RESUMO
AIMS: We contribute to the methodological literature on the assessment of health inequalities by applying an algorithmic approach to evaluate the capabilities of socioeconomic variables in predicting the prevalence of non-communicable diseases in a Norwegian health survey. METHODS: We use data from the seventh survey of the population based Tromsø Study (2015-2016), including 11,074 women and 10,009 men aged 40 years and above. We apply the random forest algorithm to predict four non-communicable disease outcomes (heart attack, cancer, diabetes and stroke) based on information on a number of social root causes and health behaviours. We evaluate our results using the classification error, the mean decrease in accuracy, partial dependence statistics. RESULTS: Results suggest that education, household income and occupation to a variable extent contribute to predicting non-communicable disease outcomes. Prediction misclassification ranges between 25.1% and 35.4% depending on the non-communicable diseases under study. Partial dependences reveal mostly expected health gradients, with some examples of complex functional relationships. Out-of-sample model validation shows that predictions translate to new data input. CONCLUSIONS: Algorithmic modelling can provide additional empirical detail and metrics for evaluating heterogeneous inequalities in morbidity. The extent to which education, income and occupation contribute to predicting binary non-communicable disease outcomes depends on both non-communicable diseases and socioeconomic indicator. Partial dependences reveal that social gradients in non-communicable disease outcomes vary in shape between combinations of non-communicable disease outcome and socioeconomic status indicator. Misclassification rates highlight the extent of variation within socioeconomic groups, suggesting that future studies may improve predictive accuracy by exploring further subpopulation heterogeneity.
RESUMO
BACKGROUND: Differences in the sociodemographic characteristics of participants and non-participants in population-based studies may introduce bias and reduce the generalizability of research findings. This study aimed to compare the sociodemographic characteristics of participants and non-participants of the seventh survey of the Tromsø Study (Tromsø7, 2015-16), a population-based health survey. METHODS: A total of 32,591 individuals were invited to Tromsø7. We compared the sociodemographic characteristics of participants and non-participants by linking the Tromsø7 invitation file to Statistics Norway, and explored the association between these characteristics and participation using logistic regression. Furthermore, we created a geographical socioeconomic status (area SES) index (low-SES, medium-SES, and high-SES area) based on individual educational level, individual income, total household income, and residential ownership status. We then mapped the relationship between area SES and participation in Tromsø7. RESULTS: Men, people aged 40-49 and 80-89 years, those who were unmarried, widowed, separated/divorced, born outside of Norway, had lower education, had lower income, were residential renters, and lived in a low-SES area had a lower probability of participation in Tromsø7. CONCLUSIONS: Sociodemographic differences in participation must be considered to avoid biased estimates in research based on population-based studies, especially when the relationship between SES and health is being explored. Particular attention should be paid to the recruitment of groups with lower SES to population-based studies.
