Just twenty years ago only 3 of 100 adolescents with diabetes had Type 2 diabetes (T2DM); today that proportion nears 1 in 2. Moreover, this epidemic of T2DM disproportionately affects disadvantaged children, many of whom grow up in environments that lack healthy food options or safe exercise facilities. Though it is clearly urgent to devise medical, behavioral, and cultural interventions to decrease incident cases of T2DM, it is equally pressing to prevent or delay the early onset of the known chronic complications of T2DM. Indeed, the efficacy of various approaches, alone or in combination, remains unknown.
In this week’s NEJM, the Treatment Options for Type 2 Diabetes in Adolescents and Youth (TODAY) investigators present their findings from a combined medical and behavioral approach. The researchers randomized 699 youth with Type 2 Diabetes to either metformin, metformin plus rosiglitazone, or metformin plus lifestyle intervention. The lifestyle intervention consisted of in-person visits to the child’s home and used family-based interventions to encourage weight loss through activity and healthy eating. The youth were between the ages of 10 and 17, had a BMI at or above the 85th percentile, and had carried a diagnosis of T2DM for less than two years. The participants were followed for at least two years. The primary outcome was time to treatment failure, as defined by a hemoglobin AIC >8% for six months, or metabolic decompensation requiring insulin that could not be weaned after three months. Adherence, based on assessment of pills that remained in study blister packs, was monitored every two to three months.
For the group as a whole, the results are sobering — 45.6% failed treatment with an average time to failure of 11.5 months. Those in the rosiglitazone group reached treatment failure later than those in the metformin alone group, with a 25.3% decrease in the occurrence of the primary outcome. Though lifestyle intervention plus metformin seemed to temper disease progression more than metformin alone, the difference was not statistically significant. Meaningful weight loss occurred among only 31% of those in the lifestyle intervention arm and was apparently not sustained.
Subgroup analyses were performed to assess interactions between gender and ethnicity and treatment assignment. Overall, more boys than girls failed treatment (48.2 versus 44.3%), and the benefit seen with dual pharmacotherapy was more pronounced in girls than in boys. Non-Hispanic blacks failed treatment more than either Hispanics or non-Hispanic whites, but there was no interaction between ethnicity and treatment effect, except that metformin alone seemed less effective in blacks. Adherence went from 84% to 57% at month 60, but did not differ significantly across treatment groups, and was not thought to account for the treatment failure rate. Serious adverse events, including hospitalizations and hypoglycemia, occurred in just over 19% of all study participants; however, 87% of these events were not considered related to study treatment
What to make of these data? For now, our solution to this problem clearly does not lie in medications alone, nor in this particular behavioral intervention. Foremost, the alarming rate of treatment failure these data suggest demands the attention of our entire society. Nevertheless, the banning of 16-ounce sugar-based soft drinks two weeks ago in New York City stirred up tremendous controversy, as many viewed this measure as yet another example of government as “nanny state,” limiting people from choosing freely. Perhaps. But the question these data pose is not whether we need social interventions, but what types of interventions, combined with our arsenal of drugs and lifestyle changes, actually work. Admittedly, we have no idea if measures such as a soda ban actually work; one can easily refill a 12-ounce cup, or simply buy a bigger soda at a corner “convenience” store.
Thus, if the results of the TODAY study are used to fuel our social willingness to implement, and rigorously analyze, social, as well as behavioral and medical interventions for our nation’s diabetic youth, then our glass is half-full. But if we fail to heed the caution these results dictate, continuing to depend upon the status quo to save us, then we should consider our glass, and a big one at that, half-empty.