Assessing the potential utility of commercial ‘big data’ for health research: enhancing small-area deprivation measures with Experian™ Mosaic groups

In contrast to area-based deprivation measures, commercial datasets remain infrequently used in health research and policy. Experian collates numerous commercial and administrative data sources to produce Mosaic groups which stratify households into 15 groups for marketing purposes. We assessed the potential utility of Mosaic groups for health research purposes by investigating their relationships with Indices of Multiple Deprivation (IMD) for the British population. Mosaic groups showed significant associations with IMD quintiles. Correspondence Analysis revealed variations in patterns of association, with Mosaic groups either showing increasing, decreasing, or some mixed trends with deprivation quintiles. These results suggest that Experian's Mosaics additionally measure other aspects of socioeconomic circumstances to those captured by deprivation measures. These commercial data may provide new insights into the social determinants of health at a small area level