Implementation Research for
Digital Technologies and TB

Implementation Research for Digital Technologies and TB

Case study

Electronic medication monitors in China

(Patient care)
Case study : 1 Photography by Safar Safarov


This example is taken from a study conducted during the piloting of EMM in 30 counties across China which aimed to assess and evaluate the early implementation experiences

Research objectives

To determine the i) proportion of eligible TB patients using EMM and the factors associated with using EMM within the first month of diagnosis ii) treatment adherence among those using EMM, including the proportion of patients who were shifted to DOT as indicated by adherence data


At time of evaluation, 72% of target counties had implemented electronic medication monitors (EMM) as standard of care.

Areas for further investigation: What proportion of patients who are offered EMM accept the technology? What are the sociodemographic characteristics of patients who do/do not accept EMM?


An evaluation was conducted on the EMM involving 432 newly registered TB patients across nine counties in four provinces in China to compare EMM dosing indications with actual medication ingestion based on random urine tests. The evaluation found that there was a high degree of correlation between both indicators of treatment adherence (adherence based on urine test was observed among 99.5% patients vs 96% of patients deemed adherence based on EMM records), suggesting that EMM could be effectively used as evidence of TB treatment adherence.


Areas for further investigation: What are the technological requirements needed to adopt EMM (e.g. network coverage, feature/smartphone availability, device failure, supply chain capacity) and how do they vary across the country? Are there differences between adoption across public and private facilities?


An acceptability trial was undertaken, involving both patients and providers, to identify any unrecognised defects in the design and function of the EMM, to evaluate the effectiveness of labels, instructions and training for patients and providers, and to determine the acceptability among patients and providers of the technology and technology-enabled patient management approach. The trial found that the EMM system had a high degree of acceptance and satisfaction among TB patients and medical staff. The system was reported as easy to use and operate, which thereby improved patients' compliance, improved patients' medication experience, and also indirectly reduced the workload of medical staff.


EMM will be scaled up as standard of care for drug-susceptible TB patients.

Areas for further investigation: How does the impact of EMM change over time? What are the long-term cost requirements associated with scaling up the technology? Are there any components of the EMM system that must be sustained, discontinued or modified after initial roll out, and why? Has acceptability of the EMM changed over time? What are various risk factors/externalities/threats for continued implementation? How will EMM continue to function as the technology infrastructure and availability continue to change or improve?

Study design

This was a mixed methods study that included: i) longitudinal descriptive study (to assess eligibility, uptake and treatment adherence); and ii) cross-sectional study to assess factors associated with EMM uptake during first month of diagnosis.

Research outcomes

This study explored various outcomes related to implementation:

  1. Uptake of EMM among eligible TB patients: defined as the proportion of eligible people who used EMM within the first month of diagnosis based on dates of diagnosis and start of EMM use.
  2. Treatment adherence among those using EMM: defined as the proportion of month-wise missed dose as a proportion of expected doses among patients who use EMM within the first month of diagnosis, until the date of shift to DOT (for those shifted to DOT), or the date of stopping EMM (for those not shifted to DOT).
  3. Proportion of patients who were shifted to DOT as indicated by adherence data: defined as the proportion of TB patients eligible for shifting to DOT and who actually shifted to DOT.

Study population

All notified people with TB in the 30 implementation counties between July and December 2018 were included. Exclusion criteria: TB patients with known rifampicin- or multi-drug resistance, or patients with any communication impairment (mental, visual, auditory or speech) and requiring hospital admission at notification. Patients transferred into one of the 30 implementation counties were also excluded.


The EMM scale-up plan identified three provinces (comprising 138 counties). This study was conducted in the thirty counties that had begun implementation at the time of the study (May and June 2018). Implementation began in the remaining 108 counties by January 2019. The thirty counties belong to three provinces in China (Zhejiang in the east region, Jilin in the middle region and Ningxia Autonomous Region in the west).

