Is a BMI of 17.3 healthy?

Well, if you are inpatient in Cheshire, apparently  it is.  It is a perfectly acceptable "maintenance weight", despite the patient being diagnosed 5 months ago at a BMI of 17.

The likelihood of a severe relapse and a readmission at this weight is pretty certain.

Sigh.  Another rant about how BMI is not a diagnostic tool but it seems impossible to get anyone to listen.  I have found a wonderful paper from the Council of the Royal College of Psychiatrists, that would appear to refute the consultant's claim that a BMI of 17.3 "is fine".  I wonder if he is a member and would like to take it up with the Council?  Or perhaps the Council would like to take it up with him?  Then again, pigs may fly.

Refeeding

In chronic starvation the energy requirement is depressed. It is therefore possible to promote weight gain with a relatively low energy intake at first and increase it gradually. An individualised approach may be best for those not being treated in a specialist eating disorders unit. In specialist units, a standardised programme can be used. A weekly weight gain of 0.5–1.0 kg is generally regarded as optimum and an intake of 2200–2500 kcal (9200–10 000 kJ) daily will achieve this in most patients. The rate of gain will slow down as weight increases, owing to an increase in metabolic rate and physical activity. Vegetarian diets can normally be accommodated without difficulty. Vegan diets present particular problems in achieving adequate energy intake and provision of sufficient phosphate.

It is a common practice to set a target weight at the beginning of treatment.  There is no clear consensus as to how this should be determined, but it is often set at a body mass index (BMI) of 19–20 kg/m2. However, this represents a minimum healthy weight based on population norms and is not necessarily the optimum weight for the individual. At present, there are inadequate data on which to base recommendations about healthy weights in specific ethnic groups.

Target weight

It is a common practice in many units to set a target weight at the beginning of treatment. This gives definition to the treatment programme and may help to allay the patient’s anxiety about being allowed to become overweight. There is no clear consensus as to how the target weight should be determined. A reasonably

common practice is to base it on a low normal body weight, such as a BMI of 19 kg/m2  or 20 kg/m2. This may have to be modified in the light of individual circumstances, for example if the patient’s premorbid stable weight was significantly higher or lower than this. It may sometimes be appropriate to agree a lower target weight, for example as part of a specialised treatment plan or in intractable cases where the patient has repeatedly failed to attain a normal weight.

In some units a target weight range is used in preference to a single weight.  A BMI range of 19–25 kg/m2  is accepted as ‘healthy’ in European and North

Although setting the target at a normal weight results in lengthy admissions, clinical experience suggests that discharge before this point may allow the patient to avoid the difficult psychological transition to a normal weight. There is limited research evidence that discharge at a low weight is associated with a poorer outcome and a higher readmission rate (Baran et al, 1995; Howard et al, 1999). It should be emphasised that the target represents a minimum healthy weight rather than an ideal. The return of menstruation may be used as a physiological marker of adequate weight restoration, although patients should be informed that this may be delayed for several months after attaining a normal weight. An alternative approach is to ask the patient to attain a BMI of 19 kg/m2  and then begin sequential ovarian ultrasound examinations; weight gain is then continued until a dominant follicle is observed. This approach may avoid the arguments with patients which often occur when professionals define the ‘healthy’ weight.