| Unique ID issued by UMIN | UMIN000060567 |
|---|---|
| Receipt number | R000068230 |
| Scientific Title | Heterogeneous Treatment Effects of Early High-Dose Rehabilitation in Acute Ischemic Stroke: Target Trial Emulation and Causal Forests |
| Date of disclosure of the study information | 2026/04/30 |
| Last modified on | 2026/05/25 15:07:14 |
Heterogeneous Treatment Effects of Early High-Dose Rehabilitation in Acute Ischemic Stroke: Target Trial Emulation and Causal Forests
Heterogeneous Treatment Effects of Early High-Dose Rehabilitation in Acute Ischemic Stroke
Heterogeneous Treatment Effects of Early High-Dose Rehabilitation in Acute Ischemic Stroke: Target Trial Emulation and Causal Forests
Heterogeneous Treatment Effects of Early High-Dose Rehabilitation in Acute Ischemic Stroke
| Japan |
Ischemic Stroke
| Neurology | Neurosurgery | Rehabilitation medicine |
Others
NO
This study aims to operationally define and evaluate an "early high-dose" regimen (initiated <48 hours, duration 7 days) for acute ischemic stroke within the Target Trial Emulation (TTE) framework. In addition to estimating the average causal effect, we will use Causal Forests to identify Conditional Average Treatment Effects (CATE) tailored to patient characteristics.
Focusing on the clinically feasible first 7 days of hospitalization, we will compare dose designs to identify "high-benefit populations" (maximizing benefit while minimizing harm), thereby providing evidence to establish individualized rehabilitation strategies.
Safety,Efficacy
Confirmatory
Not applicable
Barthel Index at discharge
modified Rankin Scale (mRS) score at discharge (including
achievement of mRS 0-3)
Observational
| 18 | years-old | <= |
| Not applicable |
Male and Female
Eligibility criteria were as follows: (1) ischemic stroke (ICD-10: I63*) identified in the DPC data as either the reason for admission or the principal diagnosis; (2) age >= 18 years; (3) pre-admission modified Rankin Scale (mRS) score <= 4; and (4) confirmed stay in a stroke care unit (SCU) on the day of admission (Day 1-SCU). Time zero was defined as the admission date (Day 1). Day 1-SCU was identified using a claims-based proxy definition: the first billing date of an SCU-related management fee in the DPC data coinciding with the admission date.
Exclusion criteria were: (1) cases suggestive of in-hospital onset; (2) cases with clear evidence of long-term care (LTC) residence prior to admission; and (3) cases with an explicit contraindication to rehabilitation immediately after admission (e.g., immediately after decompressive craniectomy, ventricular drainage/external ventricular drain [EVD] placement, initiation of extracorporeal membrane oxygenation [ECMO], or initiation of high-dose vasopressors).
40000
| 1st name | Hirofumi |
| Middle name | |
| Last name | Nagayama |
Kanagawa University of Human Services
Department of Occupational therapy, The School of Rehabilitation
2388522
1-10-1 Heiseicho, Yokosuka, Kanagawa, Japan
046-828-2724
nagayama-ktr@kuhs.ac.jp
| 1st name | Hirofumi |
| Middle name | |
| Last name | Nagayama |
Kanagawa University of Human Services
Department of Occupational therapy, The School of Rehabilitation
2388522
1-10-1 Heiseicho, Yokosuka, Kanagawa, Japan
046-828-2724
nagayama-ktr@kuhs.ac.jp
Mie University
Ministry of Education, Culture, Sports, Science and Technology
Japanese Governmental office
Ethics approval: Not applicable; ethics review was not required for this study.
Ethics approval: Not applicable; ethics review was not required for this study.
Ethics approval: Not applicable; ethics review was
Ethics approval: Not applicable; ethics review was not required for this study.
NO
三重大学
| 2026 | Year | 04 | Month | 30 | Day |
Unpublished
Preinitiation
| 2026 | Year | 02 | Month | 21 | Day |
| 2026 | Year | 02 | Month | 22 | Day |
| 2026 | Year | 03 | Month | 31 | Day |
The aim of this study is, within a target trial emulation (TTE) framework, to operationally define an "early high-dose" rehabilitation regimen initiated within 48 hours and evaluated over the first 7 hospital days among patients with acute ischemic stroke, and to rigorously estimate its average causal effect. In addition, we will use causal forests to identify heterogeneity of treatment effects according to patient characteristics, expressed as conditional average treatment effects (CATE). We focus on a 7-day regimen from admission because it is clinically feasible to implement, and we will compare implementable dose designs and clarify a "high-benefit" subgroup from the perspective of maximizing benefit and minimizing harm, thereby generating evidence to inform personalized rehabilitation strategies.
[Protocol amendment: 2026-05-25]
Following consultation among the co-authors, the analytical scope of this study has been refined to focus on the core effectiveness outcomes, and the secondary outcomes have been reduced to mRS at discharge only.
[Removed secondary outcomes]
- BI shift categories (0-39, 40-59, 60-84, 85-100)
- Achievement of BI >=60 at discharge
- Safety outcomes (in-hospital mortality, new-onset aspiration pneumonia, and trauma/injury)
[Rationale]
This is a retrospective analysis of an existing DPC database, with data collection completed prior to registration. After review of the analytical plan among the co-authors, we decided to focus the methodological scope of this study on the primary effectiveness outcome (discharge BI) and the heterogeneity of treatment effect (HTE) analysis.
[Unchanged]
- Primary outcome (Barthel Index at discharge)
- Study design (Target Trial Emulation + Causal Forest)
- Eligibility criteria and intervention definition
| 2026 | Year | 02 | Month | 03 | Day |
| 2026 | Year | 05 | Month | 25 | Day |
Value
https://center6.umin.ac.jp/cgi-open-bin/ctr_e/ctr_view.cgi?recptno=R000068230