Preventing diabetes. Indivudal vs population based approaches – on emptying an ocean with a teaspoon

PHE and SSHaRR have published the diabetes prevention programme return on investment tool. It’s fantastic. Really like it and good effort to them!

I ran the numbers for my patch –
I hope I have interpreted correctly….and I would be happy to be corrected, just what I had a look at whilst I was having a sandwich at lunch.

DPP and return on investment of services delivered individual by individual 

Whilst I’ve suggested in these paramaters that the uptake is 32% of those offered and the DPP is 25% less efficacious than the DPP trials (NNT = 7 over 5yrs), I think that’s also highly optimistic……..

(I retain my concern about all the efficacy / effectiveness real world issues etc)

And before I get started on the very term pre diabetes – see here

So…. If I have interpreted rightly –

we will spend £170k in 1 (or £1.7m over 10y) year on 480 people (or 4,800 over 10),

of whom only 120 (or 1,200) will see any benefit (and that’s if we achieve 32% uptake and clinically beneficial impact in 25% of this 32%)….I will believe that if as and when I see it!

Cumulatively out to 5 years avoids 25 cases of diabetes, 15 out to 10 years if one years implementation.

Times by 10 – assume you intervene in the same way over those 10 years – and get about 150 cases of diabetes avoided…

The NHS will save £193k (seems NOT, I repeat NOT net of intervention costs – ie it will cost MORE) out to 10 years

Very crude fag packet maths.. and Im happy if others sharpen up.

The initial results of the NHS DPP are also worth considering in this context

NHS diabetes prevention programme helps weight loss, analysis shows

Conference presentation

Over 21 months –

  • 154 000 referred (50% men, 25% from BME group) to programme (13 education and exercise sessions of one to two hours)
  • 66 000 have taken up places.
  • 4384 completed to date
  • 2277 completed at least 8 tailored support sessions over a nine month period, losing an average of 3.3 kg or 3.7 kg, when excluding participants who already had a normal BMI
  • No data re reduced the risk of developing type 2 diabetes among participants.

Compare this to a policy intervention – in this case the sugar tax and some modelling done in Philadelphia (1.5m – three times as big as Sheff so do the maths)

out to 10 years:

intervention cost annually $238k (so lets say $100k or £60k in a pop of 500k)

2280 cases of diabetes prevented ( – or 700 cases in 500k pop

saving in health care costs of $200m – so lets say $60m or £40m (depending on exchange rate) in pop of 500k (I wonder whether this is also a little on the optimistic side, and bear in mind that USA health system is t like U.K. Cost base wise!)

$84 : $1 ROI (see above – optimistic?)

Notwithstanding over optimism, point seems well enough made – In this area, like so many others….. policy eats services for breakfast. Can prevent more cases, more effectively, effieicntly and equitably with policy sort of stuff than services delivered to one individual at a time (emptying an ocean with a teaspoon?). Simon Capewell and others have been telling us this (and quantifying it – IMPACT model) for years.

And the above is  before we get into product placement, marketing, advertising etc etc…. all the things that weren’t included in the Naitonal Obesity “strategy”……obviously we can all draw our own conslucions on why some interventions that were included in the (excellent) PHE Sugar review were not included in the eventual strategy.
I wonder if food industry advocacy suggested services delivered individual by individual would be more effective and have better return on investment.

So whilst Im not arguing against efforts to encourage us to sweat more and eat less pies….. I AM suggesting that the DPP may not be the panacea we all hope it is…..and we neglect policy interventions at our peril.



See this NIHR Signal on the Let’s Prevent RCT

Services delivered one by one to pre diabetics is NOT a good way of preventing diabetes!

Leicestershire – high BME / indian sub continent

44 practices

18,000 identified at risk

3,400 – 19% of the 18,000 – agreed for GTT

880 – 25% of the 3,400 – non DM hyperglycaemia

3,400 included in RCT

Usual care – leaflet and stern talking to from GP or nurse

Structured education (6h session *1 + 3h session *2 at 12 /24m + phone calls every 3 months.


There was no difference in terms of progression from pre DM to DM at 3yrs – hazard ratio [HR] 0.74, 95% confidence interval [CI] 0.48 to 1.14). Type 2 diabetes developed in 14.3% compared with 15.5% of the usual care group.

But this was releated to no o sessions attended. Reduction in indidence reported is greater in those that attended 2 sessions (62% reduction in incidence n = 248) and 88% in those that attended 3 n=130). Not reported whether stat sig.

No differences in BP or cholesterol or 10yr CVD risk

no difference in self-reported diet or physical activity between groups.

Services delivered one by one to pre diabetics is NOT a good way of preventing diabetes!




Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Google+ photo

You are commenting using your Google+ account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )


Connecting to %s