News in brief

Apparent age of the elderly : a diagnostic tool for primary care physicians?

23 Apr 2018 | By Inserm (Newsroom) | Public health

To estimate the health of their elderly patients, healthcare professionals have various clinical tools at their disposal: geriatric assessment scales, health questionnaires and frailty scores. However, although these tests are appropriate to a medical context, they are more difficult for primary care physicians (PCPs) to implement because they require time and equipment that they do not necessarily possess. When evaluating the general health of an elderly patient, the PCP will first go on their detailed knowledge of him or her as well as the various signs observed during the consultation (physical appearance, gait, etc.).

The objective of the S.AGES cohort[1] was to describe between 2009 and 2014 the medical and paramedical management of patients over the age of 65 using a database generated by PCPs. Practitioners throughout France were enrolled to participate in this study involving the 3-year follow-up of some of their non-institutionalized patients presenting with a satisfactory state of health. In addition to collecting social, demographic and medical data, each PCP was asked to categorize their patients by their apparent age in response to the following question: “In your opinion, does your patient look younger or older than their age or do they look their age? “.

It was on the statistical analysis of the responses to this question that Inserm researchers from the Center for Research in Epidemiology and Population Health focused their attention. The objectives of this study were to analyze the characteristics associated with the PCPs’ estimations and to determine whether these made it possible to predict death within the 3-year follow-up period

The researchers observed that heart problems, depression, obesity, and low levels of autonomy in activities of daily living were primarily linked to patients whose estimated age was higher than their actual age, whereas high levels of education, normal levels of autonomy, lack of depression, lack of vascular problems and chronic hypertension, and low levels of medication were associated with patients who looked younger than their age. Only two variables impacted apparent age in either direction: quality of autonomy and the presence/absence of depression.

Of the patients who died during the 3 years of follow-up: 7.5% belonged to those who looked younger, 8.7% to those who looked their age and 13.8% to those who looked older. It can therefore be taken away from this study that patients who look older present an increased risk of mortality: the age estimation performed by the PCPs is therefore significantly linked to mortality risk.

These results suggest that such age evaluation could be represent a relatively reliable tool in the daily practice of PCPs when estimating the general health of patients. This does not mean replacing the frailty scores but using the practitioner’s clinical sense to support the diagnosis.

[1] The S.AGES cohort was funded by Sanofi and approved by the Ile de France XI ethics committee and the French medicines and health products safety agency

These contents could be interesting :

Researcher Contact

Virginie Ringa

Chercheuse Inserm

Directrice de l’équipe « Genre, santé, sexualité »

Unité 1018 Centre de recherche en épidémiologie et santé des populations

+33 (0)1 45 21 23 37


Sophie Bucher

Unité 1018 Centre de recherche en épidémiologie et santé des populations

Equipe « Genre, santé, sexualité »

+33 (0)1 45 59 60 45


Press Contact



Looking younger, dying later: General practitioners’ intuitive clinical
impression predicts mortality

Sophie Bucher a,b, Abdallah Al-Salameh b,c, Henri Panjo b, Laurent Becquemont c, Virginie Ringa b, The S.AGES investigators

a General Practice Department, Paris-Sud Faculty of Medicine, University of Paris-Sud, Le Kremlin-Bicêtre, France
b CESP, INSERM U1018, INED, Université Paris-Saclay, Université Parie-Sud, UVSQ, Villejuif, France
c Pharmacology Department, Faculty of Medicine Paris-Sud, University Paris Sud, UMR 1184, CEA, DSV/iMETI, Division of Immuno-Virology, IDMIT, INSERM Center for immunology of viral infections and autoimmune diseases, Assistance Publique-Hôpitaux de Paris, Hopital Bicêtre, Le Kremlin Bicêtre, France