Capturing Information About Multiple Sclerosis Comorbidity Using Clinical Interviews and Administrative Records: Do...

Area of Science:

• Neurology
• Epidemiology
• Health Informatics

Background:

• Multiple sclerosis (MS) is frequently accompanied by comorbidities impacting patient outcomes.
• Accurate comorbidity data is crucial but challenging to obtain in large epidemiological studies.
• Current data sources include self-reports, medical records, and administrative databases, each with limitations.

Purpose of the Study:

• To evaluate the agreement between clinical interviews and administrative records for identifying comorbidities in people with MS (pwMS).
• To assess the accuracy of administrative data in capturing common comorbidities in the MS population.

Main Methods:

• A comparative analysis of clinical interview data and administrative records was conducted in pwMS at two study sites.
• Seven key comorbidities were investigated: depression, anxiety, diabetes, hypertension, autoimmune disease, chronic lung disease, and hyperlipidemia.
• Agreement was quantified using kappa statistics, sensitivity, specificity, and predictive values.

Main Results:

• Administrative data showed excellent agreement for diabetes (κ=98.9%) and hypertension (κ=89.8%).
• Sensitivity was high for diabetes (80%) and moderate for hypertension (62%).
• Administrative records frequently underestimated autoimmune diseases, hyperlipidemia, and anxiety, while over-reporting depression.

Conclusions:

• Clinical interviews and administrative data demonstrate excellent agreement for diabetes and hypertension in pwMS.
• Administrative data exhibits lower sensitivity and potential misclassification for psychiatric, autoimmune, and hyperlipidemia comorbidities.
• Careful consideration of data source limitations is necessary when utilizing administrative records for epidemiological research in MS.