Identification of socially vulnerable cancer patients - development of a register-based index (rSVI).

BACKGROUND: Social vulnerability is a complex construct which is beyond relying on single measures. If socially vulnerable patients should be identified, we need a composite measure capturing the patient's overall circumstances. This study presents the development of a social vulnerability index (rSVI) for cancer patients based on administrative data from population-based registers. METHODS: All patients, who died from cancer within 5 years after the cancer diagnosis during 2013-2018 (n = 44,187), were identified and divided in four subcohorts according to survival; index cohort surviving 3-5 years (n = 3044 surviving 3-5 years), cohort 1 (n = 27,170 surviving < 1 year), cohort 2 (n = 9450 surviving 1-2 years), and cohort 3 (n = 4523 surviving 2-3 years). Variables from ten registries on health and social issues were linked to each individual patient. Variables of interest were weighted to construct the rSVI using the index-cohort. rSVI was subsequently tested on the three other cohorts for validation. RESULTS: The rSVI included weighted values for marital status, ethnicity, education, income, unemployment, psychiatric comorbidity, and somatic comorbidity. The validity of the rSVI was supported by the expected trend in proportions of vulnerable patients when applied on the other cohorts. Single social measures appeared insufficient in identification of vulnerable patients when compared with the rSVI. CONCLUSION: The rSVI provides a tool for identification of socially vulnerable cancer patients using administrative data. The index requires further validation in other patient groups and is tested against other measures of vulnerability. Future perspectives are to use the rSVI as predictor of advanced cancer patients' use of healthcare services.