For years, safety researchers believed that drivers were most likely to reach for their phones in slow-moving traffic — stop-and-go conditions that left them bored, impatient and reaching for distraction. A sweeping new study has turned that assumption on its head.
Drivers are actually more likely to handle their phones the faster they go, according to research published Monday by the Insurance Institute for Highway Safety. The finding, drawn from nearly 600,000 trips taken by drivers across the United States, suggests that two of the most dangerous behaviors on American roads are not random and independent — they are linked, and they tend to peak at the same moment.
“Until now, safety experts believed drivers used their cellphones most at slower speeds,” said David Harkey, the institute’s president. “But data from insurance companies’ safe-driving apps show that, in free-flowing traffic, the opposite is true.”
The Numbers Behind the Pattern
The study, conducted by IIHS senior research scientist Ian Reagan and senior statistician Sam Monfort, analyzed trip data collected between July and October 2024 by Cambridge Mobile Telematics, a platform provider that powers safe-driving apps used by auto insurers. The researchers focused exclusively on time spent driving in free-flowing conditions, setting aside periods stuck at intersections, idling in congestion or traveling below the posted speed limit — isolating, in other words, the moments when a driver is truly in motion.
What they found was a consistent and troubling gradient. On limited-access roads — freeways and highways where vehicles enter and exit only via on- and off-ramps — the share of driving time spent handling a phone rose by 12 percent for every 5 miles per hour a driver exceeded the posted limit. On arterials and other roads with traffic lights and intersections, the increase was smaller but still present: 3 percent for every 5 mph over the limit.
The pattern grew more pronounced on roads with higher speed limits. On a freeway posted at 70 mph, exceeding the limit by 5 mph was associated with a 9 percent larger increase in phone handling than the same amount of speeding on a road posted at 55 mph. A similar dynamic appeared on surface roads: the faster the posted limit, the steeper the relationship between speeding and distraction.
“It’s alarming that the relationship between cellphone manipulation and speeding was the strongest on roads with the highest speed limits,” Reagan said.
Why It Happens
Researchers offered several explanations for the pattern, none of them mutually exclusive.
One possibility is self-selection: drivers who take more risks tend to do so across the board, speeding and reaching for their phones as expressions of the same underlying disposition. A second theory centers on stress. Earlier research has shown that phone use spikes during rush hour and school drop-off periods — the same situations that tend to produce speeding. A third explanation involves the road itself: higher-speed roads often have lighter traffic, fewer pedestrians and longer stretches without stoplights, conditions that may give drivers a false sense that the environment is forgiving enough to permit a glance at a screen.
Previous research had generally found the opposite pattern — that drivers were more likely to engage in distracting activities at slower speeds, in simpler settings, adjusting their behavior based on their perception of risk. But those studies, Reagan noted, typically relied on small groups of volunteer drivers, covered only a few road types and lumped together time spent actively driving with time spent stopped at red lights.
The availability of telematics data from safe-driving insurance apps has changed what is measurable. These apps use a smartphone’s GPS and internal sensors to track speed, location, hard-braking events and phone use, building a detailed record of each trip. Drivers in the programs can receive lower premiums in exchange for allowing the data to be collected. The IIHS study used aggregated, anonymized trip data from drivers across the continental United States, excluding Alaska, California, Hawaii and New York. Only trips lasting at least 18 minutes with at least two minutes on an interstate were included.
Phone handling was detected when the phone’s gyroscope registered a significant rotation while the screen was unlocked — a threshold designed to capture intentional interaction rather than incidental movement.
Implications for Enforcement
The research carries practical implications for traffic safety policy, particularly for how law enforcement agencies allocate their resources.
Most distracted-driving and anti-speeding campaigns are run separately, each targeting one behavior at a time. The new findings suggest that pairing the two efforts — particularly on high-speed roads — could yield a larger return. Drivers who are speeding on freeways, the data indicate, are the most likely to also be on their phones, making them the highest-priority target for combined intervention.
The challenge is operational. The most common methods of catching distracted drivers — officers posted at intersections, unmarked vehicles with elevated vantage points — are difficult to deploy on highways. The researchers argued that the findings strengthen the case for automated safety cameras capable of detecting both speeding and phone use simultaneously.
They also noted a limit in the data itself: the drivers in the study were all participants in safe-driving programs, motivated by the prospect of lower insurance premiums. Even among that self-selected, incentivized group, the behavior persisted.
“Speeding and distracted driving together are especially dangerous,” Harkey said. “This research shows the risk is greater than we once thought, but it also points to an opportunity to address both problems at the same time.”