Race and Treatment Regret in Recurrent Prostate Cancer
Race and Treatment Regret in Recurrent Prostate Cancer
The prospective COMPARE registry was opened across 150 geographically diverse sites throughout the United States in February 2004 and enrolled 1120 men with biochemical recurrence after primary therapy for localized prostate cancer through March 2007. Men of any age presenting with an increasing PSA after treatment of primary nonmetastatic prostate adenocarcinoma were eligible for this observational, prospective database, with the higher PSA level defined as follows: (1) an increase of ≥0.2 ng ml on repeated testing after radical prostatectomy or (2) 2 increases greater than the nadir, with a PSA value ≥50% greater than the nadir and a minimal PSA value ≥0.2 ng ml greater than the post-radiation therapy nadir (a steering committee recommended modification of the Phoenix definition). Patients who received adjuvant or neoadjuvant androgen deprivation therapy (ADT) for ≤3 years were eligible but must have ceased ADT for >4 months before enrollment. Exclusion criteria included any ongoing treatment of prostate cancer and certain comorbidities. The full exclusion criteria have been detailed in previous reports. To be included in the current study, patients must have received radical prostatectomy, external beam radiation therapy or brachytherapy without adjuvant ADT. Furthermore, patients must have had complete information on demographic characteristics of interest and regret status. Of the entire registry, 311 patients were excluded for receiving adjuvant ADT, 250 men were excluded for not having information on regret status and 75 patients were excluded for incomplete information on demographic characteristics of interest, leaving a total of 484 patients who met inclusion criteria and who constituted the study cohort.
On enrollment into the study, data obtained from each patient included prostate cancer history, comorbidities, current physical examination parameters and pertinent laboratory data, and patient-reported common complaints. Each patient also completed the validated 'Prostate Health-Related Quality-of-Life Questionnaire' questionnaire including questions on demographics and the impact on quality of life during the previous 4 weeks of his urinary function, sexual function, bowel function and other medical problems. Patients were scheduled for follow-up visits according to their physicians' standard of care at which time follow-up quality of life questionnaires as above were completed. Patients gave informed consent and data were made anonymous. Each participating site submitted registry documents to the central or local Institutional Review Board and obtained written approval before conduction any registry-related procedures. Investigators obtained informed consent from patients and provided them with a Health Insurance Portability and Accountability Act statement before enrollment.
Treatment regret was assessed based on a validated two-question questionnaire. Patients were considered to have treatment regret if they either answered 'definitely false' or 'mostly false' to the question 'If I had known everything I could have known, I would still have chosen the same treatment approach for prostate cancer,' or if they answered 'mostly true' or 'definitely true' to the question 'During the past 4 weeks, I felt I would be better off if I had chosen a different treatment approach for prostate cancer.' Race was determined at the time of the study as non-Hispanic white, black, non-black Hispanic, Asian and American-Indian. Health literacy was determined based on the Rapid Estimation of Literacy of Medicine (REALM-SF), a reliable and validated short form derived from the widely used and validated longer (66-word) REALM form. English-speaking persons were asked to pronounce the following seven words aloud: 'Menopause, Exercise, Rectal, Behavior, Antibiotics, Jaundice and Anemia.' We collapsed health literacy findings into two mutually exclusive groups—ninth grade or higher health literacy (all seven words pronounced correctly) or less than ninth grade health literacy (less than seven words pronounced correctly). For each of sexual, bowel and urinary domains, patients were asked 'Overall, how big a problem has your function been for you during the last 4 weeks?' with the five possible answers being 'No problem,' 'Very small problem,' 'Small problem,' 'Moderate problem,' 'Very big problem.' As more than 50% of the patients answered 'No problem,' we dichotomized functional issues for the purpose of modeling as 'No problem' versus 'Any problem.'
Responses to initial questionnaires at study entry were used for all analyses. Fisher's exact test was used to perform univariate analyses to determine whether race (black versus non-black), health literacy (health literate versus not), sexual problems (yes versus no), urinary problems (yes versus no), bowel problems (yes versus no), age (continuous), cardiovascular comorbidity (yes versus no, as previously reported), living situation (lives with partners versus not), educational status (graduated high school versus not) or treatment type (radiation versus surgery) was associated with treatment regret. Fisher's exact test was also used to compare baseline clinical and demographic characteristics between non-black and black patients. Multivariable logistic regression was used to determine whether race was associated with treatment regret after adjusting for the previously listed sociodemographic factors, time since initial treatment, health literacy, treatment type and patient-reported complications after treatment (bowel, urinary and sexual). Furthermore, a multivariable logistic regression model was used to explore whether there was an interaction between race and the presence of patient-reported sexual problems with regards to treatment regret. Adjusted odds ratios (AORs) and 95% confidence intervals (CIs) were reported for each covariate and a P-value of <0.05 was used to determine statistical significance. SAS version 9.3 (SAS Institute, Cary, NC, USA) was used for all calculations.
