Aspects of health-related quality of life among clinically localized prostate cancer patients

Abstract

Medical research has yet to indicate superior survival benefit of any one treatment for localized prostate cancer. As such, health-related quality of life (HRQoL) is an important treatment outcome variable for prostate cancer patients. The purpose of the present study was to: 1) develop a disease-specific measure of HRQoL, 2) predict general domains of HRQoL from disease-specific domains of HRQoL, and 3) compare aspects of HRQoL across three treatment groups. Participants were men who had been treated with radical prostatectomy (n = 121), external beam radiation (n = 90), or interstitial radiation (n = 103). They completed the disease-specific survey and the SF-12 during a routine follow-up medical appointment. Principal components or factor analysis revealed 18 disease-specific subscales, including 9 physiologic function subscales, 3 symptom bother subscales, 3 role limitation subscales, and cancer worry, treatment regret, and treatment satisfaction subscales. Physiologic function scores significantly predicted and accounted for 14% to 53% of the variance in associated symptom bother. Disease-specific scores accounted for 29% and 19% of the variance in general physical and mental functioning, respectively. Univariate analyses of covariance, controlling for age and time since treatment, revealed expected group differences: surgery patients reported worse incontinence and worse sexual dysfunction, interstitial radiation patients reported worse irritative urinary symptoms, and both radiation groups reported worse diarrhea/frequency compared to surgery patients. Multinomial logistic regression indicated that 6 of the 18 disease specific scores significantly distinguished treatment groups, whereas general HRQoL did not greatly improve the fit of the model. The strongest predictors with all variables in the model were incontinence, irritative urinary symptoms, and diarrhea/frequency. Sexual symptoms did not significantly distinguish treatment groups in the presence of the other predictors, primarily because incontinence, which shared much common variance with erectile function, better predicted treatment group, i.e. incontinence better predicted surgery over both radiation groups than did erectile function. In summary, profiles of HRQoL were different across treatment groups primarily for disease-specific domains. Therefore, treatment decision-makers should take into consideration which of the potential treatment-related complications will be most tolerable for the individual patient.