Dr. Sharma's Obesity Notes

According to a landmark paper in this week’s issue of JAMA, I now realize that I need to gain about 75 lbs to get the most out of my pension plan.

Indeed, that is just about how many extra pounds I would need to pack on to achieve a BMI of 33, which would finally put me squarely in the Class 1 obesity range. That, according to this extensive review of the literature, would bestow me with the longest life expectancy.

Unfortunately, this will not be easy – If I recall correctly, Morgan Spurlock (the fellow in Super Size Me) only managed to gain a measly 25 lbs – and I recall how hard he had to work at this.

Or, could it perhaps be that the results of this paper are so obviously nonsensical, because the researchers asked the wrong question?

After all, who still cares about BMI?

I can only assume that my US colleagues were far too busy running their analyses to have time to read the Canadian Medical Association Journal (CMAJ or Canada’s version of JAMA), which happened to publish our analysis of data from the US(!) National Health and Nutritional Education Survey (NHANES) on this issue.

As readers of these pages may recall, our analysis of applying the Edmonton Obesity Staging System (EOSS) to two independent NHANES datasets, essentially showed that when it comes to mortality, what matters is how “sick” your are and not how “big” you are.

If you have a weight-related health problem (i.e. EOSS 1+), you die, if not, you don’t – end of story!

Neither BMI nor waist circumference were much use in predicting mortality – but whether or not you had hypertension, diabetes, or sleep apnea was.

As we outline in our paper, not only would BMI overestimate health problems in millions of US citizens, it would also completely miss about 25 million Americans, who do have weight-related health issues, despite falling well below the BMI 30 obesity range.

Perhaps, after this paper, we can finally lay BMI to rest and stop trying to predict people’s health with just scales and measuring tapes.

Hopefully, the only landmark that this paper leaves behind is a tombstone – BMI – RIP!

Let us now get back to actually taking a good medical history, doing a thorough physical exam, and running some tests before declaring someone too light or too heavy for their health.

And in the meantime, let’s not forget that prevention best starts by not losing sleep over your weight unless you have to.

AMS
Edmonton, Alberta

Excess weight is often (but not always) associated with health problems, and the latter can certainly often contribute to weight gain – even in kids!

A study by Stefan Kuhle and colleagues from the University of Alberta, published in the Archives of Diseases of Childhood, shows that overweight and obese Canadian kids use more medications than do normal weight kids.

The authors compare medication use between normal weight and overweight children (n=2,087) in a nationally representative sample from the Canadian Health Measures Survey 2007/2009, a cross-sectional survey assessing indicators of health and wellness in Canadian.

While there was no difference in the frequency of prescription, over-the-counter and natural health product (NHP) medication use between normal weight and overweight/obese 6-11 year olds, overweight/obese 12-19 year olds reported the use of prescription medication about 60% more often than their normal weight peers, especially for nervous system and respiratory problems. On the other hand, they were about 50% less likely to report the use of NHP medications.

As the authors note,

“With an overweight/obesity prevalence of 28% in the sample and 59% higher medication costs, approximately 14% of drug expenditures in this age group may be attributed to overweight and obesity.”

Less concerning than the cost associated with this increased use of medications is the question of why these kids develop obesity and related health problems in the first place.

It is particularly noteworthy that the use of drugs for the treatment of obstructive respiratory disorders was almost twice as high in the overweight/obese kids than in their normal weight peers.

Given that asthma is notoriously overdiagnosed in obese adults, I wonder how much of these drugs may be overprescribed in these kids.

In my own experience in adults, many admit that symptoms of ‘asthma’ were often a convenient way of getting out of gym class (as were ‘menstrual’ cramps). Anecdotally, I have the impression that this was more commonly reported in my female patients, who often recall gym class during their peri-pubertal years as particularly unpleasant (the fact that they often had to wear bras before any of the other girls in their class did not exactly help).

Nevertheless, the numbers are concerning and certainly an indicator that overweight and obese kids may have more health problems than their peers.

I wonder how many of my readers can remember having to take asthma or other meds as kids and how these problems may have contributed to their current health status.

AMS
Edmonton, Alberta

Small Request: Maintaining this blog costs both time and money – if you continue to enjoy these posts, please visit my website to make a small donation to the upkeep of this site by clicking here.

