Hot Off the Press: Mobile Smartphone Technology Is Associated With Out‐of‐hospital Cardiac Arrest Survival Improvement

Early cardiopulmonary resuscitation (CPR) and use of an automated external defibrillator (AED) are key components of the chain of survival in out-of-hospital cardiac arrest (OHCA).1 Unfortunately, most OHCA victims still do not receive bystander CPR, and most individuals trained in basic life support (BLS) do not get opportunities to employ their skills.2 The use of smartphone apps to direct individuals trained in BLS to victims of OHCA, as well as to map the location of AEDs, has been suggested as a method to improve OHCA survival rates.3-5 One previous study demonstrated an association between the use of an app and the rate of bystander CPR, but there was no statistically significant change in the rate of return of spontaneous circulation (ROSC) or survival.4 A scientific statement from the American Heart Association concludes that “although digital tools have tremendous potential, there is a paucity of scientific evidence for their effectiveness in improving ECCC [emergency cardiovascular and cerebrovascular care] to date. Moreover, there is potential for unintended consequences, such as incorrect information being provided via mobile apps or social media. Therefore, a key conclusion of this statement is that there is a clear need for rigorous research on digital strategies for ECCC, to build the scientific evidence base for their effectiveness and safety.”6 Derkenne et al.7 performed an observational trial looking at the use of the Staying Alive smartphone app through the Paris Fire Brigade. This is an observational cohort study that looks at the using of the “Staying Alive” smartphone app through a single emergency medicine system (EMS) in the Greater Paris area. Of 4,107 OHCAs in 2018, the app was activated 366 times, of which 46 patients received treatment from a responder. Treatment consisted of CPR in 24 cases, AED use in 18 cases, and both in four cases. The rate of ROSC was higher in patients who received treatment from an app user when compared to those who did not receive treatment (48% vs. 23%, p < 0.001). Similarly, the group that received treatment had higher survival to hospital discharge (35% vs. 16%, p = 0.004). There are always limits to the conclusions that we can draw from observational data, because the existence of confounders may mean that factors other than the variable of interest are responsible for the differences seen. Choosing an appropriate control group is of fundamental importance when using observational data. In this study, the smartphone application was available and activated for every cardiac arrest, and they compared a group of patients who received treatment to a group who did not receive treatment. Because the smartphone application was available for all patients, the comparison might be useful in telling us that CPR is a valuable intervention, but does not seem to provide any information about whether the smartphone app actually helped patients. We wonder whether a better control group might have been a similar group of patients for whom a smartphone app guided response was not available, such as patients from a different geographic location or perhaps historical controls in the same location before the smartphone app was implemented. Selection bias occurs when individuals chosen for enrollment in a trial are dissimilar to those not included in the research. When the selection occurs at the level of the trial, it tends to limit the generalizability of the results. When selection bias occurs at the level of groups within a trial, it becomes a potential confounder, potentially resulting in inappropriate conclusions. In this trial, the Staying Alive app was only activated in 366 (9.8%) of 4,107 cases of OHCA, so the trial results may not apply to 90% of OHCA patients, limiting generalizability. The group of patients in whom the app was activated were significantly younger, with higher rates of witnessed arrests and bystander CPR. This selection bias is represented in the incredibly high survival to hospital discharge seen in both the intervention and the control groups of this study (35 and 16%, respectively). Furthermore, the treatment group consisted only of the 46 patients who received lifesaving maneuvers, rather than all 366 patients in whom the app was used, which is a very important confounder in this data. Of 4,107 OHCAs during the study period, the Staying Alive app was activated 366 times (9.8%). Forty-six patients received treatment from a first responder through the app and make up the intervention group. Of these, 24 received CPR only, 18 received an AED only, and four received both. There were 226 patients for whom the Staying Alive app was activated but no treatment was given, because no one responded to the app alert, the responder could not locate the patient, or the responder arrived on scene but did not provide BLS treatment. When comparing patients who received treatment to those who did not, the rate of ROSC was higher (48% vs. 23%, p < 0.001) and the rate of survival to hospital discharge was higher (35% vs. 16%, p = 0.004). Although the use of smartphone technology to improve the number of OHCA victims who receive CPR and have access to an AED makes sense, there is still no evidence that these apps improve clinical outcomes. They seem intuitive, and implementation without evidence may seem reasonable, but it is still important to recognize costs and potentially harmful unintended consequences of any intervention. Although widespread adoption of such apps may not require a high level of evidence, we still think that clinical evidence is important. Ken Milne MD (@TheSGEM) replies: There are apparently multiple apps. Jesse Luke (@RealJesseLuke) replies: I had no idea. Actually sounds like a great idea. Justin Morgenstern (@First10EM) replies: Excellent! I assume all observational data to date? Tommaso Scquizzato (@tscquizzato) replies: 1 RCT (Ringh 2015 NEJM), 1 before-and-after study (Lee 2019 Resuscitation) and the others all observational studies. Guillaume Debaty (@gdebaty): We are currently recruiting in a multi center stepped-wedge randomized trial in France covering both rural and urban areas: NCT03633370. 1800 patients already included (2100 total). hope we will get an answer. @dispatchsamu @SAUVLife. Justin Morgenstern (@First10EM) replies: Normally we expect evidence for interventions before they are incorporated into guidelines. Should smartphone apps be treated differently than other interventions? Why? Paul Snobelen (@PSnobelen) replies: In this case, I think so. Municipal/ Regional Privacy, Liability, and Litigation teams feel more comfortable in supporting & allowing implementation of or to study these programs when it is being recommend from a body like the AHA. Low risk with potential positive outcomes. Another consideration. Do I want to measure an “outcome” like a ROSC, or action? A bystander can't control outcomes, only their actions. So can I link those who are willing to act, to those in need? A step further, can I link the arrival of an AED by human or drone? It's exciting. Smartphone apps may be a valuable tool in directing emergency medical services to patients in needed, but we lack high-quality evidence that they improve clinical outcomes.

Fetched live from OpenAlex and de-inverted. Abstracts are not stored in this database: the inverted indexes are 8.6 GB of the frame’s 9.3 GB of text, and the host has 13 GB free.