Stroke Administrative Severity Index: using administrative data for 30-day poststroke outcomes prediction

Area of Science:

• Health Services Research
• Biostatistics
• Epidemiology

Background:

• Current stroke severity scales are not suitable for retrospective analysis of archival data.
• There is a need for a reliable measure of stroke severity at hospital discharge that can be derived from administrative or billing records.

Purpose of the Study:

• To develop and validate the Stroke Administrative Severity Index (SASI) for assessing stroke severity using hospital discharge billing data.
• To enable risk adjustment in administrative data analyses for stroke outcomes.

Main Methods:

• Utilized the NIH Stroke Scale (NIHSS) as a framework to identify 285 relevant ICD-9 codes.
• Employed cluster analysis to group codes into 23 indicator variables.
• Modeled 30-day postdischarge mortality and hospice discharge risk using Medicare data, validating on separate cohorts.

Main Results:

• The developed SASI model demonstrated good fit and strong concordance (C-statistic = 0.76-0.83).
• The SASI effectively predicted NIHSS scores at discharge (C = 0.83).

Conclusions:

• The SASI provides a validated method for assessing stroke severity from administrative data at hospital discharge.
• The SASI serves as a valuable tool for risk adjustment in studies analyzing postdischarge outcomes.