Linking Clinical and Administrative Data to Inform Performance Measures Regarding Access to Specialist Care for Patients with Rheumatoid Arthritis

Abstract

Introduction
Rheumatoid arthritis (RA) is the most prevalent type of chronic adult inflammatory arthritis and requires timely diagnosis and subsequent access to specialist care and treatment from a rheumatologist. We developed a set of key performance indicators (KPIs) to evaluate access, effectiveness, acceptability, appropriateness and efficiency of care.

Objectives and Approach
The overall objective was to measure performance of a central intake system for referral to rheumatology against the KPIs. We report on one accessibility KPIs: the percentage of patients with new onset RA with at least one visit to a rheumatologist in the first 365 days since diagnosis.  We identified a cohort of RA patients using a validated case definition: >16 years, at least 1 RA related hospitalization (ICD-10-CA:M05.x-M06.x) or two RA related physician visits ≥ eight weeks apart within two years (ICD-9: 714.x).  The incident case date was date of hospitalization or second physician visit (whichever came first).

Results
This KPI assessed the proportion of patients seen by a rheumatologist within one year of first RA visit by patients in the RA cohort. 13,914 cases of RA were diagnosed between April 1 2010 and March 31 2016. The percentage of patients with new onset RA with at least one visit to a rheumatologist in the first 365 days since diagnosis increased between fiscal years 2011 and 2015. Of the 2851 incident RA cases in fiscal year 2011, 1490 (53%) met the performance measure compared to 1710 of 2710 (63%) who met the definition in fiscal year 2015. Other KPIs, including wait times, are being evaluated using both clinical and administrative data.

Conclusion/Implications
By linking multiple administrative datasets, we are able to measure system performance against a defined KPI and identify opportunities for system improvement. This is the first initiative in Alberta for patients with RA where data from different multi-custodial data repositories have been extracted, linked and analyzed for this purpose.

Introduction

Rheumatoid arthritis (RA) is the most prevalent type of chronic adult inflammatory arthritis and requires timely diagnosis and subsequent access to specialist care and treatment from a rheumatologist. We developed a set of key performance indicators (KPIs) to evaluate access, effectiveness, acceptability, appropriateness and efficiency of care.

Objectives and Approach

The overall objective was to measure performance of a central intake system for referral to rheumatology against the KPIs. We report on one accessibility KPIs: the percentage of patients with new onset RA with at least one visit to a rheumatologist in the first 365 days since diagnosis. We identified a cohort of RA patients using a validated case definition: >16 years, at least 1 RA related hospitalization (ICD-10-CA:M05.x-M06.x) or two RA related physician visits \(\geq\) eight weeks apart within two years (ICD-9: 714.x). The incident case date was date of hospitalization or second physician visit (whichever came first).

Results

This KPI assessed the proportion of patients seen by a rheumatologist within one year of first RA visit by patients in the RA cohort. 13,914 cases of RA were diagnosed between April 1 2010 and March 31 2016. The percentage of patients with new onset RA with at least one visit to a rheumatologist in the first 365 days since diagnosis increased between fiscal years 2011 and 2015. Of the 2851 incident RA cases in fiscal year 2011, 1490 (53%) met the performance measure compared to 1710 of 2710 (63%) who met the definition in fiscal year 2015. Other KPIs, including wait times, are being evaluated using both clinical and administrative data.

Conclusion/Implications

By linking multiple administrative datasets, we are able to measure system performance against a defined KPI and identify opportunities for system improvement. This is the first initiative in Alberta for patients with RA where data from different multi-custodial data repositories have been extracted, linked and analyzed for this purpose.