60% of the Sheffield adult population is overweight or obese. It’s a big deal, inaction isn’t really a viable option . Many remain obsessed by addressing obesity by weight management interventions. It’s countable, we can be seen to be doing something.
However it perpetuates the notion that obesity can be solved by treating individuals.
Here I am not saying weight management isn’t important, I’m going to suggest the population impact is limited. Further investment in weight management is welcome. But not if it as at the expense of more valuable things.
Some numbers may illustrate the population impact.
The basic numbers
In Sheffield we invest £300k in tier 2 weight management. Given that 1% of the population that lives in Sheffield, assuming we are average that’s £30m in England.
There are about 300k in Sheffield overweight or obese (ie eligible).
Last year (17/18) 880 people were referred to tier 2 weight management, 805 took up the referral and 732 completed the pathway. Of those that attended, 46% attend from bottom 2 deprivation quintiles, 20% male, 80% female.
For £300k we can provide treatment pathway for 732 / 300,000 people, or 0.24% of the target population.
How long will it take to solve obesity through investing in weight management
If we assume the treatment is 100% successful then at this volume and level of investment, it will take address the problem by one quarter of 1 percent in a year. If prevalence is 60% that will take 240 years to get population prevalence to zero.
But it will take longer, the treatment doesn’t work for many, especially out in the long term.
But it’s not a 100% successful treatment. We know that the probability of getting to normal weight is slim, and 5% loss is generally regarded as a successful outcome clinically. Our own data suggest that of those that complete (732), 84% reduced BMI (617), 60% lose 3% (432) and 20% lose 5% (142) by the end of a 12 week course.
Our long term follow up data is poor quality (huge loss to follow up and reliance on self reporting) we know we need to improve this. Commercial providers may have some decent longer term data for their programmes. In the POWER study just under a third of those using the online tool maintained 5% weight loss to 12m.
Of course we then know that at 5 years the chance of keeping that initial 5% weight loss is further attenuated. This 2015 U.K. study considered the probability of an obese person attaining normal body weight after using community based weight management interventions. Out to a max of 9 years follow up, the annual probability of attaining normal weight was 1 in 210 for men and 1 in 124 for women. This underscores the limitations of individually oriented interventions.
There are parallels for those who stop smoking, in Sheffield the 4 week quit rate is around 50 -55%, at 12 weeks quit rate is around 70% & we estimate at 12 months the quit rate is 15-20% at 12 months. The moment those who stop smoking they are in a world where smoking is generally inconvenient and not the norm. For weight management clients have to go back into a world where sitting down and eating cheap, calorie laden food is extremely normal and convenient. Stopping smoking is binary – you’ve stopped or you haven’t.
There’s also the issue of demand. The above only applies to those people who recognise themselves as overweight, see it as something to worry about and are motivated and able to do something about it. Only a small proportion of the target population will ever access weight management services.
So, onto the population impact
So of the 300,000 people eligible, in a year – £300k provides treatment for 732, or 0.24% of the eligible population of whom 20% or 142 loose clinically relevant weight. (Even if I was kind and used 30% at 12m that’s still only 220 people). So £300k enables us to change population prevalence of obesity from 300,000 people to 299,876 people.
It might take a while to make a significant dent in population prevalence at this rate.
So additional investment is welcome but isn’t the thing that will get anywhere near making a difference. Emptying an ocean with a teaspoon?
Individual even interventions may never have the population impact some hope they will unless we massively scale up coverage, probably to tens of thousands of people. Currently our investment of £300k provides completed treatment pathways for 732 people, or £409 per completer. So to get to 20,000 treated we’d need in the order of £8,100,000. Everyone wants scale up, as long as someone else is paying.
Getting better impact through greater coverage
the POWER study suggests brief intervention plus online support, with a bit of remote support from a practitioner along the way, seems effective.
818 eligible individuals in the study. Of the 818, 80% (666) had lost weight at 12 months
By 12 months 32.4% of POWeR+R participants maintained a 5% weight reduction. Interestingly the control group had a brief intervention (by a practice nurse I think), were signposted to some online info about food swaps etc. and were weighed a couple of times over 12m by the practice nurse. 20% of those had lost 5% by 12m, not bad given the minimal investment.
