The Jazz Age Lawn Party, which usually occurs twice a summer on Governor’s Island, in New York, has become a delightful tradition. Typically, it features live music, social dancing, dance performances and many gorgeous 1920s-style dresses and dapper suits on two weekends in June and August. This year, it became an inadvertent example of why we need better health communication and policy on COVID-19.

Because it had been canceled last year, many folks at this year’s June event were especially eager to do the Charleston, the Peabody and otherwise swing their way into the 2020s, which are already echoing the 1920s in eerie ways.

Unfortunately, health department requirements meant that there was no dance floor allowed on the spacious lawn. And, even though proof of vaccination or recent negative COVID tests were required, masks were supposed to be worn at all times except when seated, eating or drinking. While the music was, in fact, grand, I almost expected there would be a rule that the trumpets and trombones must use mutes.

Perhaps these limitations can be seen as a reenactment of 1918 flu-related precautions. But to me, they represented yet another failure of public health communication. Despite the fact that we know far more about which situations post great risk for infection than we did last year, clear guidance from government remains scarce.

Now, however, more than ever, that is exactly what we need. Overly cautious rules, like making vaccinated people wear masks at an outdoor party, don’t make sense at a point when pandemic fatigue affects almost everyone. Instead, we need to apply the principles of “harm reduction,” which is a public health philosophy that recognizes this reality and has gained currency in recent years.

The idea of harm reduction was born in the fight against AIDS. By the mid-1980s, it was already clear that cracking down on drugs and needle possession did not prevent HIV infections. In fact, the places that had the toughest laws and the greatest scarcity of needles actually had the highest rates of AIDS.

Essentially, failing to accept that banning needles doesn’t eradicate IV drug use meant that syringe shortages guaranteed sharing—and with it, the spread of HIV. (Sadly, some states are now rerunning this experiment by cutting syringe access, and they can expect the same results seen in Edinburgh, New York and Charleston, W. Va., under such restrictions.)

And so, people who inject drugs came together with public health officials and researchers to devise an alternative. Rather than insisting that we must fight drug use at any cost, policy should focus on reducing injury to individuals and to the public health in general, they argued. Saving lives mattered more than trying to stop behavior that some see as immoral.

In other words, the best way to slow the spread of disease is not to make unrealistic demands for total abstinence, but to work to reduce the greatest harms. This idea was soon taken up by gay men who had injected drugs, and was applied to sexual behavior as well.

In the mid-1980s, one drug counselor, Luis Palacios-Jimenez, realized that he sometimes saw the same pattern of behavior among his fellow gay men that he’d seen in his addiction clients. They would try heroically to maintain abstinence or celibacy for months—and then they’d slip and make no attempt to protect themselves at all during a binge or spree. To counteract this, he co-founded the now highly regarded Eroticizing Safer Sex workshop and wrote a related book.

Psychologist Alan Marlatt observed similar behavior among cigarette smokers and drinkers, labeling it the “abstinence violation effect.” Basically, if the only acceptable goal is continuous abstinence, even a minor slip can rapidly snowball. When people believe they’ve already “blown it,” they figure they may as well just go for it and throw any prior caution to the wind. This issue affects dieters, too—and pretty much anyone trying to change behavior. With extremely high-risk activities, it can be deadly—and research in addictions shows that it can make relapses both more likely and worse.

These days, we’re seeing it with COVID measures, too: if we have to be 100 percent careful 100 percent of the time, we’ll tend abandon further attempts at risk mitigation when we fail, at least in the short term.

But harm reduction—and the many papers that now show its effectiveness in everything from reducing the spread of HIV to moderating drinking—offers a better way. Since new variants may make existing vaccines less effective, it is more crucial than ever to emphasize prevention measures primarily when they could do the greatest good so that they’ll have the biggest impact.

That means, no, don’t require masks outdoors at events when people have demonstrated their vaccination status. But do recommend them for everyone in tight, crowded indoor spaces where air circulation is limited, vaccination status is unknown, and distancing isn’t possible.

Don’t fall into the trap of “hygiene theater,” with constant washing of surfaces that show little evidence of driving transmission—but do help people maintain better handwashing habits and avoid unnecessarily touching high-contact surfaces like lobby elevator buttons.

Rather than aiming for no risk-taking ever, instead help people recognize which situations genuinely are risky—and, critically, explain why.

In the drug world, harm reduction advocates discovered that treating people as though they have agency and with respect in itself helps reduce risk-taking, likely by making people feel wanted and worthwhile.

The same applies here: talking down to people or assuming that they can’t or won’t make good decisions about risk for themselves only amplifies mistrust. In contrast, when you provide accurate information—including data related to uncertainties—people are far more likely to take heed.

Essentially, when people feel like their choices matter for not only themselves but their communities, they are more likely to do the right thing. And, when they trust that authorities are leveling with them, they are again more likely to reduce risk behavior. In fact, one study of COVID-19 risk reduction found that in countries with the highest levels of public trust, compliance with the most stringent restrictions was doubled.

As we emerge from the pandemic in the U.S., we need to learn from mistakes made early on such as communications about masking that were meant to protect the supply for healthcare workers but instead discouraged an effective preventative measure and increased mistrust.

On June 15, New York’s Governor Andrew Cuomo lifted most of the state’s remaining pandemic restrictions, citing the fact that 70 percent of the state’s adults are now at least partially vaccinated. But people remain confused, and many are afraid to let down their guard—especially with ominous reports about variants regularly making headlines. Clear, honest and accurate risk communication is critical.

Harm reduction research shows that even the most unlikely people—those who are actively addicted to drugs like methamphetamine and heroin—will act to protect themselves and others if given accurate information and access to resources like clean needles and overdose antidotes.

We should respect the general public to be able to do at least as well with COVID. (And it seems that the Lawn Party will be back in full swing in August).

This is an opinion and analysis article; the views expressed by the author or authors are not necessarily those of Scientific American.