In this setting, people with TB (not known to be rifampicin-resistant) receive daily fixed-dose combination treatment over six to eight months (depending on their status as newly diagnosed or previously treated case). As per recommendations of China's NTP, patients are recommended to use one of two treatment adherence support systems: i) directly-observed treatment (DOT) (by township or village doctor or family member or volunteer); or ii) self-administered treatment (SAT) with or without support by digital adherence technologies. Locally developed adherence technologies are recommended by the China NTP. If on SAT, patients are visited by village doctors every ten days during the first two months of treatment, followed by once a month to check on their health status and treatment adherence. All notified patients and their treatment outcomes are recorded in a web-based TB information management system.

Intervention/implementation strategy

The EMM was locally designed by a producer selected by competitive tender. The EMM was designed to monitor treatment adherence for a one-month fixed-dose regimen. The device is relatively small in size (16cm × 12.5cm × 7.3cm) and included a plastic box and electronic module that records each opening of the box by the patient, which is used to indicate that the patient has taken their medication. The electronic module is removable and is reused for at least three treatment cycles. The device also alerts users through audible and visual alerts of daily dosing, monthly refill and low battery. Battery life is approximately two months, and the average cost per patient is approximately US$5. Prior to the pilot, training materials and standard operating procedures were developed and provided to all doctors from designated hospitals within the pilot counties.

All newly notified TB patients meeting the inclusion criteria were offered EMM supported self-administered therapy at the beginning of outpatient treatment. EMMs were programmed by doctors to provide a medication reminder at the same time every day at a time decided by the patient and a treatment dose equivalent to one-month. Patients would return to the health facilities on a monthly basis to collect their next monthly dose. During these monthly visits, an adherence report was generated for the last month via the EMM operating system (offline software). If <20% of doses were missed, the patient was counselled on importance of treatment adherence. If 20–49% of doses were missed, the frequency of home visits by village doctors was increased to once every seven days for rest of the treatment. Patients whose data indicated continued instances of missing 20–49% of doses or a single instance of missing ≥50% of doses were shifted to DOT. In line with guidelines, continued use of EMM was encouraged by doctors even after shifting to DOT.

Data from EMM operating system was uploaded on a regular basis to an online EMM information system by designed county-level hospital staff. This data was used by the NTP to develop a progress report which was provided on a monthly basis to the health facility.

Data collection, management and analysis

Data collection methods

This study was a secondary analysis of routine programme data extracted from the following sources:

  1. TB information management system: notification date, diagnosis date, treatment initiation date, sex, age, occupation, migrant status, category of TB, classification of TB.
  2. EMM information system: date of starting and stopping EMM, number of days/doses missed, number of days patient should take medicines from first to eighth month.
  3. Paper-based patient records at county: eligible for EMM, reasons for non-eligibility, shifted to DOT anytime during treatment along with date of shifting.

Data management and analysis

Data from the paper-based records were digitized and merged with data from the TB and EMM information systems. Patient data were matched using the TB information system patient notification number. Stata was used for all analysis.

The following analysis was conducted:

  1. Uptake among eligible TB patients: among eligible patients, EMM use within the first month of diagnosis based on dates of diagnosis and start of EMM use was derived. For factors associated with uptake, adjusted prevalence ratios and 95% confidence intervals using log-binomial regression were calculated across sociodemographic and TB-related variables to identify factors associated with not using EMM within the first month among all eligible people. All available variables were included in the regression model, after ruling out any collinearity among variables. Among eligible subjects who did not use EMM within the first month, the number and proportion that started using EMM later during treatment was described.
  2. Treatment adherence among those using EMM: monthly missed doses as a proportion of expected doses were calculated and classified into the following categories: <20% missed doses; 20 – 49% missed doses; ≥50% missed doses.
  3. Proportion of patients who were shifted to DOT as indicated by adherence data: the number and proportion of patients eligible for shifting to DOT and actually shifted to DOT were described and stratified by those who started using EMM within and after the first month.

Based on: Wang N, et al (2020). Electronic medication monitor for people with tuberculosis: Implementation experience from thirty counties in China (

Share this story

See all case studies

Interested in using the toolkit and would like assistance?