Smartphone CPR alerts matched standard 911 response in public cardiac arrest cases
A smartphone system designed to summon nearby volunteers did not improve the chance that a person in cardiac arrest received bystander CPR or an automated external defibrillator before professional responders arrived. In a randomized analysis drawn from two population areas, the app-based alert strategy produced nearly the same early care as standard emergency dispatch alone.
The finding is striking because the intervention targets the narrowest, most time-sensitive window in resuscitation: the minutes between collapse and first chest compressions. During that interval, perfusion falls fast, and every additional layer of delay erodes the odds of survival. Yet in these public-location cases, the added digital alert did not move the needle.
Among 355 people included in the primary analysis, bystander CPR or AED use occurred in 123 of 178 cases, or 69%, when PulsePoint Respond alerts were activated, compared with 127 of 177 cases, or 72%, when alerts were not activated. The site-adjusted difference was not statistically significant. Survival to hospital discharge was about 21% in each group.
Why a faster notification system did not change the bedside outcome
PulsePoint Respond works by linking local 911 dispatch centers with app users who have indicated they are willing to help during an emergency. When a suspected arrest occurs in a public place, the app can send a “CPR Needed” notification to nearby volunteers and display the location of the emergency as well as nearby AEDs. The logic is straightforward: if trained help can arrive before EMS, the person in arrest might receive compressions and defibrillation sooner.
The trial suggests the system may have been operating in an environment where bystander response was already unusually strong. In the control group, 73% of patients received bystander CPR even without app alerts. That high baseline leaves little room for an additional dispatch tool to show a measurable gain, especially in urban settings where lay rescuer activity is already common and professional response times are relatively short.
Median time from the 911 call to EMS arrival was 5.6 minutes in the alert group and 5.9 minutes in the no-alert group. Those intervals are short, and they compress the margin in which app alerts can influence care. If trained volunteers are only a few minutes ahead of EMS, the intervention has to overcome not just response time but also the availability of a nearby responder who is willing, able, and physically close enough to intervene.
A second constraint emerged from the responder geography. In 57% of activated incidents, only one app user was within 400 meters of the event, and 91% of incidents had four or fewer volunteer first responders nearby. Sparse responder density limits the probability that the alert will translate into immediate action. A notification is only as useful as the person who receives it.
Urban response conditions may have narrowed the effect size
Most cardiac arrests in the analysis occurred in urban locations, accounting for 91% of cases. Urban response systems can be efficient, but they can also be crowded with competing signals: multiple nearby witnesses, street-level traffic, buildings that complicate access, and dispatch intervals already optimized for speed. In that setting, the incremental gain from a volunteer notification may be modest unless the network is dense and highly responsive.
The study’s structure also matters. Researchers randomly assigned 911 calls for possible cardiac arrest to either standard EMS dispatch or standard dispatch plus PulsePoint alerts, then later identified which events met eligibility criteria. The design strengthens internal comparison, but it also means the final analysis focused on public-location arrests with at least one PulsePoint user within 400 meters. The result says something specific about that deployment model, not about every possible use of smartphone-based volunteer alerts.
The mechanism behind early resuscitation remains biologically compelling. Chest compressions preserve coronary and cerebral perfusion; AED shocks can terminate a shockable rhythm before irreversible injury accumulates. A system that shortens the interval to either intervention should, in theory, help. Here, the theory ran into a real-world ceiling created by already-high bystander performance and limited volunteer density.
The limits of the study are clinically important
The trial does not address whether PulsePoint-style alerts might work better in private homes, where most out-of-hospital cardiac arrests occur, or in settings with longer emergency response times such as rural communities and high-rise buildings. It also leaves open whether a larger pool of nearby volunteers could change the result. Those are not minor exclusions; they define the settings where early intervention might matter most.
The investigators reviewed 17,547 randomly assigned 911 calls for possible cardiac arrest in public locations from June 8, 2021, to March 27, 2026, in British Columbia and Columbus, Ohio. Only a small fraction met the criteria for the primary analysis, which makes the study highly specific to the circumstances in which the alert system was actually tested. The narrow eligible cohort also means the findings should not be stretched beyond the deployment conditions that were examined.
One sentence captures the central result: more notifications did not produce more early resuscitation in the public spaces studied. That is not a failure of CPR training or volunteer willingness; it is a signal that implementation environment can overpower elegant technology.
Steven Brooks, MD, MHSc, who served as senior lead author and co-principal investigator, said the unexpectedly high bystander CPR rate in the control group may have limited the opportunity for alerted responders to add benefit. That explanation fits the data. If ordinary 911 response plus community readiness already delivers CPR in most cases, the value of an overlay app will be hard to prove unless it reaches a population with slower access to help.
What resuscitation teams should take from the data
For emergency systems and policymakers, the most practical lesson is not that smartphone alerts have no role. It is that their effect depends on where they are used, how many volunteers are nearby, and how quickly standard EMS already arrives. In a dense urban network with strong baseline lay response, the incremental benefit may be small. In other settings, the answer could be different, but this study does not provide that evidence.
The abstract was presented at the American Heart Association’s Resuscitation Science Symposium 2026 and remains preliminary until publication in a peer-reviewed journal. For now, the trial leaves resuscitation leaders with a concrete question for future work: whether volunteer alert systems can outperform standard dispatch when the geography, response times, and responder density are less favorable than those seen here.