Background
Population-based studies using patient-reported outcomes (PROs) provide essential information on health-related quality of life (HRQoL) and patient experiences, but declining response rates in todays emphasis on technology centered data collection increasingly threaten representativeness and validity. These challenges are particularly relevant in surveys that include both cancer survivors and the general population. High-quality evidence  on participation patterns is needed to support the use of PROs in cancer epidemiology  and health service research.
Objective
To describe a mixed method recruitment strategy and  response rates of the SURV-ICE cohort, and to examine participation patterns.
Methods
SURV-ICE is a population-based, cross-sectional study of adults residing in Iceland. Individuals diagnosed with cancer between 2014 and 2024 were identified through nationwide cancer registration, and a control group without cancer was randomly sampled from the general population and frequency matched by age and gender. Data were collected using validated questionnaries assessing HRQoL and health literacy, including both EORTC QLQ C-30 and SURV-100. A mixed-mode recruitment strategy was employed, including postal invitations, web-based questionnaires accessed via a secure national health portal, and a paper questionnaire if requested, in addition to age-adapted text message and telephone follow-up. Questionnaire data were linked with national registry information using unique personal identification numbers.
Results
After exclusions, 15,668 individuals were invited, including 10,005 cancer survivors and 5,663 controls. In total, 7,791 individuals aged 19-99 year old responded. Response rates were higher among participants with a cancer diagnosis (54.9%) than among controls (40.6%). Within the cancer group, participation varied by cancer type, with the highest response rates observed among individuals with breast cancer (62.7%). Response rates were markedly lower among participants aged 81 years and older (29.5%), compared to participants aged 19-80 year old (53.2%). Females had a higher response rate than males across most age groups. Questionnaire completion was high, with more than 97% of respondents completing at least 80% of items in one or more survey instruments. Cumulative response increased following each contact during the data collection. Statistically significant differences between responders and non-responders were observed with respect to age, gender, region of residence and study group. Older individuals and members of the general population without cancer were more likely to be non-responders.
Conclusions/Implications
The SURV-ICE study demonstrates the feasibility and acceptability of large-scale, population-based PRO data collection using a mixed-mode recruitment strategy combined with national registry linkage. The resulting cohort provides a valuable resource for cancer survivorship research. The observed response rate is comparable or higher to that reported in other recent population-based PRO surveys. Observed differences in participation across subgroups highlight the importance of accounting for potential non-response bias when interpreting PRO data. These findings provide methodological insights that can inform the design, implementation and interpretation of future population-based surveys.