A comparison of demographic, medical, social, and lifestyle behaviour correlates between cancer and non-cancer populations: a secondary analysis of the English Longitudinal Study of Aging cohort

Abstract

Abstract Purpose This secondary analysis of the English Longitudinal Study of Aging (ELSA) dataset aims to explore differences in sociodemographic, medical, psychosocial, and behavioural characteristics between cancer and non-cancer populations. Methods Data was collected in waves via computer-assisted interviewing, self-led questionnaires, and nurse-led physical measurements. Datasets were arranged according to variables of interest. IBM SPSS Statistics v27.0 software was used to perform parallel cross-sectional analyses. Chi-squares determined differences in categorical variables and Analyses of Variance (ANOVAs) determined differences in continuous variables. Results Cancer groups were more likely to report poor or fair general health than those without cancer (p < .001 in Waves 2, 6, 9 and COVID Wave 2; p = .016 in COVID Wave 1). They were also more likely to report often being troubled by pain (Wave 2 p = .037, Wave 6 p = .036, Wave 9 p = .006) and a difficulty with one or more activities of daily living (ADLs) and mobility (p < .001 in Waves 2, 6 and 9). In Waves 6, 9 and COVID Wave 2, cancer groups were less likely to report being current smokers (p < .001) and more likely to report being ex-smokers having smoked regularly in the past (p < .001). Cancer groups reported less frequent engagement in mild, moderate or vigorous physical activity in Wave 2 (p < .001, p = .005, p = .008) and Wave 9 (p = .003, p < .001, p < .001). Those with cancer were more likely to experience depression (p = .026) in Wave 6. Conclusion These findings indicate pain management may be a useful therapeutic aim in combination with targeted physical activity advice and psychosocial interventions for cancer populations.