Quality of care indicators and their related outcomes: a population-based study in prostate cancer patients treated with radiotherapy.

Abstract

We describe variations across the regional cancer centres in Ontario, Canada for five prostate cancer radiotherapy (RT) quality indicators: incomplete pre-treatment assessment, follow-up care, leg immobilization, bladder filling, and portal film target localization. Along with cancer centre volume, we examined each indicator's association with relevant outcomes: long-term cause-specific survival, urinary incontinence, and gastrointestinal and genitourinary late morbidities. We conducted a population-based retrospective cohort study of 924 prostate cancer patients diagnosed between 1990 and 1998 who received RT within 9 months of diagnosis. Data sources included treating charts and registry and administrative data. The associations between indicators and outcomes were analysed using regression techniques to control for potential confounders. Practice patterns varied across the regional cancer centres for all indicators (p<0.0001). Incomplete pre-treatment assessment was associated with worse cause-specific survival although this result was not significant when adjusted for confounding (adjusted RR=1.78, 95% CI=0.79-3.98). Treatment without leg immobilization (adjusted RR=1.72, 95% CI=1.16-2.56) and with an empty bladder (adjusted RR=1.98, 95% CI=1.08-3.63) was associated with genitourinary late morbidities. Treatment without leg immobilization was also associated with urinary incontinence (adjusted RR=2.18, 95% CI=1.23-3.87). We documented wide variations in practice patterns. We demonstrated that measures of quality of care can be shown to be associated with clinically relevant outcomes in a population-based sample of prostate cancer patients.