Of all the technological innovations of recent decades, the red-light camera has to rank among the least popular. A new study suggests that, in terms of their intended effect, they are also among the least successful.

These devices, which snap photos of vehicles making illegal maneuvers at intersections, have long been the subject of heated debate. Some public officials consider them nothing but “revenue generating” devices for municipalities, which receive money from the increased number of traffic citations. Others insist they help cut down the number of drivers who recklessly run red lights.

A just-released paper by researchers at the University of South Florida College of Public Health comes down dramatically on the side of the doubters. Lead author Barbara Langland-Orban, a professor and chair of health policy and management, analyzed studies from a number of states and concluded that “red-light cameras don’t work. Instead, they increase crashes and injuries as driver attempt to abruptly stop at camera intersections.”

“Police in one community concluded that nearly 90 percent of infractions at a trial camera were split-second violations visible only to the camera lens,” the report states. “The majority of the red light running safety issue can be resolved through inexpensive engineering remedies that address infraction sin the first second after the light changes (such as) lengthening yellow light timings and/or adding a brief all-red-light interval, which permits traffic to clear the intersection prior to releasing cross traffic.”

Other, similarly simple engineering improvements, such as improving signal visibility and timing series of signals on a street, are far more likely to reduce red-light running, the report adds.

Langland-Orban adds that insurance companies love these devices, for obvious reasons: More citations mean higher premiums. But in terms of highway safety, they are definitely not the ticket.