Session Information
Date: Sunday, November 8, 2015
Session Type: ACR Poster Session A
Session Time: 9:00AM-11:00AM
Background/Purpose:
Care fragmentation is associated with poor health outcomes and is hard to assess in a single-site data registry. The Chicago HealthLNK Data Repository (CHDR) is a city-wide database of de-duplicated patient Electronic Health Record data from 6 institutions in Chicago, with the capability to match patient data from outside registries to supplement registries and provide a multi-institution view of a patient population. The Chicago Lupus Database (CLD) is a registry of 880 patients meeting ≥3 of the 1982/1997 revised ACR classification criteria for SLE. CLD participant data was combined with CHDR to understand factors influencing fragmentation.
Methods:
Using a HIPAA compliant algorithm1, patients from the CLD were hashed and matched into the CHDR to create a comprehensive set of patient visits from all institutions. Patients’ demographics, insurance, fragmentation (defined as having ≥ 1 visit at ≥ 2 institutions) and visit data were pulled from the CHDR. Differences between groups were computed using chi-squared tests and odds ratios. Multivariate regression was used to assess factors contributing to percent inpatient visits and probability of death.
Results:
625 patients from the CLD matched with a CHDR patient having at least 2 visits from 2006-2012. 92% were female; 52% white and 32% black; 65% private and 21% public insurance; 29% had fragmented care. Sex had no effect on fragmentation. Blacks were more likely to have fragmented care than whites (OR: 2.52 CI: 1.68,3.79). Fragmented care was increased for both blacks (OR: 2.81 CI: 1.44,5.50) and whites (OR: 2.01 CI: 1.09,3.68) on public compared to private insurance. Overall, 6% of patient visits were inpatient. Table 1 shows percent change in inpatient visits for categories of fragmentation, race, and insurance; in the multivariate model, fragmentation, black race, public and no insurance all contributed to higher percent of inpatient visits. For each percent increase in inpatient visits, the odds of death increase by 1.05 (CI: 1.03,1.07).
Conclusion:
By combining a disease specific registry with a city-wide database, this study examines factors contributing to care fragmentation among SLE patients, and whether fragmented care impacts health outcomes in patients with a complicated chronic illness. For blacks and whites, public insurance correlated with fragmented care and increased inpatient visits, which is suggestive of more serious health impacts due to SLE. With the expansion of public insurance, patients with chronic illness seeking care at multiple sites may need improved care coordination or access to primary care to avoid more severe situations, hospitalization and increased likelihood of death.
Reference:
1. Kho AN, Cashy JP, Jackson KL et al. Design and Implementation of a Privacy Preserving Electronic Health Record Linkage Tool in Chicago. JAMIA 2015.
Table 1. Change in Percent of Inpatient Visits
Group |
Percent Change |
95% CI |
p-value |
Race |
Ref=White |
||
Black |
3.7 |
(1.6,5.9) |
<0.001 |
Other |
-0.4 |
(-2.9,2.1) |
0.76 |
Insurance |
Ref=Private |
||
Public |
3.8 |
(1.8,5.9) |
<0.001 |
No Insurance Reported |
9.7 |
(4.9,14.6) |
<0.001 |
Fragmentation |
Ref=Not Fragmented |
||
Fragmented |
2.2 |
(0.2,4.2) |
0.03 |
Combinations |
|||
Race and Insurance |
Ref=White, Private |
||
Blacks w/ Public Insurance |
7.6 |
(5.1,10.1) |
<.0001 |
Whites w/ Public Insurance |
3.8 |
(1.8,5.9) |
<0.001 |
Race and Fragmentation |
Ref=White, No Fragmentation |
||
Blacks w/ Fragmentation |
5.9 |
(3.2,8.6) |
<.0001 |
Whites w/ Fragmentation |
2.2 |
(0.2,4.2) |
0.03 |
Race, Insurance, and Fragmentation |
Ref= White, Private, No Fragmentation |
||
Blacks w/ public and fragmentation |
9.8 |
(7.0,12.6) |
<.0001 |
Whites w/ public and fragmentation |
6.0 |
(3.4,8.7) |
<.0001 |
To cite this abstract in AMA style:
Jackson K, Walunas T, Chung A, Ramsey-Goldman R. Utilizing City-Wide Electronic Health Record Data to Assess Care Fragmentation in Patients with Systemic Lupus Erythematosus (SLE) [abstract]. Arthritis Rheumatol. 2015; 67 (suppl 10). https://acrabstracts.org/abstract/utilizing-city-wide-electronic-health-record-data-to-assess-care-fragmentation-in-patients-with-systemic-lupus-erythematosus-sle/. Accessed .« Back to 2015 ACR/ARHP Annual Meeting
ACR Meeting Abstracts - https://acrabstracts.org/abstract/utilizing-city-wide-electronic-health-record-data-to-assess-care-fragmentation-in-patients-with-systemic-lupus-erythematosus-sle/