Kuhle S, Fung C, & Veugelers PJ (2012). Medication use in normal weight and overweight children in a nationally representative sample of Canadian children. Archives of disease in childhood PMID: 22833408

.

Regular readers will have followed the track of studies demonstrating that BMI is but a poor measure of health. This is why we came up with the Edmonton Obesity Staging System, that defines obesity based on the actual presence of physical, mental, and/or functional problems and limitations. Readers will recall, that we have previously shown that Obesity Stage – in contrast to BMI – is a powerful predictor of CV and all-cause mortality.

Now, yet another study, which looks at the relationship between BMI and mortality, again shows that ‘metabolically healthy’ obese individuals do not have an increased risk for CV mortality and in fact have a lower risk than ‘metabolically unhealthy’ non-obese individuals.

This study by Mark Hamer and Emmanuel Stamatakis from University College London, UK, published in the Journal of Clinical Endocrinology and Metabolism, looks at data from an observational study with prospective linkage to mortality records in community-dwelling adults from the general population in Scotland and England.

The sample includes 22,203 men and women [aged 54.1 y, 45.2% men] without known history of CVD at baseline followed for an average of about 7 years.

Based on blood pressure, high-density lipoprotein-cholesterol, diabetes diagnosis, waist circumference, and low-grade inflammation (C-reactive protein ≥ 3 mg/liter), participants were classified as metabolically healthy (0 or 1 metabolic abnormality) or unhealthy (two or more metabolic abnormalities). Obesity was defined as a body mass index of 30 or greater.

Compared with the metabolically healthy nonobese participants (about 20% of the obese population), their obese counterparts were not at elevated risk of CVD or all cause mortality.

On the other hand, both non-obese and obese participants with two or more metabolic abnormalities were at elevated risk.

Not surprisingly, metabolically unhealthy obese participants were at elevated risk of all-cause mortality compared with their metabolically healthy obese counterparts.

These data fit nicely with our findings using the obesity staging system. Thus, in our studies of the NHANES and Cooper Longitudinal Fitness Study, we found that Stage 0 and 1 obese individual (i.e. those with no or pre-clinical risk factors) did not have an increase risk compared to Stage 2+ individuals (those with comorbidities and/or endorgan damage).

Consistent with the present study, we also found that about 50% of non-obese individuals (in the BMI 25-30 range), would in fact fall into the Stage 2+ categories based on metabolic or other abnormalities.

Thus, simply using BMI as an indication for treatment would ‘over treat’ a substantial number of ‘obese’ individuals while missing an even greater number of ‘non-obese’ individuals, who are at risk.

As pointed out in an accompanying editorial by JP Després,

“Thus, obesity assessed by the BMI cannot properly estimate CVD and all-cause mortality risk. Furthermore, the therapeutic objective of achieving a normal BMI to prevent/manage cardiometabolic diseases may also be questioned on the basis of the emerging evidence…..The study by Hamer and Stamatakis provides robust evidence that a paradigm shift is needed: obesity can no longer be assessed the old-fashioned way.”

This is why the 5As of Obesity Management recommend the assessment of Obesity Stage in addition to BMI prior to giving patients any advice on managing their weight.

AMS
Edmonton, Alberta

Hamer M, & Stamatakis E (2012). Metabolically healthy obesity and risk of all-cause and cardiovascular disease mortality. The Journal of clinical endocrinology and metabolism, 97 (7), 2482-8 PMID: 22508708

Després JP (2012). What is “metabolically healthy obesity”?: from epidemiology to pathophysiological insights. The Journal of clinical endocrinology and metabolism, 97 (7), 2283-5 PMID: 22774209.

The math is simple – of the one in five adults in most Western countries, who are now considered ‘obese’, very few of them were obese as kids or even as young adults.

In fact, most obese adults are 40 or older. And, despite the alarming increase in childhood obesity (now affecting about one in twelve kids), obesity rates in younger adults remains at about 10% or less. Indeed, the greatest increase in obesity is seen in 45-54 year old men.

These numbers alone, should tell us that most obese adults (and thus, the vast majority of obese people alive today), developed obesity as adults – not as kids!

