From the article:

> "As an example, consider a hypothetical 2-year wellness program that costs $400 per month per person (say, $13 per day for healthy food) and reduces the risk of heart failure by 1 percentage point, from 5% to 4%. For the first 90 days alone, heart failure treatment is much more expensive at $22 000 than the wellness program, but that doesn’t mean that the wellness program saves money. In this case, 100 people must participate in the 2-year program to reduce the incidence of heart failure by 1, from 5 people to 4 people. In other words, the number of people who must be treated with the prevention program to avoid 1 heart failure is 100 (NNT = 100)"

1. Why is the person's food cost attributed to the wellness program, instead of the person's food budget?

2. Most people are overweight or obese. I would expect $400/person/month of healthy food to lead to weight loss as part of the intervention. This skim-read study of 5000 people[1] concludes that "Modest weight losses of 5 to <10% were associated with significant improvements in CVD risk factors at 1 year". The hypothetical uses 1 percentage point, which sounds insignificant.

3. This Novo-Nordisk page[2] says: "According to a post-hoc analysis from the Look AHEAD trial, individuals with type 2 diabetes who lost ≥10% body weight had a 21% decrease in 4-point MACE incidence (defined as CV death, non-fatal acute myocardial infarction, non-fatal stroke, or hospitalisation for angina). A weight loss of 5% has significant health benefits". Again relatively small weight loss shows significant improvements.

(Comparison: someone 300lbs losing weight at 1 lb per week takes eight months to lose ten percent of their bodyweight. Healthy weight loss is often discussed at twice that rate or 2 lbs per week. In the TV show My 600lb Life Dr Nowzaradan gives morbidly obese people targets of 5-10lbs/week initially. The hypothetical wellness program is two years?).

The improvements from healthy diet which leads to weight loss ought to be more than just reduced risk of heart attack. Reducing hypertension (high blood pressure) leads to reduced risk of stroke and chronic kidney disease. Reduced body fat and healthier diet leads to reduced risk of Type 2 Diabetes and its complications (vision loss, neuropathy, slowed wound healing, worse immune function). Reduced weight leads to reduced risk of back pain, knee pain, sleep apnoeia, non-alcoholic fatty liver disease. Reduced body fat leads to reduced risk of some cancers, better diet reduces risk of other cancers. The hypothetical only considers treatment for heart attack in the savings but there should be a lot more savings.

The hypothetical only considers the treatment cost of a heart attack for 90 days but is that all the treatment they need? Are they not on regular drugs and checkups long term? And someone who survives a heart attack is at increased risk of another heart attack from all the same reasons they had the first one. American Heart Association says "about one in five people who have had a heart attack will be readmitted to the hospital for a second one within five years"[3].

And avoiding the prodcutivity loss from all the time off work from heart attack, back pain, sickness, medical appointments and treatments over years and decades. Intervention may not save money, but it may lead to increased ability for the economy to generate more money than it cost.

There's probably a point in that article, but why use such an easily criticised hypothetical to try and make a point? Not even going into $13/day vs 'beans and rice' or saving a person money on takeaway, cigarettes, alcohol, gasoline, or the costs to a person of moving to a walkable area or joining a gym, or all that kind of lifestyle intervention.

[1] https://pmc.ncbi.nlm.nih.gov/articles/PMC3120182/

[2] https://pro.novonordisk.com/disease-area/cardiovascular-dise...

[3] https://www.heart.org/en/news/2019/04/04/proactive-steps-can...

In the limit, one would expect that to be the case: at some point the marginal cost curves will intersect. I’m quite surprised that we are already at that point. I suppose more things are at the efficiency frontier than I had imagined.

Thanks for the link. Will have to read it later.

I would read the comments on that article. The conclusion you've come to seems murky at best.