Middle-aged spread and mortality

29 may 2009--“People who are overweight or obese in middle-age run the risk of being frail in later life,” reported BBC News. It said a study found that men who put on weight in their 40s but lost it when they got older were at the highest risk of death in their 70s. It quoted the study leader as saying, “the unhealthy pattern of weight in their 40s was causing frailty in later life probably due to underlying cardiovascular problems, such as high blood pressure and early stages of diabetes”.

This study had several limitations that restrict its reliability. In particular, it is important to note that it was not necessarily losing the excess weight they carried in their 40s that raised the men’s risk of death. Instead, it may be that these men lost weight because they had undiagnosed health problems, or because of other factors that the study did not investigate. The authors themselves note that “life-long normal body weight is the best option”, and the study’s findings should not be interpreted as encouragement to maintain an unhealthy weight.

Where did the story come from?

The research was carried out by Dr Timo E Strandberg and colleagues from the University of Oulu and other universities and research centres in Finland. The study was funded by Päivikki and Sakari Sohlberg Foundation, the Helsinki University Central Hospital and the Finnish Foundation for Cardiovascular Research. The study was published in the peer-reviewed European Heart Journal.

What kind of scientific study was this?

This study was a new analysis of data collected in a previous cohort study called the Helsinki Businessmen Study. This study involved initially healthy men, mostly business executives, who were born between 1919 and 1934. They started in the study in the 1960s and 70s, and had structured health check-ups as part of the study. The current study aimed to look at how changes in body mass index (BMI) throughout life affected death rates in old age. In particular, they wanted to look at how risk factor for cardiovascular disease (such as obesity) in midlife affected death rates.

In 1974 the 1,815 healthy middle-aged participants (average age 47 years) were examined, had their current height and weight measured, and were asked to recall their weight at age 25. They were also asked to rate how they perceived their health on a five-point scale, ranging from very good to very poor. Men with a history or signs of chronic diseases such as diabetes, high blood pressure or heart problems were not included in the study. Overweight was defined as a BMI (weight in kilogrammes divided by height in metres squared) over 25kg/m² and normal weight defined as a BMI of 25kg/m² or less.

In 1985-6, 909 of the men were again assessed, and a measurement taken of their BMI and waist circumference.

In 2000, at an average age of 73 years, all participants that were still alive (1,390 men) were sent questionnaires about their health, present weight, lifestyle (including smoking and alcohol consumption) and demographic factors, and whether they had a history of chronic diseases. This information was used to calculate a standard index that showed how many concurrent medical problems (comorbidities) the men had. Their health was assessed using a standard scale that gave summary scores for overall physical and mental health.

Data on BMI at age 25, and in 1974 and 2000, were available for 1,114 men (61% of original participants, 80% of those surviving to 2000), and these men were included in the analysis. The men were grouped according to their weight pattern from 1974 to 2000: those who had normal weight at both times (345 men), those who were overweight at both times (494 men), those who were of normal weight in 1974 but overweight in 2000 (136 men) and those who were overweight in 1974 but normal weight in 2000 (139 men). At the end of 2006, the researchers used a national population registry database to identify those men that had died and the causes of their deaths. They used statistical methods to look at whether BMI change from 1974 to 2000 was associated with risk of death. These analyses took into account smoking and the men’s perceived health at the start of the study, and self-reported history of disease in 2000.

What were the results of the study?

Of the 1,815 healthy middle-aged men evaluated at the start of the study, about 24% (425 men) had died by 2000. Men who were overweight at the start of the study were more likely to die in this period (about 26%) than those who were of a normal weight (20%).

Among the 1,114 participants with full data from both 1974 and 2000, almost half (44%) were constantly overweight, 31% were constantly of normal weight, 12% became overweight and 12% were overweight in midlife (in 1974) but became a normal weight by their 70s (in 2000). From 2000 to 2006, 188 of these men died (17%). The actual number of men who died in each group was not reported, but deaths were more common in the group of men who had gone from overweight in midlife to normal weight in their 70s than among men in the other groups. Men in the group that lost weight were about twice as likely to die between 2000 and 2006 than men who stayed a normal weight.

The other groups (those that remained overweight and those that became overweight) did not differ significantly from the group that remained a normal weight. These results were largely unchanged by making adjustments for age, smoking, perceived health in 1974 and self-reported diseases in 2000.

What interpretations did the researchers draw from these results?

The researchers conclude that men “who had normal weight in late life, but had been overweight in midlife, had the greatest mortality risk in old age. In contrast, the risk of those men who did not become overweight until after midlife did not differ from that of men with constantly normal weight”. They say that this may “suggest that cardiovascular risk factors may increase risk of frailty” and that their findings “support the implication that some weight gain may be beneficial for those who are not overweight in early adult life”.

However, they say that if deaths before later life are taken into account, men of a normal weight were at a lower risk of death than overweight men, and that “life-long normal body weight is the best option”.

This study has a number of limitations:

• As with all studies of this type, it is possible that factors other than BMI change (known as confounders) were responsible for the difference seen. Although the authors took some potential confounders into account, these were not assessed in a very thorough way (for example, smoking was assessed only once and amount smoked was not assessed) and this may have reduced the ability of these adjustments to remove their effect. There may also have been other unmeasured and unknown confounders having an effect.
• It is possible that there were some inaccuracies in weight and health assessment in the study. For example, the men may not have been able to remember accurately their weight at age 25, and, in 2000 when men had to report their own weight, these measurements may not have been accurate. Men also self-reported any diagnosed health problems in 2000, and these reports may not have been accurate.
• The men were divided into four weight categories based on measurements of their weight on two occasions, 27 years apart. This is a relatively crude measure of weight change over a period this long, and within these categories the men’s weight may have fluctuated in different ways between these two times, which may have influenced the results.
• This study included only men who were healthy in midlife and who were largely businessmen. The results may not apply to women, men in different socio-economic groups or men who are not healthy in midlife.
• Regarding their analysis of “frailty” (the analysis that adjusted for self-reported disease in 2000), the authors themselves state that this analysis was “inconclusive and principally aims to be hypothesis-generating for future studies”. Therefore, no firm conclusions can be drawn about the effect of BMI on frailty.
• In addition, about a quarter of men who were overweight at the start of the study (in 1974) had already died by 2000, and had these been included in the group that were “constantly” overweight between 1974 and 2000, this may have affected the results.
• The suggestion by the authors that “some weight gain may be beneficial for those who are not overweight in early adult life” is not supported by the results. Those who were of normal weight in midlife and who became overweight in later life did not differ in their risk of death to those who remained a normal weight. This does not indicate that gaining weight is “beneficial”. In addition, death is not the only negative outcome associated with being overweight. Men who became overweight between 1974 and 2000 were more likely to report high blood pressure, diabetes, congestive heart failure and cerebrovascular disease, among other diseases, than men who maintained a constantly normal weight. Again, this does not suggest any “benefit” from gaining weight.

The points outlined above limit the reliability of the findings, which will need to be confirmed in further research. It is important to note that it was not necessarily losing excess weight they had carried in their 40s that put the men at risk of worse outcomes. Instead, it may be that these men lost weight because they had, as yet undiagnosed, health problems. The study’s findings should not be interpreted as encouragement to maintain an unhealthy weight or to gain weight.

Links to the headlines

Middle age spread link to frailty. BBC News, May 26 2009

Links to the science

Strandberg TE, Strandberg AY, Salomaa VV et al. Explaining the obesity paradox: cardiovascular risk, weight change, and mortality during long-term follow-up in men. European Heart Journal 2009; Advance Access published online on May 9