The effect of WeChat-based training on improving the village doctors’ knowledge of tuberculosis management: a pre-and-post intervention study from China

Lei Gao, Henan Xin, Haoran Zhang, Xuefang Cao, Boxuan Feng, Zhusheng Quan, Ying Du, Yijun He, Dakuan Wang, Bin Zhang, Qi Jin, Zisen Liu, Shouguo Pan, Ling Guan, Fei Shen, Jianmin LiuResearch Square 2020 · DOI

TL;DR

This study evaluates a year-long, WeChat-based mobile health (mHealth) intervention designed to boost tuberculosis (TB) management knowledge among rural village doctors in China. By delivering multimedia training content directly to doctors' smartphones, the researchers achieved a statistically significant improvement in knowledge scores, suggesting that ubiquitous social media platforms can serve as scalable, low-cost tools for professional medical education in resource-limited settings.

What Problem Does This Paper Address?

In China, approximately 70% of tuberculosis cases occur in rural areas, where "village doctors" serve as the primary frontline for healthcare. These practitioners are essential for early detection, referral, and patient management. However, their knowledge of TB protocols is often insufficient due to a lack of accessible continuing education.

Traditional in-service training—typically delivered via face-to-face lectures—faces significant hurdles: it is expensive, requires doctors to leave their clinics, and is difficult to scale across vast rural geographies. The authors sought to determine if WeChat, a social media platform used by over 90% of mobile phone users in China, could bridge this "training gap" through a structured, asynchronous curriculum.

Methods

The researchers conducted a pre-and-post intervention study (a "self-control" design) over one year in Zhongmu County, Henan Province.

  • The Intervention: They created a WeChat Subscription Account (SA) titled "TB control commune of village doctors." Every week for a year, they released original content covering TB detection, treatment, prevention, and patient management.
  • Multimedia Approach: Content was delivered in various formats, including text articles, posters, videos, and cartoons, to test which formats resonated best with the audience.
  • Evaluation: Participants took an offline, 30-question test before the training began and the same test one year later. The test included both single-choice and more difficult multiple-choice questions, with scores normalized to a 100-point scale.

Datasets

The "dataset" in this study consists of the demographic and performance data of the participating healthcare workers:

  • Study Population: Initially, 594 village doctors participated; 467 completed both the pre-test and post-test.
  • Demographics: The cohort was predominantly male (73%), aged 40–60, with most having only a high school education.
  • Testing Instrument: A 30-question self-administered questionnaire (9 single-choice, 21 multiple-choice) divided into four domains: general TB knowledge, detection/treatment, patient care, and Latent TB Infection (LTBI).

Key Insights and Ideas

  • The "Engagement-Knowledge" Paradox: While the training was successful, the average reading time per article was quite low (144 views on average for a group of nearly 500). This suggests that even modest engagement with high-quality digital content can lead to measurable knowledge gains.
  • Format Matters: Text-based articles had the highest average reading times (245), followed by videos/cartoons (199), while posters were the least engaged (38). This indicates that rural practitioners may prefer detailed text or engaging video over static summaries.
  • Demographic Sensitivity: Education level was the strongest predictor of high scores. Interestingly, after adjusting for other factors, female doctors showed a higher rate of improvement than their male counterparts, suggesting that mHealth interventions might be particularly effective for certain demographic subgroups.
  • LTBI: The Weakest Link: At the start of the study, knowledge regarding Latent TB Infection (LTBI) was dismal (median score of 20/100). Although it doubled to 40/100 by the end, it remained the lowest-scoring category, highlighting a major area for future educational focus.

Key Results

  • Overall Improvement: The median knowledge score increased from 50 to 60 (p < 0.001). While a 10-point jump may seem modest, it represents a 20% increase in baseline knowledge across a large population.
  • Domain Gains: Significant improvements were seen across all four tested areas. General knowledge of TB rose from 71.4 to 85.7, while detection and treatment knowledge rose from 50.0 to 62.5.
  • Retention: Despite the year-long duration, the majority of doctors (78.6%) stayed with the program, though those who dropped out were more likely to be female or have no prior history of managing TB patients.

Limitations and Future Work

The authors acknowledge several hurdles:

  • Passive Learning: The intervention relied on "self-learning willingness." If a doctor didn't click the link, they didn't learn. Future iterations might include "push" notifications or gamified elements to increase engagement.
  • The "Ceiling" of Knowledge: A 20% increase is a start, but a median score of 60 still leaves significant room for error in clinical practice.
  • Follow-up Bias: About one-fifth of the participants dropped out before the final test, which may have slightly skewed the results toward more motivated individuals.

Future Direction: The researchers suggest that future mHealth programs should explore more interactive methods—such as quizzes within the app or shorter, punchier content—to further improve engagement and knowledge retention among rural health workers.

Summary updated

Abstract

<title>Abstract</title> <bold>Background:</bold> In China, around 70% tuberculosis (TB) cases occurred in rural areas. Village doctors played a crucial role in TB cases referral and management. The current study aimed to evaluate the effect of WeChat-based training program on improving the village doctors’ knowledge on TB management.<bold>Methods:</bold> A<bold> </bold>One-year WeChat-based training was conducted by means of releasing original contents (in forms of text, poster, video or cartoon), through WeChat subscription account (WeChat SA) once a week, with the topics of TB detection, treatment, prevention and patient’s management. Based on a self-control design, pre-and post-training offline tests were hold using the same self-administered questionnaire to evaluate the training effect.<bold>Results</bold>: A total of 467 village doctor registered in Zhongmu County (Henan province, China), who worked in the local village clinic or community health center, were included in the study. During the training, sixty original articles were posted through WeChat SA during the one-year training. There are 19, 26 and 15 were displayed in the form of text, poster and other types (such as video or cartoon), and the corresponding average reading times were 245, 38, 199, respectively. With respect to the two tests, the median score increased from 50 (40.0-60.0) to 60 (53.0-70.0) (p&lt;0.001) after training. The median scores were always higher after training no matter what kind of article contents and forms. In addition, the effectiveness of such training was found to be influenced by gender and education level.<bold>Conclusions: </bold>Our results indicated that WeChat-based training could improve the knowledge of village doctors on TB management. It is worthy to explore more effective new media-based training methods that might promote TB control work in rural China.
BibTeX
@inproceedings{gao2020effect,
  author = {Lei Gao and Henan Xin and Haoran Zhang and Xuefang Cao and Boxuan Feng and Zhusheng Quan and Ying Du and Yijun He and Dakuan Wang and Bin Zhang and Qi Jin and Zisen Liu and Shouguo Pan and Ling Guan and Fei Shen and Jianmin Liu},
  title = {{The effect of WeChat-based training on improving the village doctors’ knowledge of tuberculosis management: a pre-and-post intervention study from China}},
  booktitle = {Research Square},
  year = {2020},
  doi = {10.21203/rs.3.rs-92812/v1},
}

Loading related content...

Figure