A population-based study of pain and quality of life during the year before death in men with prostate cancer

Abstract

In order to explore how health-related quality of life changes towards the end of life, a questionnaire including the EuroQOl form and the Brief Pain Inventory form was sent to all men with prostate cancer in the county of Ostergötland, Sweden, in September 1999. Responders who had died prior to 1 January 2001 were later identified retrospectively. Of the 1442 men who received the questionnaire, 1243 responded (86.2%). In the group of responders, 167 had died within the study period, 66 of prostate cancer. In multivariate analysis, pain as well as death within the period of study were found to predict decreased quality of life significantly. Of those who died of prostate cancer, 29.0% had rated their worst pain the previous week as severe. The same figure for those still alive was 10.5%. On a visual analogue scale (range 0-100), the mean rating of quality of life for those who subsequently died of prostate cancer was 54.0 (95% confidence interval +/-5.2) and those still alive was 70.0 (+/-1.2). In conclusion, health-related quality of life gradually declines during the last year of life in men with prostate cancer. This decline may partly be avoided by an optimised pain management.

Figure 1

Quality of life, as rated on the EuroQol VAS, during the last 16 months of life. ± 95% confidence interval.

Figure 2

Quality of life, as estimated by the EQ-5D score, during the last 16 months of life. ± 95% confidence interval.

Figure 3

Distribution of ratings of ‘worst pain last week’ in BPI (number of responders within parenthesis).

Figure 4

(A) Distribution of answers to the BPI question of how pain interferes with general activities (number of responders within parenthesis). (B) Distribution of answers to the BPI question of how pain interferes with mood (number of responders in brackets). (C) Distribution of answers to the BPI question of how pain interferes with walking (number of responders within parenthesis). (D) Distribution of answers to the BPI question of how pain interferes with work (number of responders within parenthesis). (E) Distribution of answers to the BPI question of how pain interferes with relations with other people (number of responders within parenthesis). (F) Distribution of answers to the BPI question of how pain interferes with sleep (number of responders within parenthesis). (G) Distribution of answers to the BPI question of how pain interferes with enjoyment of life (number of responders within parenthesis).