SAT-439 Impact on Health Care Resources: A Comparison of Costs Following Treatment with Pegvisomant and Somatostatin Analogues Using Optum Claims Database

OBJECTIVES: To estimate the health care resource use (HCRU) and costs for patients diagnosed with acromegaly, following treatment with either pegvisomant, a growth hormone receptor antagonist (GHRA); or a somatostatin analogue (SSA). METHODS: De-identified data from Optum’s Clinformatics(TM) Data Mart (CDM), a US claims database, were used to retrospectively identify patients with acromegaly from January 2011 - June 2018 who were treated with the GHRA or an SSA. Propensity score matching (PSM) was used to address the small sample size of acromegaly patients and to ensure a direct comparison between the two treatment groups. PSM was fitted using a logistic regression model that included age, gender, total cost and Charlson Comorbidity Index (CCMI) score at index¹. Follow up (FU), HCRU (including Emergency Room [ER] costs, outpatient and inpatient costs) were compared. Average differences in change from baseline costs between the GHRA and SSA from the matched groups are shown with p-values obtained from a t-test. All costs are average across visits. RESULTS: Following PSM 91 patients were treated with GHRA and 269 patients were treated with SSAs. Mean ER costs were higher in the SSA vs GHRA group ($63 vs -$19, p=0.5646). On average outpatient visits were higher in the SSA vs GHRA group (2.88 and 2.82 respectively, p=0.8583); associated costs were higher ($1181 and $789 respectively p=0.3832). Regarding office visits and associated costs, these were also higher in the SSA group (Visit: 0.77 vs 0.74 and Cost: $1 vs -$7). The number of inpatient visits was equal (average =0.03) for both groups, with slightly higher costs in the GHRA group ($131 vs $161, p=0.9644). The drug cost and overall total cost of SSAs vs GHRA were significantly different ($2630 vs $6933, p<0.0001; $3681 vs $7561, p<0.0004 respectively). In the matched group, 18.7% of GHRA and 26.0% of SSA patients had new diabetes as comorbidity at FU. DISCUSSION: This analysis suggests that patients treated with SSAs generally utilize slightly more healthcare resources. In particular, outpatient costs were higher in SSA group compared to GHRA. Looking at the matched CCMI at FU, the value was higher (with more comorbidities) for the SSA group (63%) compared to the GHRA group (57%), with a trend that higher rates of new diabetes comorbidities occur in the SSA group. CONCLUSIONS: Although the drug cost and total cost of treating with SSAs is less than that for GHRA, these data analyses show that patients treated with SSAs generally have an increased trend in patient visits, and hence healthcare utilization costs; this could be a consideration when making treatment choices. Reference: 1. Coding Algorithms for Defining Comorbidities in ICD-9-CM and ICD-10 Administrative Data. Quan H et al Medical Care Vol 43, Number 11, Nov 2005