responses (mock patient) as viewed in the mobile app on a smartphone (left) and on the web app (right) by the specialist. A red triangle icon is displayed to alert the care team to reported pain scores that are above the set threshold.

Figure 6

Clinician web app view of patient information (top) and clinical record of the mock patient record (bottom).

Figure 7

Example screenshots of the patient symptom-focused educational resource (displayed in Kiswahili for patients). Patients are able to select specific content areas they would like to learn more about.

Figure 8

Screenshots of the emergency contact module. When the patient clicks Yes on the first screen of the module (left), they are advanced to the next screen (right), where they can click on a hyperlink to directly call the designated emergency contact.