Background/Purpose: Systemic Lupus Erythematosus (SLE) is a chronic auto-immune disease resulting in significant excess morbidity and health care utilization. Since time of disease onset is often unknown, we hypothesized that there is an excess of health care utilization prior to diagnosis of SLE cases (but after disease onset).

Methods: We identified a cohort of incident cases of SLE among individuals documented in the Clinical Practice Research Datalink (CPRD) and linked with Hospital Episode Statistics (HES). CPRD is a database of anonymized longitudinal medical records from primary care from over 600 practices in the United Kingdom. HES is a database documenting all admissions to the National Health System hospitals in England that can be linked with CPRD since 1997. All individuals documented in GPRD and HES prior to October 1st, 2010 and aged 18 years or older were retained in the analysis.

Patients with SLE were identified as individuals with at least one relevant diagnosis code in CPRD or HES (list of codes available upon request).

Incident cases were defined as patients with at least 12 months of registration in CPRD prior to the date of first diagnosis. All incident SLE cases were matched with up to 5 controls registered in CPRD and linked with HES at the time of and during the one year prior to first diagnosis of the SLE case, and matched by age, gender, and practice. The index date for the controls is the index date (or date of first diagnosis) of the matched case.

Results: The proportions of individuals in the controls group who were hospitalized at least once, had an encounter with a health care practitioner (HCP) or were treated with corticosteroids during each 6 month period over the 3 years prior to the index date remained relatively stable (Table 1).

The proportions of SLE cases who were hospitalized at least once or were treated with corticosteroids per 6 month period grew from 14% to 26% and from 20% to 33%, respectively, over the 3 years prior to the date of incident diagnosis. The proportion of patients who had at least one encounter with a HCP grew also from 73% to 81%.

There is a significant excess of health care utilization in SLE cases versus matched controls during the 3 years prior to the date of incident diagnosis. The excess was still significant between M.-36 and M.-31, when the proportions of individuals hospitalized at least once and treated with corticosteroids were twice as high among SLE cases than in matched controls: 14% versus 7% and 20% versus 10%, respectively.

Table 1: Health care utilization three years prior to diagnosis of SLE or prior to index date of the matched controls

Matched Controls	M.-36/M.-31 n=3,954	M.-30/M.-25 n=4,344	M.-24/M.-19 n=4,752	M.-18/M.-13 n=5,139	M.-12/M.-7 n=5,557	M.6-to D.0 n=5,687
Hospitalization	7%	8%	9%	8%	9%	9%
Treatment with corticosteroids	10%	10%	11%	10%	10%	11%
Encounter with HCP	75%	74%	75%	75%	75%	74%
SLE Cases	M.-36/M.-31 n=973	M.-30/M.-25 n=1,024	M.-24/M.-19 n=1,096	M.-18/M.-13 n=1,126	M.-12/M.-7 n=1,159	M.6-to D.0 n=1,159
Hospitalization	14%	16%	16%	17%	20%	26%
Treatment with corticosteroids	20%	20%	21%	23%	27%	33%
Encounter with HCP	73%	73%	73%	75%	76%	81%

Conclusion: Excess health care utilization prior to incident diagnosis of SLE should be taken into account to assess the total burden of disease. This analysis suggests that there is an excess health care utilization for at least 3 years prior to diagnosis.

---

« Back to 2012 ACR/ARHP Annual Meeting

ACR Meeting Abstracts - https://acrabstracts.org/abstract/excess-health-care-utilization-prior-to-diagnosis-of-systemic-lupus-erythematosus-in-england/