This study did not constitute human subjects research. The Premier Healthcare Database is an all-payer database representing over 1 billion inpatient and outpatient encounters from a diverse range of over 1000 U.S. hospital-based facilities [2]. The database was searched from January 2010 to March 2020 for outpatient encounters containing a billing code for pelvic MRI (regardless of contrast medium use) with an ICD (9th or 10th revision) code indicating a diagnosis of elevated PSA, abnormal digital rectal examination, or prostate cancer. Patient-, facility-, and MRI-level characteristics were extracted. Charge per MRI encounter was adjusted to 2019 U.S. dollars by use of the consumer price index. Supplemental Methods are available in the online supplement.

A total of 552 facilities performed 37,073 prostate MRI examinations over the study period. Median facility volume was 55 (IQR, 20–110) examinations per year, and median charge per examination was $4419 (IQR, $3304–5520). There was 26-fold variation between the lowest ($593) and highest ($15,150) median facility charges. Among self-pay patients, the median charge per examination was $4350 with 25-fold variation between the lowest ($550) and highest ($13,815) median facility charges. Imaging-associated charges accounted for 91.4% ($154,648,204/$169,190,899) of total charges, and contrast medium–associated charges accounted for 6.0% ($10,087,954/$169,190,899) (Fig. 1). The median inflation-adjusted price per examination increased from $4192 in 2010 to $4586 in 2020.

In multilevel hierarchic linear regression analysis, interfacility variability explained 63.9% of charge variability (Table 1). The next-largest contributor to charge variability was IV contrast media, which explained 10.3% of variability. Facility size explained 0.6% of variability, facility volume explained 0.0% of variability, and geographic region explained 2.0% of variability. Patient characteristics, including insurance type, explained 0.0% of variability.

This analysis revealed large variation in prostate MRI charges nationally, primarily representing facility-to-facility variation. Patient insurance type and facility characteristics explained a small portion of charge variation. The explanation for the large variation among facilities is unclear. Absence of incentives to provide competitive prices may enable hospitals to set arbitrary prices.

Insurance coverage for prostate MRI is inconsistent; only 11% of insurance providers covered MRI of biopsy-naïve patients in 2018 [1]. Even with prostate MRI insurance coverage, patients may be responsible for a significant proportion of charges if the study is performed out of network, an increasingly common problem given the growth of narrow-network insurance plans [3]. Additionally, only a fraction of hospitals are compliant with the CMS mandate to publicly post chargemasters, and a centralized price comparison resource is lacking [4]. Patients may be responsible for paying for the examination but poorly equipped to compare highly variable charges among facilities. These concerns compound population disparities in insurance rates [5, 6].

This study included only hospital-affiliated facilities, which account for approximately 60% of U.S. imaging volume [7]. We could not account for MRI technique or quality. Additionally, hospital reimbursement data were unavailable. Nonetheless, using a national sample of 37,073 encounters, we shed light on large variability in prostate MRI charges, which are otherwise not explained by insurance claims data. Expanded coverage, broadening of narrow networks, and stricter enforcement of price transparency regulations may ameliorate this problem.