The time taken to complete quality of life questionnaires in an advanced cancer trial

Abstract

8592 Background: Respondent burden is important when designing trials in advanced cancer. We determined the time taken to complete a set of quality of life questionnaires (TTTCQ) in a randomized trial of sertraline in advanced cancer. Methods: Participants recorded the times they started and finished an 11 page questionnaire of 187 items including the Centre for Epidemiologic Studies Depression Scale, Hospital Anxiety and Depression Scale, Functional Assessment of Cancer Therapy General and Fatigue scales, and 4 other validated instruments. Determinants of the TTTCQ were sought amongst the self-ratings of anxiety, depression, fatigue, trouble concentrating, other aspects of quality of life, and other characteristics including age, Karnofsky Performance Status (KPS) and treatment. Distributions are summarised by their median, interquartile range (25th to 75th percentiles, IQR) and interdecile range (10th to 90th percentiles, IDR). Associations were assessed with Spearman’s rank correlation coefficient (r), and differences with the Mann-Whitney test. All p-values are two-sided. Results: The TTTCQ was available for 128 of 150 subjects (85%). Median age was 66 (IQR 59 to 73), median KPS was 80 (IQR 70 to 90), and 63% were male. The median TTTCQ was 30 mins (IQR 23 to 40, IDR 15 to 50), equating to a median of 10 secs per item (IQR 7 to 13, IDR 5 to 16). The TTTCQ was modestly correlated with age and KPS (r = 0.19, p = 0.04 for both). Age and KPS were not correlated (r = 0.03, p = 0.8). Subjects aged 60 or older had a median TTTCQ 10 mins longer than those 59 or younger (35 v 25 mins, p = 0.002). Subjects with a KPS of 100, 90, 80, and 70 or less had a median TTTCQ of 20, 25, 31, and 35 mins respectively. TTTCQ was modestly correlated with self-ratings of: problems coping with treatment, dry mouth, feeling dizzy or lightheaded (r = 0.22 to 0.25, p = 0.01). Correlations were weak or non-existent with all measures of depression, anxiety, trouble concentrating, emotional well-being, and overall well-being (r = 0.10 to 0.18, p = 0.04 to 0.25). TTTCQ was unaffected by gender, opioids, steroids and number of previous cytotoxic regimens. Conclusions: Respondent burden was less of a problem than anticipated and little affected by treatments or self-rated psychological problems. No significant financial relationships to disclose.