#Newpub#Hotofthepress : A concrete application of our recent paper stressing the importance of assessing reliability of trial outcomes: validation of the simplified mRS for treated brain aneurysm pts! lnkd.in/guqziccC
2/ What previous paper, you say? In case you missed it, please check out why evaluating reliability for RCT outcome measures is as important as for diagnostic tests, if not more: lnkd.in/g4piwx6T
3/ As everyone knows, mRS is widely used for evaluating functional outcome in stroke pts.
A simplified version (smRS) was proposed by Bruno et al. for easier adjudication.
While reliability was extensively studied for stroke pts, it wasn't for patients with treated aneurysms.
4/ Moreover, mRS assessment in RCTs is usually only performed by certified stroke neurologists, which can be costly and of limited availability.
Expanding mRS evaluation to non-trained/certified staff could allow for a wider & more effective use & center recruitment.
5/ We first did a sys review of all Kappa studies of RS showing:
- 7 modif. to initial RS
- Almost all studies were for disab. after stroke
- Higher K ➡️ Ranking Foc. Assessment (RFA) : 0.99
- Lower K ➡️ standard mRS: 0.25
- NO paper assessing mRS reliability after aneurysm Tx
6/ Inter-rater reliability of smRS was then evaluated between:
For 60 patients with clipped📎 or coiled🧶 intracranial aneurysms🎈 from two ongoing RCTs (CURES & ISAT-2).
7/ Almost perfect inter-rater agreement (K>0.8) was reached for all pairs of raters, as well as overall.
Intra-rater was also >0.8.
High level of reliability was maintained when good outcome was defined as smRS=0-2 (K=0.87), but NOT when defined as smRS=0-1 (K=0.59).
8/ Moreover:
Significant difference (P=0.003) in smRS adjudication of good outcome between raters was found when smRS 0-1, while not the case when smRS 0-2 (P=0.52).
9/ That's it! a few take-home points:
1. smRS is a RELIABLE tool for assessing disability in patients with treated brain aneurysms.
2. Defining good outcome as smRS 0-2 is both more reliable AND clinically relevant.
3. Non-trained staff can RELIABLY perform smRS assessments.