Who Benefits from Telemonitoring? Uplift Modelling of a Nationwide Home Blood Pressure Monitoring Programme
Organizations: MOH Office for Healthcare Transformation, Singapore
Abstract
Home blood pressure (BP) telemonitoring programmes are being deployed at national scale, yet enrolment resources are finite and the benefit of telemonitoring is unlikely to be uniform across patients. We study the Primary Technology-Enhanced Care for Hypertension (PTEC-HT) programme, a nationwide telemonitoring programme for hypertensive patients in Singapore. Using retrospective data on 16,694 patients registered between August 2020 and May 2024, we first estimate the average treatment effect of the programme on 6-month BP control against a propensity-score-matched non-participant cohort, finding an absolute improvement of 5 percentage points in the rate of BP control (74% vs. 69%, ), rising to 12 percentage points (63% vs. 50%, ) among patients with uncontrolled BP at baseline. We then move beyond the average effect and apply uplift modelling to estimate individual treatment effects. Across five uplift learners evaluated with 5-fold cross-validation and Qini curves, an uplift random forest performed best on the full cohort and an S-learner on the uncontrolled-baseline subcohort, although performance bands overlapped across learners. On held-out patients, realised uplift was largest in the top predicted-uplift groups and declined across the ranking. These patients had higher baseline BP grade, more recent hypertension diagnosis and lower medication and diagnosis burden, a profile we offer as hypothesis-generating for targeted enrolment. A handcrafted rule built from these features was competitive with the learned models, outperforming all of them on the uncontrolled-baseline cohort, suggesting that uplift models are useful for discovering heterogeneity that can be distilled into simple enrolment criteria for directing limited care team capacity.
Figures & tables
| Full | Uncontrolled BP | |
| Matched pairs | 7,278 | 2,053 |
| PTEC-HT, baseline | 70.0% | 0.0% |
| Control, baseline | 70.0% | 0.0% |
| PTEC-HT, 6 months | 74.2% | 62.6% |
| Control, 6 months | 69.0% | 50.4% |
| ATE (pp) | +5.2 | +12.2 |
| Full cohort | Uncontrolled BP | |||
|---|---|---|---|---|
| Model | CV | Test | CV | Test |
| S-learner | 0.30 0.08 | 0.371 | 0.11 0.06 | 0.335 |
| T-learner | 0.32 0.08 | 0.342 | 0.10 0.05 | 0.171 |
| X-learner | 0.32 0.06 | 0.333 | 0.09 0.03 | 0.230 |
| R-learner | 0.31 0.05 | 0.324 | 0.09 0.08 | 0.214 |
| Uplift forest | 0.41 0.08 | 0.520 | 0.08 0.04 | 0.281 |
Appendix figures & tables9 assets
Supplementary material from the paper’s appendix.
Appendix
| SMD | ||
| Covariate | Full | Uncontrolled BP |
| Age | 0.007 | 0.012 |
| Male | ||
| Years since diagnosis | 0.000 | 0.011 |
| Total diagnoses | 0.007 | |
| Total prescriptions | 0.016 | |
| SMD | ||
| Covariate | Full | Uncontrolled BP |
| Age | ||
| Male | 0.333 | 0.243 |
| Years since diagnosis | ||
| Total diagnoses | ||
| Total prescriptions | ||
| Feature | Definition |
|---|---|
| Age | Age at PTEC-HT registration date (treated) or assigned observation date (control) |
| Gender | Male vs. female |
| Years since diagnosis | No. of years from first coded hypertension diagnosis to observation date |
| No. hypertension diagnoses | Total number of coded hypertension diagnosis events prior to observation date |
| No. antihypertensive prescriptions | Total number of antihypertensive medication prescriptions prior to observation date, defined via ATC codes for diuretics (C03), beta blockers (C07), calcium channel blockers (C08) and RAAS agents/ARBs (C09) including single- and multi-agent combination products |
| BP grade (3 months) | Mean clinic BP over the 3 months preceding observation, graded: Normal, High-normal, Grade 1, Grade 2, Grade 3 |
| Model | Full cohort | Uncontrolled BP cohort |
|---|---|---|
| S-learner | Logistic regression ( ) | Random forest ( , , , ) |
| T-learner | Logistic regression ( ) | Logistic regression ( ) |
| X-learner | Outcome: SGD classifier (penalty= , loss=log loss, ) Effect: linear regression | Outcome: SGD classifier (penalty= , loss=log loss, ) Effect: elastic net ( , ) |
| R-learner | Outcome: SGD classifier (penalty= , loss=log loss, , random_state=seed) Effect: linear regression | Outcome: SGD classifier (penalty= , loss=log loss, , random_state=seed) Effect: elastic net ( , , random_state=seed) |
| Uplift random forest | , , , , | , , , , |
| SMD | ||
|---|---|---|
| Covariate | Train | Test |
| Male | 0.050 | 0.078 |
| Age | 0.231 | 0.069 |
| Years since diagnosis | 0.395 | 0.056 |
| BP grade (3-month) | 0.139 | 0.053 |
| BP grade (1-year) | 0.113 | 0.089 |
| (1) | (2) | (3) | |
|---|---|---|---|
| (1) Years since diagnosis | 1.00 | ||
| (2) Total diagnoses | 0.82 | 1.00 | |
| (3) Total prescriptions | 0.74 | 0.79 | 1.00 |