Lifestyle Plus the Drug

The tiresome argument about whether people should “just eat better instead” has an answer now, from 98,261 people followed across 632,543 person-years. Doing both beat doing either. People taking a GLP-1 and keeping six to eight healthy habits had 43% fewer major cardiovascular events than people with three or fewer habits and no drug.^[1]

What was measured

The authors’ conclusion is that combining lifestyle change with pharmacotherapy produced a greater reduction than either approach alone.^[1]

The comparison you cannot make from this table

It is tempting to set 0.40 against 0.84 and conclude lifestyle is roughly four times more powerful than the drug. That comparison is not available, and it is the misreading this study will be used for.

The rows are different contrasts. One compares the extremes of an eight-point gradient; the other compares taking a drug against not taking it.

“Eight habits versus one or fewer” contrasts the healthiest people in the dataset against the least healthy across a wide range of behaviors. “GLP-1 user versus non-user” is a single yes-or-no. Those produce numbers that cannot be ranked against each other, any more than you could compare the effect of “the top versus bottom of a scale” with “on versus off a switch”.

There is a second problem specific to lifestyle findings in observational data. Adhering to eight healthy habits is not a randomly distributed trait — it correlates with income, housing, working hours, disability, and every other advantage that independently predicts cardiovascular outcomes. Adjustment helps and does not remove it.

What the finding does support

• The two are not alternatives. This is the useful part, and it is the opposite of how the choice is usually framed publicly.
• Taking the drug does not make lifestyle irrelevant. The combined group did better than drug users generally.
• The drug is not a substitute for the habits, and the habits are not a reason to refuse the drug. Both directions of that sentence are supported here.
• It is consistent with the randomized picture. In a four-arm trial we covered in exercise plus a GLP-1, the drug alone did nothing for physical fitness while exercise did, and the combination did most.