Predictors of Healthcare Spending Identified in Patients With Gout Using Urate-lowering Therapy

Female patient in office signing paperwork across from doctor

Identifying patients with gout who could benefit from interventions to reduce long-term spending has become a central priority for U.S. healthcare providers and insurers. The prevalence of gout is escalating, and it is highly comorbid with cardiometabolic diseases and other conditions that increase the risk of adverse outcomes.

Using routinely collected claims data, researchers at Brigham and Women’s Hospital identified three distinct groups of gout patients according to their patterns of total spending on health care. Julie C. Lauffenburger, PharmD, PhD, an epidemiologist in the Division of Pharmacoepidemiology and Pharmacoeconomics, Seoyoung C. Kim, MD, ScD, formerly a rheumatologist in the division, and colleagues provide specific information in Arthritis Care & Research (Hoboken) that could be useful for designing cost-containment interventions.


The data source for the study was Optum’s Clinformatics Data Mart of commercially insured individuals in all 50 states. The analysis included patients with gout who:

  • Had one year of baseline data (January 1 to December 31, 2017) and two years of follow-up data (January 1, 2018, to December 31, 2019)
  • Were ≥40 years old on January 1, 2018
  • Filled at least one prescription for urate-lowering therapy in 2017

57,980 gout patients met these criteria (70% male; mean age 71). The researchers recorded 60 baseline variables for each patient, including demographics, healthcare utilization, medication use and adherence, healthcare costs, and gout-related comorbidities.

The team also measured monthly healthcare spending for each patient (all inpatient care, outpatient care, and prescription drug claims) during the follow-up period.

Three Spending Patterns

Three clusters of patients were identified who showed similar spending patterns on health care over time:

  • Minimal spending ($0–$5 per month) (13.2% of patients)
  • Moderate spending (37.4%)
  • High spending (49.4%)

Predictive Factors

When all 60 baseline variables were considered, predictors of being in the high-spending group were:

  • Total spending in the baseline year
  • Number of medications, out-of-pocket spending, and level of insurance benefits during the follow-up years

When only gout-related comorbidities and medication use were considered, predictors associated with increased odds of assignment to the spending group included:

  • Higher adherence to urate-lowering therapy (OR for each 1-unit change in the proportion of days covered, 1.31; 95% CI, 1.07-1.61)
  • Use of glucocorticoids (OR, 1.26; 95% CI, 1.11-1.42)
  • Diabetes (OR, 1.18; 95% CI, 1.02-1.38)
  • Having renal stones (OR, 1.69; 95% CI, 1.44-1.98)

The use of nonsteroidal anti-inflammatory drugs or coxibs was associated with lower odds of membership in the high-spending group (OR, 0.78; 95% CI, 0.71-0.87).

Developing Targeted Interventions

Being able to discriminate between patients with gout who fall in these clusters could help healthcare systems target population-level health management programs to those at greatest need. The predictor variables identified in this study, such as higher medication burden and concomitant diabetes, could also be key areas for the development of cost-containment interventions.

Leave a Reply