Age and Obesity
In 1998, the World Health Organization (WHO) and the National Heart, Lung, and Blood Institute (NHLBI) issued comprehensive reports setting guidelines for the classification of overweight and obesity. These reports confirmed the well-known fact that weight increases with age (see Chapter 75). The key questions are then whether this is healthy, and how ideal weights should be determined for people at different ages.
The WHO and NHLBI reports agreed in recommending the use of body mass index (BMI) in contrast to weight-for-height tables, and in setting cut points at 25 and 30 kg/m2 to define the separation of normal from overweight, and overweight from obesity (see Chapter 68). They also agreed in recommending the same cut points for adults of all ages.
These recent proposals have largely replaced the use of the tables issued periodically for the past 60 years by the Metropolitan Life Insurance Company, most recently in 1983. Those tables, which provided separate ranges of weights for men and women according to their heights and “body frame,” for adults ranging in age from 25 to 59 years, provided no recommendations for those over age 60. A reanalysis of the insurance data upon which those tables are based showed that the relation of BMI to all-cause mortality was decidedly U-shaped (a quadratic function) in men and women for the decades of life ranging from 20–29 to 60–69 years. Furthermore, minimal mortality (the nadirs of the Ushaped curves) increased progressively, from about 20–27 kg/m2 over the 40-year age span. Thus, the question of the need for age-specific weight tables was raised and became a topic of controversy that has continued for the past 15 years.
The literature is now vast: The NHLBI Evidence Report cites 768 references, and the WHO Consultation on Obesity cites over 600. In the present chapter, I have limited my analyses to reports of all-cause mortality in older men and women, published since the reviews noted earlier, that is, during the years 1997–2000.
Obesity is associated with a remarkably broad spectrum of health problems, including cause-specific mortality and morbidity, and disability (see Chapters 76 and 84). Yet,