This is not to say that excess weight in childhood is not a major predictor of excess weight in adulthood – many of my patients recall being teased and bullied about their weight 30 years ago, when they were growing up in rural Alberta, helping with chores around the farm, riding their bikes to school everyday, and playing shinny hockey on a frozen pond all winter.

But the majority of my patients did not have a weight problem till well into adulthood.

Why do I bring this up?

Simply because, I believe that better understanding, or even fully preventing, childhood obesity is unlikely to have a noticeable impact on adult obesity rates anytime soon.

The problems that lead to obesity for the vast majority of obese adults occurred during their adult years.

Their obesity was not caused by lack of phys-ed classes, poor school lunch programs, hallway vending machines, or parents too busy to cook dinners from scratch.

Their obesity was probably also not caused by too much video gaming, too much TV watching, or not playing outside till the lights came on.

Remember, the demographic group with currently the highest obesity rates (almost one in three) were kids in the 50s and 60s – an era, when a 6 oz serving of pop was considered a rare treat.

Why is any of this important?

1) Focussing all of our efforts solely on better understanding the drivers of childhood obesity and trying to prevent it likely means continuing to ignore the drivers of adult obesity, which account for the vast majority of obesity in the population.

2) Even if we successfully eliminate childhood obesity, by say, changing our kids’ lifestyles back to the lifestyles of kids back in the 1960s, we may still see obesity affecting as many people in 50 years as it does today.

3) Adults with obesity today will live another 30-40 years (or longer) – many more adults not obese today, will become obese tomorrow – this is why preventing and treating ADULT obesity must be the most important priority for any health care system.

4) While childhood obesity is certainly a risk factor for adult obesity – the main driver of childhood obesity is ADULT obesity – having just one obese parent markedly increases the risk for childhood obesity – having two obese parents is virtually a guarantee. Thus, any solution to childhood obesity must focus on the ‘root cause’ of kids’ excess weight, i.e. having obese parents – these parents need treatment.

5) As far as I can tell, the strongest ‘halo’ affect of treating adult obesity is on their kids (the most extreme examples of this come from the kids of parents who have undergone bariatric surgery). Simply stated: successfully treat the parents and you ‘automatically’ prevent obesity in the kids.

I am not implying that childhood obesity is not a concern and that improving the lifestyles of all our kids (irrespective of their shapes and sizes) should not be an important goal.

All I am saying is that we need to stop ignoring the adults if we hope to make any dent in the obesity epidemic in our lifetime.

Imagine if the obesity epidemic was due to a virus that mostly affects adults – would we just be vaccinating the kids?

AMS
Vancouver, BC

Regular readers will appreciate the importance that I have given to pain both as a driver and consequence of weight gain as well as a barrier to treatment.

Just how closely obesity and pain are associated is now documented by Arthur Stone and Joan Broderick from Stony Brook University, NY, in a paper just published in OBESITY.

This study is based on a Gallup ‘poll’ of 1,062,271 randomly selected US individuals surveyed between 2008 through 2010.

BMI and pain yesterday were reliably associated (even when adjusted for a wide range of demographic variables): the overweight group reported 20% higher rates of pain than Low-Normal group, 68% higher for Obese I group, 136% higher for Obese II group, and 254% higher for Obese III group.

All of the tests of association between the pain conditions and BMI groups were significant, with the strongest association for the knee and leg condition.

The association held for both men and women but in women, the trend to more pain as BMI increases was steeper than in men.

The association between BMI and pain increases moving from the younger categories to the older categories; for those in the Obese III group, the odds ratio for the youngest group is 1.72 compared with a ratio of 3.79 for those in the highest age group.

As the authors note,

“The association is robust and holds after controlling for several pain conditions and across gender and age. The increasing BMI-pain association with older ages suggests a developmental process that, along with metabolic hypotheses, calls out for investigation.”

Despite the possible limitations due to the nature of the survey (telephone, self-reported height and weight and pain levels, etc.), the relationship between higher weight and pain is striking.

Assessing for pain (the 2nd ‘M’ or ‘Mechanical’) should be routine part of any exam for obesity and may have to be tackled in any obesity management program.

AMS
Edmonton, Alberta

Stone AA, & Broderick JE (2012). Obesity and pain are associated in the United States. Obesity (Silver Spring, Md.), 20 (7), 1491-5 PMID: 22262163

.