Although we agree that society should not necessarily wait for perfect knowledge to support nationwide actions that promote public health, on the subject of diabetes prevention we do not agree with the arguments put forth by Goff et al. (1).

First, we did not call for a community-based randomized controlled trial (RCT) that would replicate the well-known prevention trials, only that there should be more evidence than is currently available before national health care resources are diverted into efforts to prevent diabetes (2). In brief, RCTs have told us only that diabetes can be delayed on average 2–4 years, but we have no evidence that such a delay has a favorable impact on clinical outcomes for a population. Moreover, we do not know how much weight loss in those overweight or obese people, maintained for some period, is sufficient to have a clinical benefit.

Second, Goff et al. (1) cite a number of public health initiatives that were implemented without evidence derived from an RCT. For two that they cite, the “cost” to society is hardly appreciable (speed limits and smoking), and for smoking the burden has been placed almost entirely on the offender and not society. For those policies, and the others, there was strong empirical evidence that the intervention would very likely have known medical benefits. In contrast, there is no evidence that modest weight loss in those with prediabetes will have any medical benefit. Also for the interventions Goff et al. cite, it is clear what exactly is to be done (e.g., wear a seat belt, get an air bag, stop putting chemical X in Y amount in our water). Conversely, for diabetes prevention there is no clear knowledge of the components of a community lifestyle intervention program that will, together, be beneficial for a large population. What exact resources are critical and whether they will result in the maintenance of weight loss are vague. Moreover, it is even more unclear whether a program would be equally effective in people of different ethnicities, cultures, and socioeconomic status.

Third, we do not believe an intervention has to work for everyone. It may well be cost-effective for a limited number of individuals even though everyone pays for the intervention. Whereas the community-based studies performed by Goff and his colleagues are encouraging, formal cost-effectiveness studies indicate that much more needs to be accomplished. For example, prevention services were shown to be cost-effective when substantial weight loss (∼4% of body weight) was maintained for at least 30 years and the intervention was no more than ∼$200/year/person. Furthermore, a prevention program only became cost-effective after ∼15 years, which assumes that if we start a program now, only after 15 or more years it might pay off.

Finally, we think Goff et al. (1) are correct to assert that lifestyle interventions are unlikely to impose a significant health risk. But absent knowing more what “healthy eating and active living” (1) specifically entail, what they would cost, and what we expect will happen to society at large given the money spent, it seems like government or health plans are being asked to act on hope and not good science. A lifestyle of “healthy eating and active living” would seem to be appropriate for individuals who may benefit from an individualized plan under the guidance of their health care provider.

Duality of Interest. No potential conflicts of interest relevant to this article were reported.

1.
Goff
DC
,
Katula
JA
,
Blackwell
CS
,
Isom
SP
,
Pedley
CF
,
Vitolins
MZ
.
Comment on Kahn and Davidson. The reality of type 2 diabetes prevention
.
Diabetes Care
2014
;
37
:
943
949
(Letter). Diabetes Care 2014;37:e185–e186. DOI: 10.2337/dc14-0854
2.
Kahn
R
,
Davidson
MB
.
The reality of type 2 diabetes prevention
.
Diabetes Care
2014
;
37
:
943
949