Multi-component Stroke Intervention and Long-term Biofunctional Outcomes: A Secondary Analyses of the SINEMA Trial

Bolu Yang, Xiaoyan Yan, Zhi Zheng, Fei Wu, Xiong Ding, Xingxing Chen, Brian Oldenburg, Haizhu Song, Yiying Zhou, Liebin Zhao, Bifeng Yuan, Lijing L Yan, Gong EnyingmedRxiv 2026 · DOI

TL;DR

A one-year digital health and primary care intervention for stroke patients in rural China led to significantly lower cumulative blood pressure (BP) burden over a seven-year follow-up period. Intriguingly, while higher cumulative BP was strongly linked to functional and cognitive decline in the control group, this association was largely "decoupled" in the intervention group, suggesting the program may have conferred long-term resilience against vascular-related deterioration.

What Problem Does This Paper Address?

Stroke is a global leading cause of long-term disability, often resulting in a "biofunctional decline" that includes physical disability and cognitive impairment. While the link between high blood pressure and poor post-stroke outcomes is known, most research relies on "snapshot" BP measurements at single time points. These snapshots fail to capture the long-term, cumulative burden of BP exposure.

Furthermore, while short-term interventions can improve BP control, we have limited evidence on whether these benefits persist years later or if they actually change the underlying relationship between BP levels and functional health in resource-limited settings like rural China.

Methods

This study is a post-hoc secondary analysis of the SINEMA trial, a cluster-randomized controlled trial.

  • The Intervention: A 12-month program involving a "system-integrated and technology-enabled" model. Village doctors were trained and equipped with a dedicated app to conduct monthly follow-ups, while patients received daily voice messages regarding medication adherence and physical activity.
  • Long-term Tracking: Researchers followed up with participants at 72 and 84 months (6-7 years) post-baseline.
  • Cumulative BP Calculation: Instead of using single readings, the authors used the trapezoidal rule to calculate the Area Under the Curve (AUC) for systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP), and pulse pressure (PP). This provided a "mean annual cumulative BP," representing the average long-term burden.
  • Biofunctional Metrics: Outcomes included health-related quality of life (EQ-5D-5L), disability (modified Rankin Scale), activities of daily living (ADL), physical function (grip strength and "Timed Up and Go" test), and cognitive function (CSI-D).

Datasets

The study utilized data from the SINEMA trial, which recruited 1,299 adults with stroke from 50 rural villages in Hebei, China, between 2017 and 2018. The final analysis focused on 897 participants who had complete data across all assessment points up to 84 months.

Key Insights and Ideas

  • The "Legacy Effect": A relatively short 12-month intervention produced a "legacy effect" on blood pressure that was still detectable seven years later. The intervention group maintained lower cumulative BP levels long after the active phase ended.
  • Modification of Risk (The "Decoupling" Effect): The most striking finding wasn't just lower BP, but a change in the relationship between BP and health. In the control group, higher cumulative BP was a death sentence for functional health (strongly associated with decline). In the intervention group, this association was significantly attenuated or non-existent.
  • Resilience through Care: The authors hypothesize that the intervention might have improved "health literacy" or "self-management skills" that protected patients. Even if their BP rose later, they might have been better equipped to manage their overall health, conferring a form of "biofunctional resilience."
  • Cumulative BP as a Superior Metric: The study reinforces that cumulative BP is a more informative indicator for long-term stroke management than isolated measurements, as it better reflects the sustained vascular burden on the brain and body.

Key Results

  • BP Reduction: Compared to usual care, the intervention group showed significantly lower mean annual cumulative:
  • SBP: −2.2 mm Hg (95% CI, −3.9 to −0.6)
    • DBP: −1.6 mm Hg (95% CI, −2.4 to −0.7)
    • MAP: −1.8 mm Hg (95% CI, −2.8 to −0.8)
  • Interaction Effects: Significant interaction effects were found between the trial arm and cumulative SBP for seven different outcomes, including quality of life and disability.
  • Functional Outcomes: While the absolute levels of functional decline were similar between the two groups at 84 months, the intervention fundamentally changed how BP burden translated into that decline.

Limitations and Future Work

  • Post-hoc Nature: As a secondary analysis, the study was not originally designed to test these specific hypotheses, meaning the results should be viewed as exploratory.
  • Survivor Bias: Biofunctional outcomes could only be measured in participants who survived to the 72- and 84-month marks, which might underestimate the true association between BP and decline.
  • The "Black Box" of Mechanisms: While the study shows that the relationship changed, it doesn't definitively explain why. Future work is needed to investigate the specific mediating pathways—such as improved diet, exercise, or social support—that might be driving this resilience.
  • Baseline Gaps: Some functional measures (like grip strength and cognition) were not recorded at the very start of the trial, preventing a true "baseline to end-line" comparison for those specific metrics.

Summary updated

Abstract

Abstract Background The one-year SINEMA trial demonstrated improved blood pressure (BP) control and reduced mortality up to 72 months after the intervention. This article aims to assess between-arm differences in mean annual cumulative BP and to explore whether the associations between cumulative BP and biofunctional outcomes differed by trial arm. Methods Post-hoc secondary analysis of the SINEMA cluster-randomized trial, which recruited 1299 adults with stroke from 50 rural villages in Hebei, China, between 2017 and 2018. The 12-month intervention was followed by observational assessments at 72 and 84 months post-baseline. BP was measured during each face-to-face follow-up, assessed by blinded assessors at baseline, 12, 72, and 84 months. Mean annual cumulative systolic BP (SBP), diastolic BP (DBP), mean arterial pressure (MAP), and pulse pressure (PP) were calculated. Biofunctional outcomes included health-related quality of life, modified Rankin Scale, activities of daily living, physical function, and cognition function. Results Among 897 participants (mean age 62.7 years; 40.8% female) with complete data across all assessment, the intervention arm demonstrated significantly lower mean annual cumulative SBP (−2.2 mm Hg; 95% CI, −3.9 to −0.6), DBP (−1.6 mm Hg; 95% CI, −2.4 to −0.7), and MAP (−1.8 mm Hg; 95% CI, −2.8 to −0.8), not PP, compared with usual care. Significant associations between cumulative BP and biofunctional outcomes were observed in the control arm while not in the intervention arm. Interaction effects between trial arm and cumulative BP were significant for multiple outcomes, most prominently for cumulative SBP. Conclusions The one-year SINEMA intervention was associated with lower cumulative BP burden over 72-84 months but did not improve overall biofunctional outcomes. Secondary analyses revealed that the association between cumulative BP burden and biofunctional decline differed by intervention arm, suggesting cumulative BP exposure may be an important long-term risk indicator and the intervention may modify BP-outcome relationships through mechanisms requiring further investigation.
BibTeX
@misc{yang2026multi,
  author = {Bolu Yang and Xiaoyan Yan and Zhi Zheng and Fei Wu and Xiong Ding and Xingxing Chen and Brian Oldenburg and Haizhu Song and Yiying Zhou and Liebin Zhao and Bifeng Yuan and Lijing L Yan and Gong Enying},
  title = {{Multi-component Stroke Intervention and Long-term Biofunctional Outcomes: A Secondary Analyses of the SINEMA Trial}},
  year = {2026},
  doi = {10.64898/2026.01.28.26345092},
  howpublished = {medRxiv},
}
Figure