Patient-reported outcomes (PROs) in lung cancer: Experiences from a nationwide feasibility study

Our objectives were to examine the feasibility of a nationwide collection of patient-reported outcomes (PROs) in a lung cancer population as well as in various sub-populations, and to describe the characteristics of responders compared to non-responders.

All patients diagnosed with lung cancer in Denmark are registered in the Danish Lung Cancer Registry (DLCR). The 7,295 patients registered in DLCR from 1 October 2013 until 30 September 2015 who had received treatment were eligible. Using the European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 and QLQ-LC13 questionnaires, we employed two different methods of delivery, resulting in two different project parts. In project part 1, the baseline questionnaire was handed out at the hospital departments before treatment. The following questionnaires were sent out as paper versions three times within one year. In project part 2, all questionnaires were electronic versions delivered in association with planned hospital visits.

Of the 7,295 lung cancer patients 4,229 (58%) completed at least one questionnaire, and 2,459 completed two or more. Only 562 baseline questionnaires were returned before treatment (7.7%), whereas 43.4%-57.4% of the potential responders completed the following questionnaires. The best response rates were achieved among patients treated with surgery and among patients who discussed their questionnaires with health care personnel. When comparing patient characteristics, responders had a significantly better health and a higher socioeconomic position than non-responders.

A decentralised model used for delivering the initial questionnaire to the patients was insufficient. It is our estimation that sending out electronic versions of the baseline questionnaires, as was done with the following questionnaires, would result in a significantly better patient coverage. Despite the severe morbidity and high mortality rate in lung cancer, reasonable response rates were achieved at follow-ups to this method, and PRO collection in this population was feasible.