Self-Assessment of Mental and General Health Status in Population Surveys : Strengths and Limitations

« If the notion of health is known by everyone, not so many people would be able to give a clear definition of this concept. For many years, the concept of good health was equated with the reductive concept of « organ silence », but in 1946, the WHO described it as “a state of complete physical, mental and social well-being and not merely the absence of disease and infirmity”. As it can be seen, it is not limited to physical issues, but it stresses the importance of mental problems or “unconscious” diseases. Hence, during the past decades, mental health knowledge has grown, because firstly, it is really often hard to diagnose any kind of mental issue than a physical issue, and this is the case for many reasons: symptoms are less obvious and not detectable via techniques such as MRI or x-rays for example, it includes a main part of subjectivity by the patient, etc… Also, patients affected by mental disorders will not only suffer from unhappiness but will also be more likely to have physical diseases. In order to increase public awareness of this topic, many political decisions have been decided, for example, the « World Mental Health Day » has been celebrated since 1992 on October, 10th.

Despite the fact that developed countries have a high life expectancy for their population, large inequalities still exist depending on social groups. For example, we observe a 7-year gap between engineers and workers and a 13-year gap between the lowest 5 percent and the highest 5 percent in terms of revenue. And unfortunately, these gaps are increasing, even in countries like France, where the health system is supposed to be really efficient. If the first common explanation that one can think about is inequalities in access to healthcare, this argument is not enough, as sources of financial aid are available, as differences in terms of behaviors should also be minded.

However, the main limitation of this focus is the difficulty to determine one’s health. But what is good health? Even if WHO has given a definition that holds since 1946, it mostly shows that it has lots of dimensions: physical, physiological, mental, and even within these concepts the notion of good health is not obvious. Realizing numerous medical checkups for every patient is difficult and can be very expensive, in order to detect those who suffer from any kind of health issues. Otherwise, several approaches exist for health measurement: it can be objective by using known criteria like mortality rates for example, or subjective. Many questionnaires have been developed in order to integrate the health dimension into general population surveys without necessarily having recourse to a doctor’s expertise or a medical diagnosis. In these questionnaires, individuals are asked to make a subjective assessment of their health status. These are called SAH (Self-Assessed Health), and they are composed of very simple questions such as “In general, would you say your health is:” with simple responses from “Very Good” to “Very Poor”. Nevertheless, their quality has been proven, as it has been shown that they very frequently present efficient results concerning the actual health of the patients. Effectively, the topics discussed by these questionnaires are multiple (physical, emotional, social, etc…), so depending on the answers given by the patient, physicians will be able to determine whether or not he should be given healthcare. Similarly, a very short questionnaire called MINI has been developed concerning mental health.

Although it is an efficient and particularly economical tool, studies have been able to highlight heterogeneities in the reporting of subjective health status, i.e. differences between individuals or social groups in the way they assess their own health: for example, on a daily basis, farmers are more subjected to pain than other people. All of these elements are likely to bias the results. Many studies have addressed this problem of reporting heterogeneity and some, such as Shmueli (2003), have proposed methods to deal with the biases associated with these reporting heterogeneities.

Therefore, the aim of the project is to improve the understanding of measures of perceived health and mental health in general population surveys and to understand how to account for the reporting heterogeneities associated with them.

To do so, the rest of our work is as follows: the first part presents a review of the literature, then we present the data we used in our study. The third part presents some descriptive statistics and the fourth and fifth parts present the methodology adopted in our study as well as our results. »