AI improves AF and rhythm irregularities detection and diagnosis rates in Community

Atrial fibrillation (AF) is a serious public health problem significantly effecting the health care costs. It is associated with severe complications such as stroke, systemic embolism, heart failure, and cognitive impairment, leading to high morbidity and mortality; thus, its early diagnosis and proper treatment are of great importance. AF associated Stroke indicates greater severity, a higher mortality rate, and greater disability, leading to high socioeconomic impact due to the costs derived from hospital admissions and home care required by the patients. Therefore, the management of patients with AF should include treatment not only of the fibrillation itself, but for the prevention of stroke and other thromboembolic events as well.

High-quality epidemiological studies at the local, national, and regional levels are needed to estimate the regional and global burden of AF. In low- and middle-income countries systematic population based investigation of AF is limited. Where surveys have been conducted, there is marked heterogeneity in diagnostic methods, clinical settings, and the presence or absence of symptoms. The 2 approaches to increase the rate of detection to enable early intervention, especially for the prevention of stroke and other sequel of AF, are 1) systematic and 2) opportunistic screening for AF. In a review of evidence from the United Kingdom, Moran et al recently showed that although both approaches compared favorably on AF diagnosis rate, from the health service provider perspective, the cost of systematic screening’s was high than that of opportunistic screening. There are specific challenges involved in AF detection because the condition can often be asymptomatic as well as sporadic (or paroxysmal). This will require the development of innovative approaches like mobile and wireless technologies.

Health care systems in the developing world are challenged both by significant public health issues related to communicable disease and, increasingly, by a rising prevalence of noncommunicable lifestyle diseases, including obesity, tobacco addiction, hypertension, diabetes, and cardiovascular disease.

The limited studies available suggest that in the developing countries there is a high prevalence of AF, associated with hypertension and valvular heart disease, and carries a risk of stroke. Highly variable use of anticoagulants may be related to different health care and socioeconomic settings. More studies are needed for better understanding of the epidemiology and management of AF in developing countries.

Valvular AF is much more common in Africa than in Europe or the US due to the high burden of rheumatic heart disease. While undertreatment of AF remains a worldwide problem, patients with AF in Africa are even more likely to be undertreated than those in other regions due to major differences in the availability of health care resources and socioeconomic conditions. In most of Africa, fewer patients with AF receive anticoagulant therapy compared with those in other parts of the world.

Rates of stroke-related disability and mortality are ten fold higher in medically underserved regions of the world compared with the most developed nations. This is primarily due to the lack of access to primary care to screen for and mitigate stroke risk factors. In the developed world, AF is the leading preventable cause of stroke, accounting for 15%–20% of all strokes. Therefore, effective strategies to reduce the risk of stroke in patients with AF must be considered in the context of local health care systems.

AF burden has regional variations, with high-income countries experiencing a higher prevalence, incidence and mortality than low-middle income countries (LMIC), as the lower rates of AF documented in developing countries may be due to under reporting, limited access to health care services and geographical disparity in published data.

Chugh, et al. reviewed all population based AF studies between 1980 to 2010, from 21 global burden of disease regions. These investigators estimated global/regional prevalence, incidence, and morbidity and mortality related to AF. The estimated number of individuals with AF globally in 2010, was 33.5 million (20.9 million men and 12.6 million women) with significant regional variations and heterogeneity. Mortality associated with AF was increased by 2-fold in both genders from 1990 to 2010.

Identifying individuals at risk of developing AF is important as early detection and treatment of modifiable risk factors can reduce morbidity and mortality due to AF. Current guidelines advocate that “all patients who present with symptoms of AF―breathlessness, palpitations, syncope, chest discomfort or stroke―should have their pulse checked for irregularities as well as 12-lead ECG”. Prolonged ECG monitoring may be especially useful in patients with heart failure and post-stroke, in order to enhance detection and reduce health resource utilization and costs, depending on local resource and expertise. A recent randomized trial comparing routine practice versus targeted population-based screening and opportunistic screening, opportunistic palpation (pulse-taking) of patients aged 65 and over, with or without known AF risk factors was found to be the cheapest and most effective method of screening for AF . One limitation of opportunistic pulse palpation is the high number of false positives that can result in unnecessary ECGs. A recent meta-analysis has suggested that newer technologies such as modified blood pressure monitors (BPMs) and single-lead ECGs may be more accurate in detecting AF, and at-home BPMs have been estimated to reduce strokes and save costs by the UK National Institute of Clinical Evaluation.

