All Behavioural Interventions Carry Risks



sharma-obesity-kids-scale1Following the recent release of the Canadian Task Force on Preventive Health Care guidelines for prevention and management of adult obesity in primary care, the Task Force yesterday issued guidelines on the prevention and management of childhood obesity in the Canadian Medical Association Journal (CMAJ).

Key recommendations include:

  • For children and youth of all ages the Task Force recommends growth monitoring at appropriate primary care visits using the World Health Organization Growth Charts for Canada.
  • For children and youth who are overweight or obese, the Task Force recommends that primary health care practitioners offer or refer to formal, structured behavioural interventions aimed at weight loss.
  • For children who are overweight or obese, the Task Force recommends that primary health care practitioners not routinely offer Orlistat or refer to surgical interventions aimed at weight loss.

The lack of enthusiasm for the prevention of childhood obesity is perhaps understandable as the authors note that,

“The quality of evidence for obesity prevention in primary care settings is weak, with interventions showing only modest benefits to BMI in studies of mixed-weight populations, with no evidence of long-term effectiveness.”

leading the Task Force to the following statement,

“We recommend that primary care practitioners not routinely offer structured interventions aimed at preventing overweight and obesity in healthy-weight children and youth aged 17 years and younger. (Weak recommendation; very low-quality evidence)”

Be that as it may, the Task Force does recommend structured behavioural interventions for kids who already carry excess weight based on the finding that,

“Behavioural interventions have shown short-term effectiveness in reducing BMI in overweight or obese children and youth, and are the preferred option, because the benefit-to-harm ratio appears more favourable than for pharmacologic interventions.”

What caught my eye however, was the statement in the accompanying press release which says that,

“Unlike pharmacological treatments that can have adverse effects, such as gastrointestinal problems, behavioural interventions carry no identifiable risks.” (emphasis mine) 

While I would certainly not argue for the routine use of orlistat (the only currently available prescription drug for obesity in Canada) in children (or anyone else), I do take exception to the notion that behavioural interventions carry no identifiable risks – they very much do.

As readers may be well aware, a large proportion of the adverse effects of medications is attributable to the wrong use of these medications – problems often occur when they are taken for the wrong indication, at the wrong dose (too high or too low), the wrong frequency (too often or too seldom), and/or when patients are not regularly monitored. In a perfect world,  many medications that often lead to problems would be far less problematic than they are in the real world.

Interestingly, the same applies to behavioural interventions.

Take for example diets – simply asking a patient to “eat less” can potentially lead to all kinds of health problems from patients drastically reducing protein, vitamin and mineral intake as a result of going on the next “fad” or “do-it-yourself” diet. Without ensuring that the patient actually follows a prudent diet and does not “overdo” it, which may well require ongoing monitoring, there is very real potential of patients harming themselves. There is also the real danger of promoting an eating disorder or having patients face the negative psychological consequences of yet another “failed” weight-loss diet. Exactly how many patients are harmed by well-meant dietary recommendations is unknown, as I am not aware of any studies that have actually looked at this.

The same can be said for exercise – simply asking a patient to “move more” can result in injury (both short and long-term) and coronary events (in high-risk patients).  Again, ongoing guidance and monitoring can do much to reduce this potential harm.

In short when patient apply behavioural recommendations at the wrong dose (too much or too less), wrong frequency (too often or too seldom), and/or are not regularly monitored, there is indeed potential for harm – I would imagine that this potential for harm is of particular concern in kids.

This is not to say that we should not use behavioural interventions – we should – but we must always consider the potential for harm, which is never zero.

I’d certainly be interested in hearing from anyone who has seen harm resulting from a behavioural intervention.

@DrSharma
Edmonton, AB