Underestimation of health status

I am strongly convinced that health surveys used to estimate morbidity differ from objective measures. Such large differences are unknown and too often health policy and planning is exclusively based on self-assessed measures. A recent chapter in the book "Active ageing and solidarity between generations in Europe: First results from SHARE after the economic crisis" confirms my impression. Why is this so?. The authors say:

"Being female, older or highly-educated implies a lower probability to underestimate health, and this probability is higher if people are wealthier and have confidants in their social network. Besides, people are more likely to overestimate their health if they are older or wealthier; on the contrary, this probability is lower if they are homeowners or have someone in entourage to talk to."