Leveraging Providers’ Preferences to Customize Instructional Content in Information and Communications Technology–Based Training Interventions: Retrospective Analysis of a Mobile Phone–Based Intervention in India
Bibliographic record
Abstract
BACKGROUND: Many public health programs and interventions across the world increasingly rely on using information and communications technology (ICT) tools to train and sensitize health professionals. However, the effects of such programs on provider knowledge, practice, and patient health outcomes have been inconsistent. One of the reasons for the varied effectiveness of these programs is the low and varying levels of provider engagement, which, in turn, could be because of the form and mode of content used. Tailoring instructional content could improve engagement, but it is expensive and logistically demanding to do so with traditional training. OBJECTIVE: This study aimed to discover preferences among providers on the form (articles or videos), mode (featuring peers or experts), and length (short or long) of the instructional content; to quantify the extent to which differences in these preferences can explain variation in provider engagement with ICT-based training interventions; and to compare the power of content preferences to explain provider engagement against that of demographic variables. METHODS: We used data from a mobile phone-based intervention focused on improving tuberculosis diagnostic practices among 24,949 private providers from 5 specialties and 1734 cities over 1 year. Engagement time was used as the primary outcome to assess provider engagement. K-means clustering was used to segment providers based on the proportion of engagement time spent on content formats, modes, and lengths to discover their content preferences. The identified clusters were used to predict engagement time using a linear regression model. Subsequently, we compared the accuracy of the cluster-based prediction model with one based on demographic variables of providers (eg, specialty and geographic location). RESULTS: The average engagement time across all providers was 7.5 min (median 0, IQR 0-1.58). A total of 69.75% (17,401/24,949) of providers did not consume any content. The average engagement time for providers with nonzero engagement time was 24.8 min (median 4.9, IQR 2.2-10.1). We identified 4 clusters of providers with distinct preferences for form, mode, and length of content. These clusters explained a substantially higher proportion of the variation in engagement time compared with demographic variables (32.9% vs 1.0%) and yielded a more accurate prediction for the engagement time (root mean square error: 4.29 vs 5.21 and mean absolute error: 3.30 vs 4.26). CONCLUSIONS: Providers participating in a mobile phone-based digital campaign have inherent preferences for instructional content. Targeting providers based on individual content preferences could result in higher provider engagement as compared to targeting providers based on demographic variables.
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 | 0.004 | 0.014 |
| Meta-epidemiology (narrow) | 0.000 | 0.001 |
| Meta-epidemiology (broad) | 0.001 | 0.001 |
| Bibliometrics | 0.001 | 0.002 |
| Science and technology studies | 0.001 | 0.001 |
| Scholarly communication | 0.001 | 0.001 |
| Open science | 0.001 | 0.001 |
| Research integrity | 0.001 | 0.001 |
| Insufficient payload (model declined to judge) | 0.001 | 0.000 |
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".