Should function and bother be measured and reported separately for prostate cancer quality-of-life domains?

OBJECTIVES: To evaluate the psychometric properties of the three domains bowel, urinary, and sexual function as they were measured in the Prostate Cancer Outcomes Study and examine their use in different research and practice settings. Leading prostate cancer health-related quality-of-life questionnaires include questions that measure patients' bowel, urinary, and sexual function and their perceived annoyance (or bother) caused by limited functioning. The published results are mixed on reporting function and bother independently or together as a single domain.
METHODS: Statistical tools from classical measurement theory and factor analytic methods were used to evaluate the psychometric properties of the Prostate Cancer Outcomes Study disease-specific scales. The findings from studies of other prostate cancer outcomes scales and clinical input were included to formulate the conclusions.
RESULTS: Factor analysis results uncovered a multidimensional structure within two of the three domains. The urinary domain consisted of items measuring two factors: incontinence and urinary obstructive symptoms. Sexual dysfunction consisted of two dimensions: interest in sexual activity and erectile function.
CONCLUSIONS: These empirical results suggest that bowel dysfunction and urinary incontinence can each be combined with measures of bother to produce overall measures of function; however, evidence was present for the need for separate measures of sexual function, sexual interest, and perceived bother with sexual function. For informing patient-doctor communications, function and bother on all three domains should be reported separately, because treatment decisions or symptom management may vary depending on a patient's perceived concern about their health-related quality-of-life.