{
  "abstract": "Background The male-female health-survival paradox describes the phenomenon that women tend report worse health than men despite the fact that men have higher mortality rates. This paper attempts to answer a question that has plagued the literature on this topic: can over-reporting by women or under-reporting by men explain this paradox? We compare gender differentials in subjective self-reported conditions and objective physiological functioning from health assessment data to answer this question.Methods Our analysis uses data from the first wave of TILDA (2009–2010), a nationally representative cohort study of ageing in Ireland involving 8,171 people aged over-50 participating in the personal interview. Data was collected through personal interviews, health assessments and self-completion questionnaires. We compare self-reported with objectively measured health for four health conditions (hypertension, high cholesterol, diabetes, and osteoporosis).We use Poisson regression models to estimate the effect of gender-by-age on the prevalence of health conditions. The models are estimated with and without adjustment for social factors. The estimates from the Poisson regression models are used to estimate the predicted marginal effect of sex on the prevalence rate ratio (PRR) from ages 50–85, using men as the reference category.Results For subjective hypertension, there are minimal gender differentials. For objective hypertension, women have a lower prevalence rate of hypertension than men at age 50 with a PRR of 0.84 (0.81–0.88) before it converges with men at age 76.There are no statistically significant differences between men and women for self-reported high cholesterol. Measured objectively, there is higher prevalence of high cholesterol among women at all ages with a PRR of 1.08 (1.03–1.13) at age 50 that grows across the lifespan. At age 85, the PRR is 1.14 (1.05–1.23).For both objective and subjective measures of diabetes, women have lower prevalence rates. Subjectively, these differences remain statistically significant throughout the lifespan. When measured objectively, the PRR at age 50 is 0.97 (0.94–0.99). The disparity remains statistically significant until age 80.For objective and subjective measures of osteoporosis, women have higher prevalence rates than men at all ages. This differential increases with age: at age 50, objectively measured osteoporosis shows a PRR of 1.02 (1.00–1.03) which grows to a PRR of 1.16 (1.08–1.24) at age 85.Conclusion There remain gender differentials in objectively measured health conditions despite adjustment for socio-demographic, lifestyle-related factors, and healthcare utilisation. This is not always the case for self-reported conditions, implicating constituent biological factors as an explanation.",
  "authors": [
    {
      "affiliations": [
        "Medical Gerontology, The Irish Longitudinal Study on Ageing, Trinity College Dublin, Dublin, Ireland"
      ],
      "name": "Katie Duffy"
    },
    {
      "affiliations": [
        "Medical Gerontology, The Irish Longitudinal Study on Ageing, Trinity College Dublin, Dublin, Ireland"
      ],
      "name": "Cathal McCrory"
    },
    {
      "affiliations": [
        "Medical Gerontology, The Irish Longitudinal Study on Ageing, Trinity College Dublin, Dublin, Ireland"
      ],
      "name": "Rose Anne Kenny"
    }
  ],
  "title": "OP26 Does reporting bias explain the male-female health-survival paradox? Evidence from the Irish longitudinal study on ageing (TILDA)",
  "uid": "ed1982da-7ddf-5ca9-8baa-4bd813e51a48"
}