Loading
In a trial measuring how people feel, the forms go missing exactly when people are most unwell. That makes the remaining scores look better than the truth. The same thing happens on a larger scale when whole groups are not asked at all.
This record exists because every figure on this front rests on the forms that came back.
Why missing data is not random here. A person who is exhausted, in hospital or dying does not complete a quality of life questionnaire. The scores that remain are therefore systematically better than the scores that would have been recorded, and the problem grows with time since randomisation, which is exactly where survivorship outcomes are measured. Statistical methods exist to handle it, including mixed models under missing-at-random assumptions and pattern-mixture or joint models that treat dropout as informative, but every one rests on an assumption about the missing values that cannot be tested from the data. A trial reporting quality of life at 24 months with half the forms missing is reporting something, and it is not the quality of life of the randomised population.
What good looks like. PRO-TECT did well: 1,066 of 1,191 participants, 89.5 per cent, completed three-month follow-up. The eRAPID trial reports that "average patient compliance with weekly symptom reporting was 64.7%" over 18 weeks, which is a realistic figure for sustained weekly self-reporting.
Who is not asked at all. Four exclusions recur across this literature, and each removes people with the most to report.
Language. Most instruments have validated translations, but many trials and services administer in one language only. PRO-CTCAE being available in more than 60 validated languages is the exception rather than the rule.
Cognitive and sensory impairment. Eligibility for the PRO-CTCAE validation study required that participants "could read English and had no clinically significant cognitive impairment". That is appropriate for a validation study and it means the instrument's performance in people with cognitive impairment after treatment, the population a cognitive outcome is most about, is less well established. The same applies to sight loss after some treatments and to hearing loss after platinum.
Digital access. Every electronic symptom monitoring system needs a device, connectivity and the confidence to use them. PRO-TECT built an automated telephone option so that people without internet access could take part, and it mattered: its own subgroup analysis found benefits "most pronounced among younger, female, Black, and less educated participants", which only became visible because those participants were included.
Setting. Trials recruit from centres. People treated in small hospitals, in rural areas, in prisons and in care homes, and people who decline trials, are not in these datasets.
What follows. A recovery figure quoted from a trial describes the people who answered. This corpus states the figure and the denominator together wherever the source gives both, and where a source does not report its completion rate, that is a limit of the source and is said so.
Missing patient-reported outcome data in cancer trials is informatively missing: whether a form is absent depends on the unobserved value. No analysis recovers that value; methods differ only in which untestable assumption they make about it, so the defence is design, meaning fewer and shorter assessments, several modes of completion, and recorded reasons for non-completion, rather than analysis.
Query for this technology: (TITLE:"The questionnaires that were never returned, and the people never asked" OR ABSTRACT:"The questionnaires that were never returned, and the people never asked") AND (cancer OR tumor OR tumour OR oncology OR carcinoma OR lymphoma OR leukemia OR leukaemia OR myeloma OR sarcoma OR melanoma OR glioma). Results are unfiltered search hits about The questionnaires that were never returned, and the people never asked, not a curated reading list.
Shares PRO-CTCAE: side effects graded by the person having them, What has actually been implemented since those trials, and what has not, Telehealth and hospital-at-home in oncology, How recovery is measured: the questionnaires behind the numbers and the tags rejuvenation, survivorship, measurement, instruments.
Shares PRO-CTCAE: side effects graded by the person having them, What a difference has to be before a person would notice it, Asking people how they are, every week, as a treatment in its own right, How recovery is measured: the questionnaires behind the numbers and the tags rejuvenation, survivorship, measurement, instruments.
Shares What a difference has to be before a person would notice it, Asking people how they are, every week, as a treatment in its own right, How recovery is measured: the questionnaires behind the numbers, Patients lack understanding, navigation and agency and the tags rejuvenation, survivorship, measurement, instruments.
Shares What a difference has to be before a person would notice it, How recovery is measured: the questionnaires behind the numbers, Patients lack understanding, navigation and agency, Quality of life and the tags rejuvenation, survivorship, measurement, instruments.
Shares What a difference has to be before a person would notice it, How recovery is measured: the questionnaires behind the numbers, Patients lack understanding, navigation and agency, Quality of life and the tags rejuvenation, survivorship, measurement, instruments.
Shares What a difference has to be before a person would notice it, How recovery is measured: the questionnaires behind the numbers, Quality of life, HR-positive / HER2-negative breast cancer and the tags rejuvenation, survivorship, measurement, instruments.
Shares How recovery is measured: the questionnaires behind the numbers, Patients lack understanding, navigation and agency, Quality of life, HR-positive / HER2-negative breast cancer and the tags rejuvenation, survivorship, measurement, instruments.
Shares What a difference has to be before a person would notice it, How recovery is measured: the questionnaires behind the numbers, Patients lack understanding, navigation and agency, Quality of life and the tags rejuvenation, survivorship, measurement, instruments.