BACKGROUND Chronic diseases are the leading cause of global mortality and morbidity, yet majority of the disease burden is poorly self-managed.(1) The rate and cost of chronic disease on the healthcare system is ever-increasing and there has been a recent emphasis on self-management in future healthcare.(2) With the quick advances in information sharing and technology, telehealth has become a useful tool and mode of intervention for health systems.(3) It has been proposed that mhealth can aid ‘compliant self-management’, where patients adhere to clinical recommendations and assimilate their own knowledge of their condition with clinical recommendations to adopt an integrated self-management regime.(2) While secondary prevention is essential in reducing associated mortality and morbidity rates, there is a case for primary prevention in reducing the preventable death and disability rates.(4) Mhealth is an emerging area of intervention delivery that has shown promise in the management of chronic diseases due to its prevalence, portability and capacity for variability and applicability.(1) As of May 2014, seven billion mobile phone subscriptions were in use worldwide.(5) The increased use of mobile phone and modern technologies has created a new medium for digital health interventions delivered by mobile devices.(5) “User-centred mobile health tracking apps have gained widespread popularity by facilitating the maintenance of health and management of chronic conditions and by empowering individuals to contribute to their own well-being and health.”(6) Majority of the World Health Organization member countries make use of mhealth for bi-directional communication between healthcare practioners and patients, emergency response, health management and information access.(3) Although interest in mhealth is increasing, and there is a sufficient research on the use of SMS or phone calls to manage health,(5) specific research into the efficacy of smartphone applications is limited. Furthermore, applications for the prevention of onset of disease is still being developed and, to the best of our knowledge, the current knowledge has not been systematically reviewed. Disease onset is defined as the first time a change has been noted in a patient’s health status with identified signs or symptoms directly attributable to a specific disease process.(7) OBJECTIVE The aim of this study was to systematically review the evidence surrounding the use of smartphone applications to prevent onset of chronic disease by improving self-management to understand the current gaps in knowledge. METHODS Search strategy Several databases including The Cochrane Library, PubMed, Google Scholar, the UQ library and Science Direct were searched. To be as inclusive as possible, following search terms were used in various combinations in all databases: “smartphone”, “mhealth”, “app”, “intervention”, “prevent*”, “long-term disease” and “chronic disease” were used. To identify other relevant studies, reference lists of similar studies were manually reviewed for additional articles. We limited the search by only including peer-reviewed randomised controlled trials (RCTs) and clinical trials, published in the English language after the year 2012. We only included studies published after 2012 as mobile health tracking technology is a relatively recent advancement.(6) Therefore, by limiting the search dates, we excluded non-smartphone interventions. Inclusion and Exclusion Criteria Additional inclusion criteria were trials that focused on treating patients at risk for future health burden due to chronic disease and trials that primarily used a smartphone application for their intervention. We maintained wide parameters because of the current dearth of literature on smartphone apps for the prevention of onset of future chronic disease. The exclusion criteria were interventions that only used short messaging systems (SMS) or telephone calls and protocols. The study selection process is outlined in the PRISMA chart below (Figure 1). Title and abstract search were performed, and relevant information extracted and tabulated by the first author (UN, Table 1). This information was reviewed independently by a second author (SE). We assessed the quality of evidence with the Joanna Briggs Institute Critical Appraisal Checklist for RCTs (JBI) (Table 2). This review was registered in the International Prospective Registry of Systematic Reviews (PROSPERO: CRD42019127184). RESULTS The search identified 561 papers, of which, 9 RCTs published within 2014 and 2017 met the criteria for review.(Table 1) Study characteristics The smallest studies contained 61 patients,(8, 9) and the largest study was an international and multi-centre trial with 2,086 patients.(10) The studies represent a total of 4,298 patients aged 26(9) to 68.(11) All of the included RCTs were from developing nations (Korea, US, Australia, China, India, Singapore, Ireland), although a trial from Ghana showed improvement in preventing future stoke in stroke survivors,(12) indicating the effectiveness of mhealth interventions in developing nations. Four studies recruited via self-referral(8, 9, 13, 14) and the rest used physician identification and other health markers to target patients. Studies recruited patients at risk for a range of chronic illnesses including cardiac-related diseases, diabetes, and blindness. Three studies aimed to prevent onset of a variety of common chronic diseases.(14-16) One study aimed to prevent disability due to chronic illness.(11) Two studies aimed to prevent diabetes.(8, 17) The remaining three studies focused on preventing cardiac-related diseases including stroke.(9, 10, 13) Delivery methods Two studies used text-messaging as a supportive measure.(13, 16) Additionally, four studies used phone calls as a means of supporting patients.