This is for a cost per licence of £2k per year per district for unlimited users
Moving further upstream
I’m not arguing against weight management. It is necessary but not sufficient.
Success in tobacco has been based on following the MPOWER model tackling the issue simultaneously from a wide range of angles. The same approach is needed in obesity, the Foresight report highlighted the issues neatly. However if we focus all our resources on treatment we will have an unbalanced approach.
Those upstream actions (at local and national level) need people to ensure they happen. Of note we are disinvesting in weight management in Sheffield in order to invest in activities further upstream.
A medicalised response to a social and environmental problem is not good enough.
The obesity epidemic is categorically *not* driven by greed or laziness, and that stigmatising perspective is both harmful and utterly wrong. Obesity is grounded in changes to physical, social, policy, commercial and other environments.
We all know that upstream interventions around the food and physical activity environment populations live in are hugely more potent than downstream behaviour change focused interventions delivered one person at a time. I’ve written before on the relative impact of individual vs population interventions. It is fair to say that those representing commercial interests, especially in the food industry, seem very happy to maintain a focus on individually oriented interventions. It takes the focus of policy attention away from addressing the upstream commercial causes (and profit margins).
Thanks to Jess Wilson, Sarah Hepworth & Toni Williams for some thoughts on an earlier draft.
After I posted the original I got a fair degree of stick from people I think assuming I was saying that we shouldn’t invest in weight management.
This wasn’t the case, as I was at pains to state in the post. I was making a point about population impact.
A number of studies cropped up in the twitter conversation that followed which sort of make the same point I was making.
- Of 1·8 million eligible individuals,
- 238 540 (13%) participated in the MOVE! programme.
- 19 367 (1% overall, 8% of participants) met criteria for intense and sustained participation
- which was associated with greater weight loss at 3 years than low-intensity or no participation (−2·2% vs −0·64% or 0·46%).
- Compared with non-participation, incidence of diabetes was reduced by intense and sustained participation (hazard ratio 0·67, 95% CI 0·61–0·74) and low-intensity participation (0·80, 0·77–0·83) in MOVE!.
- So for those that participate fully in the intense programme, incidence of diabetes was reduced by a magnitude similar to the the DPP studies. 1% of eligible individuals participate in the intense programme. There may be selection bias.
- 23 primary care practices
- The primary outcome was weight at 1 year of follow-up, followed up to 2 years
- enrolled 1269 participants. 1267 eligible participants were randomly assigned to the brief intervention (n=211), the 12-week programme (n=528), and the 52-week programme (n=528).
- At 1 year, mean weight changes in the groups were −3·26 kg (brief intervention), −4·75 kg (12-week programme), and −6·76 kg (52-week programme).
- The 52-week programme was more effective than the 12-week programme (−2·14 kg, −3·05 to −1·22; p<0·0001).
- Differences between groups were still significant at 2 years.
- £159 per kg lost for the 52-week programme , £2394 / QALY modelled our to 25y
- £91 per kg for the 12-week, £3084 / QALY modelled our to 25y
- Median weight change for all referrals was -2.8 kg [IQR -5.9 – -0.7 kg] representing -3.1% initial weight. 33% of all courses resulted in loss of ≥5% initial weight.
- 54% of courses were completed.
- Median weight change for those completing a first course was -5.4 kg [IQR -7.8 – -3.1 kg] or -5.6% of initial weight. 57% lost ≥5% initial weight.
- A third of all patients who were referred to WW through the WW NHS Referral Scheme and started a 12 session course achieved ≥5% weight loss,
(was criticised by some as being a shoddy review, use your judgement)
suggesting ineffectiveness of commercial weight-loss programs for achieving modest but meaningful weight loss
- 57% of individuals who commenced a commercial weight program lost less than 5 percent of their initial body weight.
- 49% studies reported attrition ≥30 percent.
- 37% of program completers lost less than 5 percent of initial body weight.
- Concluded that commercial weight-loss programs frequently fail to produce modest but clinically meaningful weight loss with high rates of attrition suggesting that many consumers find dietary changes required by these programs unsustainable
To repeat our own data – we think c20% lose 5% by the end of a 12 week course.
We know we can improve this.
I’ve seen data from another place suggesting 40% losing 5% at 12wks
There’s no routine national benchmarking though as all contracts are slightly different. Even at 40% my central argument still stands.