Materials and Methods
Patient Selection
The prospective COMPARE registry was opened across 150 geographically diverse sites throughout the United States in February 2004 and enrolled 1120 men with biochemical recurrence after primary therapy for localized prostate cancer through March 2007. Men of any age presenting with an increasing PSA after treatment of primary nonmetastatic prostate adenocarcinoma were eligible for this observational, prospective database, with the higher PSA level defined as follows: (1) an increase of ≥0.2 ng ml on repeated testing after radical prostatectomy or (2) 2 increases greater than the nadir, with a PSA value ≥50% greater than the nadir and a minimal PSA value ≥0.2 ng ml greater than the post-radiation therapy nadir (a steering committee recommended modification of the Phoenix definition). Patients who received adjuvant or neoadjuvant androgen deprivation therapy (ADT) for ≤3 years were eligible but must have ceased ADT for >4 months before enrollment. Exclusion criteria included any ongoing treatment of prostate cancer and certain comorbidities. The full exclusion criteria have been detailed in previous reports. To be included in the current study, patients must have received radical prostatectomy, external beam radiation therapy or brachytherapy without adjuvant ADT. Furthermore, patients must have had complete information on demographic characteristics of interest and regret status. Of the entire registry, 311 patients were excluded for receiving adjuvant ADT, 250 men were excluded for not having information on regret status and 75 patients were excluded for incomplete information on demographic characteristics of interest, leaving a total of 484 patients who met inclusion criteria and who constituted the study cohort.
Data Collection
On enrollment into the study, data obtained from each patient included prostate cancer history, comorbidities, current physical examination parameters and pertinent laboratory data, and patient-reported common complaints. Each patient also completed the validated 'Prostate Health-Related Quality-of-Life Questionnaire' questionnaire including questions on demographics and the impact on quality of life during the previous 4 weeks of his urinary function, sexual function, bowel function and other medical problems. Patients were scheduled for follow-up visits according to their physicians' standard of care at which time follow-up quality of life questionnaires as above were completed. Patients gave informed consent and data were made anonymous. Each participating site submitted registry documents to the central or local Institutional Review Board and obtained written approval before conduction any registry-related procedures. Investigators obtained informed consent from patients and provided them with a Health Insurance Portability and Accountability Act statement before enrollment.
Definition of Regret, Race, Health Literacy and Functional Problems
Treatment regret was assessed based on a validated two-question questionnaire. Patients were considered to have treatment regret if they either answered 'definitely false' or 'mostly false' to the question 'If I had known everything I could have known, I would still have chosen the same treatment approach for prostate cancer,' or if they answered 'mostly true' or 'definitely true' to the question 'During the past 4 weeks, I felt I would be better off if I had chosen a different treatment approach for prostate cancer.' Race was determined at the time of the study as non-Hispanic white, black, non-black Hispanic, Asian and American-Indian. Health literacy was determined based on the Rapid Estimation of Literacy of Medicine (REALM-SF), a reliable and validated short form derived from the widely used and validated longer (66-word) REALM form. English-speaking persons were asked to pronounce the following seven words aloud: 'Menopause, Exercise, Rectal, Behavior, Antibiotics, Jaundice and Anemia.' We collapsed health literacy findings into two mutually exclusive groups—ninth grade or higher health literacy (all seven words pronounced correctly) or less than ninth grade health literacy (less than seven words pronounced correctly). For each of sexual, bowel and urinary domains, patients were asked 'Overall, how big a problem has your function been for you during the last 4 weeks?' with the five possible answers being 'No problem,' 'Very small problem,' 'Small problem,' 'Moderate problem,' 'Very big problem.' As more than 50% of the patients answered 'No problem,' we dichotomized functional issues for the purpose of modeling as 'No problem' versus 'Any problem.'
Statistical Analysis
Responses to initial questionnaires at study entry were used for all analyses. Fisher's exact test was used to perform univariate analyses to determine whether race (black versus non-black), health literacy (health literate versus not), sexual problems (yes versus no), urinary problems (yes versus no), bowel problems (yes versus no), age (continuous), cardiovascular comorbidity (yes versus no, as previously reported), living situation (lives with partners versus not), educational status (graduated high school versus not) or treatment type (radiation versus surgery) was associated with treatment regret. Fisher's exact test was also used to compare baseline clinical and demographic characteristics between non-black and black patients. Multivariable logistic regression was used to determine whether race was associated with treatment regret after adjusting for the previously listed sociodemographic factors, time since initial treatment, health literacy, treatment type and patient-reported complications after treatment (bowel, urinary and sexual). Furthermore, a multivariable logistic regression model was used to explore whether there was an interaction between race and the presence of patient-reported sexual problems with regards to treatment regret. Adjusted odds ratios (AORs) and 95% confidence intervals (CIs) were reported for each covariate and a P-value of <0.05 was used to determine statistical significance. SAS version 9.3 (SAS Institute, Cary, NC, USA) was used for all calculations.
Source...