Longitudinal comparison of health-related quality of life at 3 and 12 months post-stroke using patient-reported outcome measures.

EQ-5D-3L demonstrated greater discriminatory capacity across mRS groups at 3 months post-stroke compared to PROMIS-10.

• Discriminatory capacity was assessed across modified Rankin Scale (mRS) groups using effect sizes.
• The study analyzed PROMIS-10 and EQ-5D-3L scores in a retrospective cohort of 741 stroke survivors.
• Patients were discharged from a tertiary center and followed at 3 and 12 months post-stroke.
• Both PROMIS-10 and EQ-5D-3L were administered at both time points for longitudinal comparison.

At 12 months, PROMIS-10 Physical Health (PH) demonstrated better discriminatory capacity than EQ-5D-3L for mild disability (mRS 0-1 vs. 2).

• The improved discriminatory capacity of PROMIS-10 PH at 12 months was consistent with ceiling effects reducing EQ-5D-3L sensitivity in the mild impairment group.
• Ceiling effects in EQ-5D-3L were identified as a limitation reducing its sensitivity for mild impairment at 12 months.
• The comparison was made specifically for the mRS 0-1 versus mRS 2 groups.
• This finding supports the use of PROMIS-10 for assessing milder levels of post-stroke disability.

Discharge ambulation consistently predicted both PROMIS-10 Physical Health (PH) and Mental Health (MH) at both 3 and 12 months post-stroke.

• Discharge ambulation was identified through linear regression as a predictor of PROMIS-10 outcomes.
• It was the only predictor consistently associated with both PH and MH at both time points.
• Linear regression was used to identify demographic and clinical predictors of PROMIS-10 PH and MH at 3 and 12 months.
• Discharge-related variables were selected as predictors given their availability at the time of hospital discharge.

Comorbidity burden, employment status, respondent type (proxy vs. patient), and discharge destination were associated with PROMIS-10 scores, particularly for 12-month Mental Health.

• These predictors were identified via linear regression in the retrospective cohort of 741 stroke survivors.
• The association of these variables was particularly pronounced for Mental Health at 12 months.
• Respondent type (proxy vs. patient self-report) was specifically noted as a factor influencing PROMIS-10 scores.
• Discharge destination was included among the discharge-related predictors examined.

Predictors and recovery trajectories differed between the Physical Health and Mental Health domains, supporting their separate assessment.

• The divergence in predictors between PH and MH was observed across both 3-month and 12-month follow-up points.
• EQ-5D-3L cannot separately measure physical and mental health dimensions, which is identified as a limitation.
• PROMIS-10 was used because it provides separate PH and MH subscores via a 10-item short form.
• The finding supports tailored care based on distinct physical and mental recovery trajectories post-stroke.

The study used a retrospective cohort design with 741 stroke survivors discharged from a tertiary center and followed at 3 and 12 months.

• Both PROMIS-10 (10-item short form) and EQ-5D-3L (EuroQoL 5-dimension 3-level) were administered at 3 and 12 months post-stroke.
• Discriminatory capacity across modified Rankin Scale (mRS) groups was assessed using effect sizes.
• Linear regression was used to identify demographic and clinical predictors of PROMIS-10 PH and MH.
• EQ-5D-3L was noted to be widely used but prone to ceiling effects and unable to separate physical and mental health.