Abstract: TH-PO1171
When Prognosis Is Known, Frailty Fades: REIN Outperforms Age and Frailty in Dialysis Decision-Making
Session Information
- Geriatric Nephrology: Advancing Care for Older Adults
October 22, 2026 | Location: Exhibit Hall A, Convention Center
Abstract Time: 10:00 AM - 12:00 PM
Category: Geriatric Nephrology
- 1300 Geriatric Nephrology
Authors
- Roy, Debajyoti M., Changi General Hospital, Singapore, Singapore
- Wilkinson, Stephen P., Changi General Hospital, Singapore, Singapore
- Koduri, Sreekanth, Changi General Hospital, Singapore, Singapore
- Yeon, Wenxiang, Changi General Hospital, Singapore, Singapore
- Lokanathan, Tharshini, Changi General Hospital, Singapore, Singapore
- Tiang Guan Bernard, Koh, Changi General Hospital, Singapore, Singapore
Background
Dialysis decisions in older adults with advanced CKD are often driven by chronological age despite heterogeneity in biological vulnerability. Whether incorporating frailty and individualized mortality risk into shared decision-making (SDM) improves clinical relevance remains unclear. This is the first prospective study integrating both into SDM in this population.
Methods
We conducted a prospective, single-center cohort study of adults ≥75 years with CKD stages 4–5 referred for dialysis decision-making. All underwent comprehensive geriatric assessment (Clinical Frailty Scale [CFS], Barthel Index, AMT, Charlson Comorbidity Index) and 3-month mortality risk estimation using the REIN score, disclosed during a joint geriatrician–nephrologist consultation.
Primary outcome was treatment selection (dialysis vs maximum medical therapy [MMT] analyzed using univariate and multivariate logistic regression. Secondary outcome was 3-year mortality (Cox regression). Patient-reported experience measures (PREMs) assessed shared decision-making quality.
Results
Among 111 participants (mean age 81±5 years), 30% died in 3 years.
On univariate analysis, age, frailty (CFS), and REIN score were all significantly associated with MMT selection (all p<0.01).
In multivariate models, age and REIN remained independently associated with MMT (both p<0.001) and demonstrated superior model performance compared with age plus frailty. When all three variables were included, frailty lost statistical significance, while age and REIN remained robust predictors.
For mortality, both frailty and REIN independently predicted 3-year outcomes. PREMs demonstrated high patient understanding, involvement, and satisfaction.
Conclusion
In older adults with advanced CKD, individualized mortality risk (REIN), not frailty, drives dialysis decisions alongside age. Frailty predicts survival but has limited impact on treatment choice after accounting for prognostic risk.
When prognosis is explicit, risk—not frailty—drives decisions.
Table1. Multivariable models for selection of Maximum Medical therapy (MMT)
| Model | Variables Included | Significant Predictors | OR (95% CI) | p-value | C-Statistics |
| Model 1 | Age | Age | 1.17 (1.07 – 1.26) | <0.001 | 0.72 |
| Model 2 | Age + Frailty | Age, Frailty | Age: 1.20 (1.09 – 1.32) Frailty: 8.16 (2.30 – 28.9) | <0.001 | 0.80 |
| Model 3 | Age + REIN | Age, REIN | Age: 1.22 (1.10 – 1.35) REIN: 1.35 (1.16 – 1.57) | <0.001 | 0.82 |
| Model 4 | Age+Frailty +REIN | Age, REIN | Age: 1.22 (1.11 – 1.36) REIN: 1.27 (1.08 – 1.51) | <0.001 | 0.83 |
Note: Highest discrimination for MMT is achieved when REIN score is included (Model 3 & 4)