Treatment and management gaps exist worldwide and older populations are the ones at highest risk but remain largely untreated due to the perceived risk of bleeding; these gaps vary in degree across countries, but are more prominent in LMIC. Data from LMIC are limited and suggest very low rates of oral anticoagulation therapy among AF patients. A few studies reported that estimated rates of anticoagulant use range from only 2.7%–50% in China 26-29 26%–44% in Pakistan 30, 16% in Malaysia 31, 46.7%–57.8% in Brazil 32, 36.8% in Mexico 33, 72.7% in Argentina 34, 33% in South Africa 35, 34.2% in Cameroon 36, from 11.5% (rural) to 26.5% (urban) in Zimbabwe 37, 62% in Senegal 38, 30.1%–67.3% in Turkey 39, 13%–53.9% in Serbia 40, 27% in Kosovo 41, and 7.1% in Moldova 42. The Gulf SAFE registry revealed similarly low rates of anticoagulation use (49% of patients) in six Gulf countries (Bahrain, Kuwait, Oman, Qatar, United Arab Emirates and Yemen).

Most evidence on AF knowledge-practice gaps LMIC focuses on gaps in management of stroke risk among AF patients with OACs. For example, research in Canada suggested that non-cardiologist physicians lack sufficient knowledge, skills and confidence to diagnose AF, with diagnosis of paroxysmal or asymptomatic AF being particularly challenging, and that continuous professional education and development is necessary to strengthen the capacity of physicians to navigate AF screening and diagnosis guidelines.

Patients with AF from middle and low income countries are significantly younger than those from high income countries, the diseases accompanying with their AF are quite different and these patients are less likely to be managed according to published AF guidelines. Despite large regional variations in prevalence, hypertension is the most common risk factor for AF globally. Thus to prevent AF, more effective strategies to diagnose, treat and control hypertension need to be implemented. This need is greatest in Eastern Europe and Africa. Rheumatic heart disease remains an important cause of AF in Africa, China, the Middle East and particularly in India, where it is present in nearly one-third of patients with AF. Although rheumatic heart disease is strongly associated with lower socioeconomic status , low-cost strategies to prevent the development and recurrence of rheumatic heart disease could have a great impact on preventing the development of AF and its complications.

Use of oral anticoagulants in these AF patients at risk for stroke was 40% or less in Eastern Europe, India and Africa where access to INR (International Normalized Ratio) testing and follow-up are limited, and was only 11% in China where physicians have concerns about the rate of warfarin-related intra-cranial hemorrhage and disagreement exists as to the appropriate therapeutic ranges of INR. INR control was also particularly poor in these regions, but only approached desired levels in Western Europe. Low cost generic strategies may be considered to overcome these hurdles in low- and middle income countries, using non-physician health workers especially at local community centers or even in people’s homes. The advent of novel oral anticoagulants may help to overcome some of these barriers; however, their current high costs may make these novel drugs affordable for only a select group of individuals who are either wealthy or have private health insurance.

Past research conducted in the UK concluded systematic screening in general practice using 12-lead ECG was not cost-effective versus routine practice for the detection of atrial fibrillation in people aged 65 and over. Various handheld or smartphone devices designed to specifically detect AF have gained much popularity. Of these state-of-the-art devices, a smartphone-based heart monitor that incorporates dry electrodes mounted on a smartphone case capable of recording a single-lead ECG, has already been US Food and Drug Administration-cleared and Conformité Européene-marked. In Europe, an automatic oscillometric blood pressure device that incorporates a specific algorithm to detect AF has been recommended by the UK National Institute for Health and Care Excellence to screen for AF during office blood pressure measurement in patients ≥65 years of age.

