Skip to main content


Higher Hospital Competition Associated With Improved Quality of Prostate Cancer Care

March 18, 2021


Ravishankar Jayadevappa, PhD, MS, University of Pennsylvania, Philadelphia, discusses results from a study that examined the association between hospital competition, process of care, and outcomes in Medicare fee-for-service beneficiaries with prostate cancer.

These results were presented at the virtual 2021 ASCO Genitourinary Cancers Symposium.


I'm Ravishankar Jayadevappa, associate professor at the University of Pennsylvania. My abstract was #235, “Hospital Competition on Quality and Cost of Prostate Cancer Care.” This study has been supported by AHRQ‑R01 grant.

Brief background about the prostate cancer, we all know that prostate cancer is the most common cancer among men in the US. To give a current picture, in 2020 alone, there were about under 19,930 estimated new cases of prostate cancer and 33,000 prostate cancer‑related death. Prostate cancer is the most expensive cancer, accounting for almost $15 billion annual expenditure.

In this study, what we are looking at, the objective is to look at the hospital because hospitals provide majority of care for prostate cancer patients. In US, we have currently about 5534 hospitals. What we are interested is how an increase or decrease in hospital competition affected quality of care as well as cost of care.

There are 2 elements of computing hospital competition. One is the hospital area, and competition intensity. I'm not going to go to detail with the methods of hospital competition. What we have used is the Herfindahl‑Hirschman Index, HHI, which is a popular measure of competition.

HHI index ranges from 0-1 with higher value representing increased market competition, that is lower competition but increased market concentration, meaning lower competition. The lower the value, the higher the competition.

The objective of this study is to assess the effects of hospital competition on racial, ethnic disparity in the outcomes of quality of care and cost of care among men with prostate cancer. Our conceptual model consists of structure, process, and outcome. The structure is the hospital competition and the process of care is the treatment the patient received, and the treatment type, and follow‑up care.

In terms of outcomes, we are interested in mortality, readmission, ER complications, health resource utilization, and finally, cost. We have used SEER-Medicare linked data for the period of 1995 to 2016. For this analysis, we are using the data from 2000 to 2016 only. What I'm presenting is the data related to 2000 to 2016.

The hospital data has been linked to the SEER-Medicare data, and this is a retrospective cohort design, and the study sample consists of all African American, Hispanic, Asian and Caucasian or white men diagnosed with prostate cancer during the period 2002 to 2003. The clinical data obtained from the SEER registry data.

The methods for the mortality we are using all‑cause mortality as well as prostate cancer specific mortality, health resource utilizations in terms of inpatient and ER visit. Cost includes inpatient, outpatient, durable medical equipment, as well as the hospice. Our covariates are marital status, geographic region, as well as clinical stage of cancer.

We identified other newly diagnosed prostate cancer patients during 2000 to 2003, and we followed them up to 15 years. HHI was computed for each hospital and it ranges from 0-1. The low HHI score index indicates lower concentration or higher competition.

To the analysis we used Cox regression for survival and log-link TLM model for cost, and quasar model for qualitative data. All our analyses are adjusted for propensity score.

In terms of summary of the research, we observed for baseline of low HHI leads to improved outcome that is higher competition, leads to improved outcome in terms of ER visit, inpatient, cost of care, and all‑cause mortality, as well as prostate cancer specific mortality across all racial and ethnic groups, that is African American, Hispanic, white, and Asian.

In terms of the cost alone, for African American increased hospital competition index, that was associated with 13% reduction in cost. In conclusion, hospital competition among older prostate cancer patients varied by racial, ethnic group. Lower HHI that is higher competition was associated with decreased use of hospitalization, ER visits, and overall cost across all racial, ethnic groups.

In addition, association between hospital computation and improved mortality also varied across racial ethnic groups. To summarize, the higher HHI, the higher competition is associated with improved long‑term mortality for all racial, ethnic groups.

In terms of the policy measures, the improved hospital competition that measures to improve hospital competition will improve not only quality of care, as well as, it reduces the cost of care.

Thank you.   

Jayadevappa R, Chhatre S, Malkowicz, et al. Hospital competition, quality, and cost of prostate cancer care. Presented at: the virtual 2021 ASCO Genitourinary Cancers Symposium; February 11-13, 2021. Abstract 235.

Back to Top