Hospital racial composition and differences in treatment for localized prostate cancer Craig Evan Pollack 1,2,3 , KJ Lia...
of 1

Hospital Racial Composition and Differences 4.4.08

Author: Craig Pollack
Published on: Mar 4, 2016
Published in: Health & Medicine      
Source: www.slideshare.net


Transcripts - Hospital Racial Composition and Differences 4.4.08

  • 1. Hospital racial composition and differences in treatment for localized prostate cancer Craig Evan Pollack 1,2,3 , KJ Liao 3,4 , Chantal M Montagnet 3,4 , Samuel H. Field 2,3,4,5 , Peter Groeneveld 2,3,4,5 , Mary Putt 4 , Katrina Armstrong 3,4,5 1 The Robert Wood Johnson Clinical Scholars Program, University of Pennsylvania, 2 Philadelphia VA Medical Center, 3 Leonard Davis Institute of Health Economics, University of Pennsylvania, 4 School of Medicine, University of Pennsylvania, 5 Center for Health Equity Research and Promotion, Philadelphia VA Medical Center <ul><li>Type of Active Treatment </li></ul><ul><li>Black men were less likely to receive prostatectomy. </li></ul><ul><li>Patients treated at a hospital with a high percentage of Black patients were significantly less likely to receive prostatectomy. </li></ul><ul><li>Effect modification was present—hospital racial composition was differently associated with type of treatment for Black and White men. </li></ul><ul><li>Figure 1. Predicted Probability of Prostatectomy holding all other variables constant at their sample means </li></ul><ul><li>Compared with White men, Black men are more likely to be diagnosed with advanced state prostate cancer, are less likely to receive active treatment and prostatectomy, and have higher rates of mortality. </li></ul><ul><li>Differences are linked to a complex interplay between individual, health care system, social, and contextual factors. </li></ul><ul><li>Black and White men tend to receive care at different hospitals which may contribute to racial differences. </li></ul>Background <ul><li>1999 Medicare Claims data was used to identify incident cases of localized prostate cancer. </li></ul><ul><li>Exclusion based on prior history of prostate cancer, matching to continental US, race other than Black or White, enrollment in HMO, unable to match to hospital assignment. </li></ul><ul><li>Patients assigned to the first hospital they presented to at or following diagnosis (89% match rate for all patients, 93% match for treated patients). </li></ul><ul><li>Outcome variables: </li></ul><ul><li>Receipt of active treatment (yes/no) </li></ul><ul><li>Type of active treatment received (prostatectomy versus radiation) </li></ul><ul><li>Primary exposure variables: </li></ul><ul><li>Patient race </li></ul><ul><li>Hospital racial composition defined as the percentage of hospital admissions who were identified as Black. Categorized as <11%, 11-30%, and >30% </li></ul><ul><li>Covariates: comorbidity, block group SES, region, hospital characteristics (teaching status, bed size, non-profit vs for-profit, general vs. specialty) </li></ul><ul><li>Analysis: </li></ul><ul><li>Hierarchical, multivariable analyses were employed to account for the nesting of patients within hospitals. </li></ul><ul><li>Sensitivity analyses examined patients < 76 years old, assigning patients to (1) the hospital where they seen most frequently, and (2) the hospital where they received active treatment. </li></ul>Methods <ul><li>Initiative of treatment </li></ul><ul><li>Black men were significantly less likely to receive active treatment. </li></ul><ul><li>Patients treated at hospitals with high percentages of Black patients were significantly less likely to receive active treatment. </li></ul><ul><li>Effect modification between patient race and hospital racial composition was not present </li></ul><ul><li>Table 2. Adjusted odds of active treatment </li></ul><ul><li>*Adjusted for sociodemographic and hospital characteristics </li></ul>Results <ul><li>Lower odds of active treatment/prostatectomy were seen in hospitals with >30% Black patients. </li></ul><ul><li>Better understanding how people come to receive care at specific hospitals and targeting interventions at specific hospitals may help mitigate differences in care. </li></ul>Conclusions Table 1. Sociodemographic characteristics of Population All Patients (N=68,232) Treated Patients (N=31,940) White Black White Black N 60833 7399 29034 2906 Median Age 73.1 (SD 5.2) 72.2 (5.1) 71.9 (4.4) 70.9 (4.2) Comorbidity 0 1 2 3 4+ 18,374 (31.9) 17,872 (29.4) 19,839 (17.8) 5,958 (9.8) 6,790 (11.1) 1,675 (22.6) 2,017 (27.3) 1,510 (2.4) 897 (12.1) 1,300 (17.6) 10,117(34.5) 9.288 (32.0) 5,107 (17.6) 2,429 (8.4) 2,093 (7.2) 695 (23.9) 929 (32.0) 630 (21.7) 308 (10.6) 344 (11.8) Block group income (SD) 52,522 (24,198) 35,123 (17,763) 53,265 (24,362) 36,318 (18,382) Treatment None Pros. Radiation 31,799 (52.3) 8,374 (13.8) 20,660 (34.0) 4,493 (60.7) 723 (9.8) 2,183 (29.5) - 8,374 (28.8) 20,660(71.2) - 723 (24.9) 2,183 (75.1) Hospital Racial Comp <11% 11-30% >30% 40,770 (67.0) 14,517 (23.8) 5,546 (9.1) 1,470 (19.9) 2,612 (35.3) 3,317 (44.8) 19,800(68.2) 6,813 (23.5) 2,421 (8.3) 642 (22.1) 1,083 (37.3) 1,181 (40.6) Baseline Model Adjusted Model* Race White Black 1 0.74 (0.70-0.79) 1 0.78 (0.74-0.83) Hospital Comp. <11% Black 11-30% >30% 1 0.86 (0.81-0.93) 0.75 (0.69-0.82) <ul><li>To determine if hospital racial composition is associated with whether men receive active treatment for localized prostate cancer and the type of treatment that men receive. </li></ul><ul><li>To examine if hospital racial composition mediates the association between patient race and prostate cancer treatment. </li></ul>Objectives

Related Documents