Stroke risk, relatively low oral anticoagulant prescription, and lack of knowledge of diagnosed AF highlight the need for community-based screening and education. ECG screening in pharmacies is a cost-effective mechanism to address these gaps, and could possibly reduce the high cost and social burden of stroke and systemic thromboembolism associated with AF. Increased physician and patient awareness through educational strategies, management and prevention approaches and ultimately to changes in public policy are required to maximize the benefit in low and middle-income countries. Government level policy changes are needed to improve hygiene and access to affordable health care to control the growing incidence of rheumatic heart disease. Strategies may also include screening for rheumatic heart disease and systems of care to include routine electrocardiograms once in five years in elderly or patients with specific risk factors for AF, such as hypertension, vascular disease or valve disease.

The advent of small devices that can diagnose AF automatically, based on either pulse irregularity or rhythm analysis of a single‐lead handheld ECG could change the cost‐benefit equation in favor of a more systematic approach to screening for AF either in the clinic or in the community.

According to Dr. Ngai Yin Chan, Princess Margaret Hospital in HK, the incidence of AF increases with age. By 2030 there will be a 50% rise in AF leading to almost 80,000 patients in honk Kong. International studies show that 30% of patients with stroke caused by AF did not know they have heart rhythm disorder until they have stroke. The Afinder screening program was conducted. Community level screening was done using single lead ecg device which is connected to mobile device and record ecg in 30 seconds which is then uploaded to iCloud. Prevalence of AF increases to 43.5% with this program. Under this program 10734 patients were screened. 244 had AF, with 1/3 previously undiagnosed. Key issues in this program was lack of education as only 64.9% identified have sought medical attention .Those that saw doctors, only 47% received appropriate treatment. Drug compliance who sought medical attention was strong, around 99%.

13,122 Hong Kong citizens voluntarily participated in a territory-wide community-based systematic AF screening programme. ECG was performed by a group of trained non-medical volunteers using the single lead ECG device. The device consists of a smartphone software application and a small handheld hardware component attached to or in close proximity to the phone. 0.4% of ECGs, which were performed by trained non-medical volunteers, were uninterpretable. A steep rise in the prevalence of AF from the age of 60 onwards with a sensitivity of 98% and specificity of 29.2% in detecting newly diagnosed AF was observed. This may serve as an age cut-off criterion to improve the cost-effectiveness of AF screening in the community.

The importance of AF to Indian public health is underscored by an ongoing stroke epidemic with an estimated 1.6 million Indians experiencing a stroke in 2015. Residents from 6 villages in Gujarat, India, were screened for AF using Mobile device. A total of 235 participants aged 50 years and older (half female) used device for 2 minutes on 5 consecutive days. Data was collected over 6 week period. Community health workers helped to screen participants. The prevalence of AF increased by the number of screenings, from 3.0% with 1 screening to 5.1% with 5 screenings.

The best device for the purpose of AF detection for screening or diagnosis will need to be individualized to the situation and the healthcare system. It will be important to optimize sensitivity for AF detection while preserving specificity, to reduce the workload and expense caused by false‐positive readings, and better noise detection and cancellation algorithms to reduce false positives.

Primary care in some countries only are potentially well-equipped and ready to deliver AF screening, with most practices having the ability to detect pulse irregularities, perform and interpret 12-lead ECGs. Compared to GPs, other HCPs have less knowledge and skills for interpreting 12-lead ECGs and diagnosing AF. Nurses may have the greatest potential to up-skill and can play an important role in supporting AF screening. However, many barriers to AF screening exist including lack of practice capacity, staffing and funding, and overcoming such barriers would enable screening implementation.

The increasing worldwide burden of AF necessitates advanced, effective and widely available approaches for diagnosis and prevention. Mobile technologies, such as the recently introduced single lead ecg devices, offer an opportunity to address these concerns. Digital health technology, such as tablet-based risk assessment tools, mobile-phone apps for physicians, and text messaging interventions, represents a new approach for stroke prevention and control. Many studies on digital health, including some in LMICs, are ongoing and are expected to provide evidence on how best to utilize these technological means for diagnosis and control. The introduction of these devices into routine practice could have a substantial impact on reducing the stroke burden and thereby decreasing the morbidity and mortality rates.