When patients start on the latest obesity drugs, they find that their food cravings drop away, and then the pounds do too. But when patients go off the drugs, the gears shift into reverse: The food cravings creep back, and then the pounds do too. Within a year of stopping semaglutide—better known by its brand names Wegovy or Ozempic—people regain, on average, two-thirds of the weight they lost. Tirzepatide, also known as Zepbound or Mounjaro, follows a similar pattern. And so the conventional medical wisdom now holds that these obesity drugs are meant to be taken indefinitely, possibly for a lifetime.

To pharmaceutical companies selling the blockbuster drugs—known collectively as GLP-1 drugs, after the natural hormone they mimic—that might be a pretty good proposition. To patients paying more than $1,000 a month out of pocket, not so much. Most Americans simply cannot afford the cost month after month after month.

This has forced some doctors to get creative, devising regimens to sub in cheaper, if less well-known, alternatives. GLP-1 drugs do work remarkably well, inducing more weight loss more quickly than any other obesity medication on the market, but some doctors now wonder whether patients need to be on GLP-1 drugs, specifically, forever. “​​What if we do a short-term investment, use it for six months to a year to get 50 pounds off?” asks Sarah Ro, an obesity-medicine doctor and the director of the University of North Carolina Physicians Network Weight Management Program. Then, as she and other doctors are now exploring, patients might transition to older, less expensive alternatives for long-term weight maintenance.

In fact, Ro has already helped patients—she estimates hundreds—make the switch out of financial necessity. Few of her patients in rural North Carolina have insurance that covers the new obesity drugs, and few can afford to continually pay out of pocket. In April, many also lost coverage when North Carolina’s health insurance for state employees abruptly cut off GLP-1 drugs for obesity. Ro switched her patients to older drugs such as topiramate, phentermine, metformin, and bupropion/naltrexone, plus lifestyle counseling. It’s not exactly an ideal solution, as these medications are generally considered less effective—they lead to about half as much weight loss as GLP-1 drugs do—but it is a far less expensive one. When prescribed as generics, Ro told me, a month’s supply of one of these drugs might cost as little as $10.

Jamy Ard, an obesity-medicine doctor at Wake Forest University, has also switched regimens for patients who lost coverage of GLP-1 drugs after retiring and getting on Medicare, which currently does not pay for any drugs to treat obesity. (Like many researchers in the field, Ard has received grants and consulting fees from companies behind obesity drugs.) Doctors I spoke with didn’t know of any studies about switching from GLP-1 drugs to older ones, but Ard says this research is a practical necessity in the United States. With GLP-1 medications exploding in popularity, more and more patients taking them will suddenly lose coverage when they hit retirement age and go on Medicare. “Now I’ve got to figure out, well, how do I treat them?” he told me.

Long-term data on the older drugs themselves are, in fact, pretty sparse, despite the drugs having been available for years and years. Until Ozempic came along, obesity drugs were not a lucrative market, so companies weren’t interested in funding the long and very expensive trials that follow patients for several years. “Studies like that cost a fortune,” Louis Aronne, an obesity-medicine doctor at Weill Cornell Medicine, told me. Some of the longest-term follow-up data about these drugs come from patients at his practice in Manhattan—not a representative population, he admits—which he published in a five-year study funded by the National Institutes of Health. (Aronne has also received grants and consulting fees from the makers of obesity drugs.)

How patients do after switching from GLP-1 to older drugs is entirely anecdotal, but so far outcomes do seem to vary quite a bit. A small minority of patients who stop GLP-1 injections are actually able to maintain their weight on diet and exercise, without any additional medications. Others may find that the older pills are simply not effective for them. In Ro’s experience, about 50 to 60 percent of her patients have so far successfully kept the weight off using one or more older drugs, on top of lifestyle changes such as cutting out fast food and sugary sodas.

The best drug to switch to may also depend on the patient. Each of the older medications works differently, hitting different biological pathways. The combination of naltrexone and bupropion, for example, makes food less pleasurable and seems to work especially well in people with a tendency toward emotional eating, Ard said. Topiramate, meanwhile, makes carbonated drinks unpleasant, which could help patients who drink a lot of soda. The older drugs also have different side effects. Aronne rattled off for me a list of health risks that might rule out a particular drug for a particular patient: seizures for bupropion, or glaucoma for topiramate. Finding the most effective and best-tolerated drug for a patient may take some trial and error.

Doctors are now discovering that some patients can maintain the weight they lost on lower or less frequent doses of GLP-1 drugs. “For the first time in my career, we’re lowering the dose of medicines,” Aronne said. Just reducing the dose doesn’t save money, though, as lower-dose injection pens cost the same as those with higher doses. However, by instead extending the time between doses from the standard seven days to a longer 10-day interval, doctors told me, some patients have been able to stretch their supplies.

But tapering off obesity medications entirely, GLP-1 or otherwise, will probably not be possible for most patients. Weight loss tends to trigger a powerful set of compensatory mechanisms in the body, which evolved long ago to protect us from starvation. The more weight we lose, the more the body fights back. The fight never quite goes away, and most patients will likely require some kind of continued intervention just to stay at a lower weight. Long-term weight maintenance has always been the “holy grail” of obesity treatment, Susan Yanovski, a co-director of the ​​Office of Obesity Research at the National Institutes of Health told me. The best maintenance strategy—whether it involves GLP-1 drugs, and at what dose—may ultimately be pretty individual. What works best and for whom still needs to be studied. “These are really good research questions,” Yanovski said. But they are not necessarily the questions that pharmaceutical companies focused on developing new meds are most keen to answer.

In time, the current crop of GLP-1 drugs will eventually become available as generics, too, and cost may no longer drive patients to seek out cheaper alternatives. But for now, it very much does.