This paper uses data from Understanding Society: the UK Household Longitudinal Study to explore the association between fuel poverty and a set of well-being outcomes: life-satisfaction, self-reported health measures and more objectively measured biomarker data. Over and above the conventional income-fuel cost indicators, we also use more proximal heating deprivation indicators. We create and draw upon a set of composite indicators that concomitantly capture (the lack of) affordability and thermal comfort. Depending on which fuel deprivation indicator is used, we find heterogeneous associations between fuel poverty and our well-being outcomes. Employing combined fuel deprivation indicators, which takes into account the income-fuel cost balance and more proximal perceptions of heating adequacy, reveals the presence of more pronounced associations with life satisfaction and fibrinogen, one of our biological health measures. The presence of these strong associations would have been less pronounced or masked when using separately each of the components of our composite fuel deprivation indicators as well as in the case of self-reported generic measures of physical health. Lifestyle and chronic health conditions plays a limited role in attenuating our results, while material deprivation partially, but not fully, attenuates our associations between fuel deprivation and well-being. These results remain robust when bounding analysis is employed to test the potential confounding role of unobservables. Our analysis suggests that composite fuel deprivation indicators may be useful energy policy instruments for uncovering the underlining mechanism via which fuel poverty may get "under the skin".