Design of an Automated Mobile Phone-Based Reminder and Incentive System: Application in a Quasi-Randomized Controlled Trial to Improve the Timeliness of Childhood Vaccinations in Tanzania
Bibliographic record
Abstract
Background: The global penetration of mobile phones has offered novel opportunities for communicating health-related information to individuals. A low-cost system that facilitates autonomous communication with individuals via mobile phones holds potential for expanding the reach of health messaging in settings with human resource and infrastructure limitations. Objective: We sought to design a flexible, low-code system using open-source software that could be adapted to different contexts and technical environments and accommodate a wide range of automation needs. We report on key details of the mobile phone-based appointment reminder and incentive system (mParis), document its use, review implementation challenges and adaptations to address these challenges in the context of a quasi-randomized trial of mobile phone-based reminders and incentives as means of increasing the timeliness of childhood vaccinations in Tanzania, and outline other use cases that highlight the versatility of the system. Methods: The mParis instance described in this paper, which is hosted in Tanzania, sent automated, individualized vaccination reminders in the form of SMS text messages to the mobile phones of mothers of young children. Process workflows, based on the national vaccination schedule of Tanzania, were programmed into mParis. Reminders for vaccinations due at ages 6, 10, and 14 weeks were sent 7 days and 1 day before and 14 days after each vaccination due date. A subset of messages included financial incentive offers to mothers for the timely vaccination of their children. We report on implementation outcomes, challenges, and adaptations to address these challenges. Results: Between August and December 2017, a total of 412 pregnant women were enrolled in the trial. After mothers reported the birth of their children, individualized vaccination reminder messages were sent for vaccination due dates between January and July 2018. From March 2018, messages contained financial incentive offers. Of 1397 messages sent, 1122 (80.3%) messages were recorded as delivered, 249 (18.8%) as expired and resent; 23 (1.6%) as failed, and 3 (0.2%) as sent but lacking a delivery confirmation. In total, 633 (45.3%) messages contained incentive offers. Of 173 women who received at least 1 message, 67 (38.7%) were sent reminders only; 106 (61.3%) women were sent at least 1 incentivized message. Numerous challenges were encountered during the system's implementation, despite its deliberate design to accommodate basic problems, such as intermittent internet access and power failures. Continuous adaptation to increase the resilience of the system resulted in a successful deployment. Conclusions: mParis' open-source nature, auditability, and ability to autonomously execute algorithms in a low-resource setting with frequent infrastructure challenges suggest favorable prospects to automate health communication in a wide range of settings. mParis' use in other applications, including enrollment and follow-up for health-related research studies, demonstrates its versatility and ability to accommodate diverse challenges that may be encountered.
Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.
How this classification was reachedexpand
Full frame machine prediction
Teacher imitationNot calibrated prevalence, not ground truth. Human validation pending. The Gemma side is a direct model label for every work in the frame, read from the title-only record. The Codex side is a classifier learned from the 10,348 direct Codex labels and calibrated to design-weighted sample rates; fields without enough sample support carry no Codex call. Candidate is the union of the two sides; consensus is their intersection. These outputs are machine_predicted_unvalidated and are not human labels.
Distilled classifier scores by category (both heads)
| Category | Codex | Gemma |
|---|---|---|
| Metaresearch | 0.020 | 0.023 |
| Meta-epidemiology (narrow) | 0.002 | 0.001 |
| Meta-epidemiology (broad) | 0.003 | 0.003 |
| Bibliometrics | 0.001 | 0.001 |
| Science and technology studies | 0.001 | 0.002 |
| Scholarly communication | 0.001 | 0.002 |
| Open science | 0.002 | 0.001 |
| Research integrity | 0.003 | 0.002 |
| Insufficient payload (model declined to judge) | 0.010 | 0.001 |
Machine scores (provisional)
The two teacher heads of the student model, read on this work. A score orders the frame for review; it never asserts a category, and the validation status ships verbatim with every row.
Baseline scores from an immature model (maturity gate not passed, 7 training rounds). Scores rank; they never assert a category.
score_only:v0-immature-baseline · verbatim from the scoring run: score_only means the number may rank works, and no category label ships from itClassification
machine, unvalidatedMachine predicted; a candidate call from one source (direct Gemma or distilled Codex), not a consensus.
How this classification was reached, model by model and score by score, is at the end of the page under "How this classification was reached".