(9, 14, 16, 17) Seven studies incorporated education in their intervention, however only the study by Crossland et al directed education at the health professionals.(11) Two studies had treatment as usual control groups,(11, 13) four used comparator trials(8, 10, 14, 16) and three used a waitlist control.(9, 15, 17) Two studies used a smartphone application exclusively.(14, 15) The rest of the studies used more than one mode of mhealth including traditional telephone calls, email and text messaging. Most studies used the smartphone application as a means of monitoring. Text messages were used as reminders in two trials, a means of education dissemination(13) or to inform participants of updates.(16) Telephone calls were implemented in four studies for update or follow-up interview.(9, 14, 16, 17) Email was utilised in two studies for education dissemination(16) or for communication between healthcare professionals.(11) Additionally, six studies implemented face-to-face contact.(8-11, 16, 17) Four studies objectively measured clinical improvements.(8, 11, 14, 17) However, all the studies reported significantly improved outcomes. Quality of evidence Sample sizes of the trial were between 61(8, 9) and 2,086(10), and four studies had a sample size below 100.(8, 9, 13, 16) Quality of evidence in the selected studies varied due to differences in blinding of outcome assessors and sampling bias. However, all the studies reported their trial adequately and majority suffered little attrition. CONCLUSIONS In the past, smartphone applications have often focused on monitoring or educating participants.(3) In more recent times, an observable shift in focus is beginning towards the prevention of disease. However, an effective and thorough review of the current evidence could identify an appropriate audience, parameters and intervention strategies, and identify key gaps in knowledge. This could inform future studies and support the uptake of mhealth in primary care. This study aimed to provide a systematic analysis of the current literature around smartphone applications for disease prevention to inform future research. These nine studies focused on chronic diseases or illnesses with a high health burden, including diabetes, cardiovascular disease, chronic heart disease, stroke, obesity, and blindness. All the studies reported a significant increase in outcomes when using smartphone applications for weight or activity management, glucose levels, atrial fibrillation, disease awareness and knowledge, and screening. Target population Although the studies included in this review focused on the nature of the patient’s disease rather than the suitability of the intervention, participants were either selected who were technology-literate or were given training upon enrolment to the study.(cite) However, the included studies ranged from young to mature adults, indicating the applicability of the intervention mode across age groups. As the studies did not include developing nations, we are unable to conclude if smartphone applications may be generalised across countries. A formal analysis of telemedicine implementation could identify appropriate patients, conditions and settings. Therefore, there is a need for a high-quality evidence-base to support uptake and sustainable integration of telemedicine into routine clinical care. Evaluation of the intervention Only two studies used a pedometer device in addition to the smartphone application.(8, 14) Six studies incorporated education into their intervention, but the study by Crossland et al implemented videoconference education for general practitioners and ophthalmologists.(11) The study by Zhang et al focused their education on the physiology of the heart, risk factors, lifestyle changes and stress management.(13) This was similar to the educational material incorporated into the smartphone app by Tian et al, Tighe et al, Block et al. Glynn et al sent all participants, regardless of randomisation, an educational brochure.(16) Alternatively, Fukuoka et al tailored their education to aiding participants meet their goals.(8) Only four studies incorporated goal setting and majority of the studies used the smartphone application to improve patient monitoring. While majority of the studies did not implement regular contact with the assessors or physicians, the studies by Oh et al and Tian et al had followed patients up at regular intervals. Limitations Some of the limitations of this review included the high number of studies that used self-reporting. Furthermore, there was bias in the recruitment methods of four studies that used self-referral, leading to potential sampling bias. Perhaps the most important limitation of this study is the dearth of published information available. This review only identified nine studies using smartphone applications to prevent the onset of chronic diseases. Additionally, three studies had an intervention period under three months.(9, 13, 16) The studies have identified this as a potential limitation in analysing the effect on long-term behaviour. Conclusion With the trend of rapidly advancing mobile technology, it is important to understand the current evidence to inform future policy aimed at reducing the health burden of a nation. This is especially relevant for policies aimed at integrating mhealth for health service delivery in developed and developing countries. Some illnesses are even associated with one another; for example, overweight and obesity are associated with several chronic diseases, including type 2 diabetes, hypertension, cardiovascular disease, arthritis, hyperlipidaemia, and asthma.(3) CLINICALTRIAL This review was registered in the International Prospective Registry of Systematic Reviews (PROSPERO: CRD42019127184).