according to a retrospective analysis of hospitalization data from a tertiary care center.
Dr. Allen Anandarajah
“The identification of the HNHC [high-need, high-cost] cohort and the risk factors for hospitalizations for this cohort will help pave the way to develop programs that improve the quality of care for high-risk lupus patients and [at the same time] lower the cost of care for all lupus patients,” first author Allen Anandarajah, MBBS, and colleagues at the University of Rochester (N.Y.) wrote in Arthritis Care & Research.
Hospitalizations and readmissions are known to be common in patients with SLE, the authors said, and they “account for a large proportion of the direct costs associated with the care of this disease.”
“While HNHC cohorts have been described with other chronic diseases, this report is the first to describe the existence of such a cohort in the SLE population,” the researchers said.
To see if a small group of SLE patients would constitute the majority of hospitalizations and consequently the costs of such care, Dr. Anandarajah and associates analyzed data from 202 SLE patients and their 467 hospitalizations at the University of Rochester–affiliated Strong Memorial Hospital during July 1, 2013, to June 30, 2016. The patients had a mean age of 46 years and included 183 females. A total of 46.5% were White, 43.1% were African American, 6.9% were Hispanic, and 3.5% were of Asian descent. These patients had median lengths of stay of 7 days per SLE patient and 4 days per admission, with median costs of $19,271 per patient and $14,375 per admission.
The researchers identified 44 patients (22%) who accounted for 275 admissions (59%) during the 3-year period. This group’s median of 4 admissions per patient was significantly higher than the median of 1 recorded in all the other hospitalized SLE patients, as was its number of readmissions within 30 days (105 total and median of 1 vs. 11 total and median of 0). The high-risk SLE patients spent a significantly greater amount of time in the hospital than did other patients (median of 30 days vs. 5 days), and their median cost was more than six times as great ($95,262 vs. $14,360). High-risk patients’ median cost per admission also was significantly greater ($19,376 vs. $12,833).
Infections were the most common cause of hospitalization among both high-risk patients and others (28% vs. 23%, respectively) and the rate of involvement of different organ systems as a cause for hospitalization were similar between the groups, except that patients at lower risk significantly more often had gynecologic/obstetric concerns (10% vs. 2%) or nervous system involvement (16% vs. 5%), and high-risk patients were significantly more likely to have gastrointestinal complaints (20% vs. 8%).
Clinically, high-risk patients had significantly higher median scores on the Systemic Lupus International Collaborating Clinics Damage Index and the Comorbidity Index, as well as a significantly higher median level of double-stranded DNA. However, they had no differences in complement factor levels or body mass index.
The high-risk patients also were younger (mean of 42 vs. 46 years) and were diagnosed at a younger mean age (26 vs. 31 years). More high-risk patients were African American (55% vs. 40%) and were more likely to live in areas identified with poverty (50% vs. 29%).
A multivariate analysis that controlled for relevant confounders showed that high-risk patients had a 10 percentage point lower medication possession ratio, which is an indicator of whether a patient had adequate medication supply in a given time frame. High-risk patients overall had a higher average number of medications to treat lupus.
“Our findings underscore the importance of identifying HNHC SLE patients when designing and implementing interventions to lower hospitalizations and improve the quality of care for lupus patients. Furthermore, it is imperative that we develop programs to address the modifiable social and behavioral factors in addition to providing high-quality clinical care targeted for this group,” the researchers wrote.
Some of the limitations in the generalizability of the results include the use of data from a large tertiary medical center serving a large catchment area, with a consequently sicker group of patients, and the potential to miss readmissions to other nearby hospitals. However, “as one of the few centers [in the region] that provides in-patient rheumatology care ... it is less likely that patients would have sought care elsewhere,” they noted.
The study involved no outside source of funding, and the authors had no relevant conflicts of interest.
SOURCE: Anandarajah A et al. Arthritis Care Res. 2020 Nov 17. doi: 10.1002/acr.24510.