Researchers in Sydney, Australia published a landmark study, the SEARCH-AF trial, this week in which the AliveCor Heart Monitor, a single lead ECG device built into an iPhone case, was used to screen 1,000 patients in community pharmacies for atrial fibrillation.
I call this study “landmark” not because of some ground breaking discovery or even because of the efficacy of the intervention. Rather, I consider it landmark because it is a pretty well done and carefully crafted study to evaluate the efficacy of a novel mobile health technology in the real world. Like any study, there are a variety of strengths and limitations to be considered when interpreting the results.
That being said, we have written many times about a need for thoughtful evaluation of these technologies to help guide clinicians about their use in practice and that’s precisely what we have here.
Ten pharmacies in Sydney, Australia were recruited to participate in a screening program in which the iECG (as AliveCor’s Heart Monitor was described) was used to screen patients over 65 years of age who walked in the door to pick up their medications. Pharmacists both palpated a pulse and screened with the single lead iECG tracing. Their interpretations were followed up with a cardiologist overread; a detection algorithm that became available later was retrospectively applied as well. For patients who were found to have atrial fibrillation, primary care physicians were contacted to verify whether this was a known diagnosis. If not, they were referred back for follow up and confirmation using 12-lead ECG.
A total of 996 patients were screened, and the following were found:
- 87 patients were identified to have atrial fibrillation or undetermined rhythm on the single-lead tracing
- 58 of those patients had prior known atrial fibrillation; 15 were “new” diagnoses of atrial fibrillation
- 10 of the 15 “new” patients had no prior history and were actually new diagnoses; 5 had remote atrial fibrillation presumed to be converted permanently back to sinus
- 14 of the 15 “new” atrial fibrillation patients had follow up and 9 of them ended up on anticoagulation
- For the 5 with follow up who did not end up on anticoagulation, 2 from the new group were in sinus on follow up
Cost effectiveness analysis was done as well using estimated costs of $20 per screening as well as cost estimates from the SAFE study (see below for details). They also assume a variety of levels of compliance with therapy. Assuming around a 60% compliance and a $20 per screen cost, the incremental cost effectiveness ratio per quality adjusted life year gained is ~$5,428.
The newly released atrial fibrillation management guidelines from the American College of Cardiology/American Heart Association/Heart Rhythm Society do not specifically make recommendations for population screening. They do, however, note the strong evidence for anticoagulation, citing a number needed to treat of 37 in primary prevention and 12 for secondary prevention with the use of warfarin. These numbers may be even better with use of newer generation anticoagulants. 
European guidelines recommend opportunistic screening using pulse palpation of patients over 65 years old based, in large part, on the SAFE trial by Fitzmaurice et al in which approximately 10,000 patients over the age of 65 years were randomized to either a strategy of screening by EKG or screening by palpation followed by EKG. In that study, they found equivalent detection rates of 1.6%; this was in comparison to a usual care control group of 5,000 patients in whom the incidence was 1.06%.
There are a few key distinctions between the approaches in this SAFE trial and here in SEARCH-AF. First, in SAFE, the quoted detection rates for each screening approach included patients discovered to have atrial fibrillation through usual care – an approach that better defines the incremental benefit of the screening program. We do not really know how many of the “new” atrial fibrillation patients in SEARCH-AF would have been diagnosed soon anyways in usual follow up care.
Second, there was a big difference in patient acceptance. Of the patients randomized in SAFE to a screening program, only 50-70% actually ended up being screened. [3,4] While it is not reported how many patients were actually approached for the SEARCH-AF, I suspect the uptake would be better with the AliveCor Heart Monitor than a 12-lead ECG. That the AliveCor Heart Monitor is far less onerous than a 12-lead ECG is an important feature in a community-based screening program.
Finally, unlike in the AliveCor study, SAFE did not do an ECG tracing at the time of pulse palpation routinely so we do not have a great sense of the actual sensitivity or specificity of just checking a pulse. We do know that the accuracy rate for palpation was pretty low; just about 35% for an irregular pulse actually being atrial fibrillation. And of the 75 patients in the “palpation” arm who were diagnosed with new atrial fibrillation, about a third had a regular pulse documented. [3,4] It is certainly possible, however, that some of this was related to paroxysmal atrial fibrillation. In the AliveCor study, 51 patients would have been identified using a pulse palpation strategy conferring an approximately 75% sensitivity and 40% specificity for this approach in the detection of atrial fibrillation. Of note, a detection algorithm retrospectively applied to the captured strips was reported to have a greater than 90% sensitivity and specificity.
The benefits of anticoagulation for atrial fibrillation in stroke prevention are well defined. With newer generation anticoagulants, having patients on therapy will likely become even less costly and burdensome. Strategies that help diagnose silent atrial fibrillation and get appropriate patients on therapy before a stroke can have a major impact.
Here, a community-based screening program using pharmacists was effective in identifying patients with silent atrial fibrillation at a rate greater than what SAFE would tell us is the usual care annual incidence. With the addition of a highly accurate automatic detection algorithm, this approach seems readily transferable to really any place that at-risk patients gather — senior centers, churches, clinics — with the actual screening being done by practically anyone with a steady hand.
This trial is a great example of the type of data that we need to help guide utilization of these emerging tools. This study makes a reasonable case for the utilization of the AliveCor Heart Monitor in community-based screening programs of at-risk populations, particularly for primary prevention.
In other settings, different strategies may be better. In some populations, perhaps in this one, we may eventually find that simple detection algorithms using smartphone cameras that detect heart rate using photoplethysmography are even lower cost and perform equally well as screening tools. For patients with a history of atrial fibrillation who are treated (i.e. cardioversion, pulmonary vein isolation) or in secondary prevention populations, devices such as implantable loop recorders or wearable patches like iRhyhm’s Zio may be more appropriate.
Overall, it is this type of thoughtful evaluation of mobile health technology that will help define where we can get the most bang for our buck with these tools, moving from novelty to efficacy and real-world impact.
1. January CT, Wann L, Alpert JS, et al. 2014 AHA/ACC/HRS Guideline for the Management of Patients With Atrial Fibrillation: A Report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines and the Heart Rhythm Society. J Am Coll Cardiol. 2014;():. doi:10.1016/j.jacc.2014.03.022.
2. Camm AJ, Lip GY, De Caterina R, Savelieva I et al. 2012 focused update of the ESC Guidelines for the management of atrial fibrillation: an update of the 2010 ESC Guidelines for the management of atrial fibrillation. Developed with the special contribution of the European Heart Rhythm Association.Eur Heart J. 2012 Nov;33(21):2719-47.
3. Hobbs FD, Fitzmaurice DA et al. A randomized controlled trial and cost-effectiveness study of systematic screening (targeted and total population screening) versus routine practice for the detection of atrial fibrillation in people aged 65 and over. The SAFE study. Health Technol Assess. 2005 Oct;9(40):iii-iv, ix-x, 1-74.
4. Fitzmaurice DA, Hobbs FD, Jowett S, Mant J, Murray ET, Holder R, Raftery JP, Bryan S, Davies M, Lip GY, Allan TF. Screening vs. routine practice in detection of atrial ﬁbrillation in patients aged 65 or over: cluster randomized controlled trial. Br Med J 2007;335:383.