Clinical relevance of smartphone apps for diabetes management: A global overview

Zhilian Huang*1,2, MPHZHUANG014@e.ntu.edu.sg

Michael Soljak1, PhDMICHAEL.SOLJAK@ntu.edu.sg

Bernhard Otto Boehm3, MDBERNHARD.BOEHM@.ntu.edu.sg

Josip Car1, PhDJOSIP.CAR@ntu.edu.sg

1Centre for Population Health Sciences, Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore2NTU Institute for Health Technologies (Health Tech NTU), Interdisciplinary Disciplinary School, Nanyang Technological University, Singapore3Immuno-Metabolism Laboratory, Lee Kong Chian School of Medicine, Nanyang Technological University, Singapore

* Corresponding author:

Zhilian Huang, Centre for Population Health Sciences, Lee Kong Chian School of Medicine, Nanyang Technological University, Clinical Sciences Building, Level 18, 11 Mandalay Road, Singapore 308232. Tel.: +65 6592396. Email: zhuang014@e.ntu.edu.sg

Key words: mHealth; health apps; diabetes management; global health; digital interventions

Word Count: 3323No. of tables: 1No. of figures: 3

12

3

456789

1011121314151617181920212223

24

25

262728293031

32

33

34

35

363738394041

ABSTRACT

AIMS: We assessed the number, proportion and clinical relevance of diabetes self-management apps in major languages spoken by the top ten countries with the greatest number of people with diabetes (PWDs).

METHODS: China, India, USA, Brazil, Russian Federation, Mexico, Indonesia, Egypt, Japan and Pakistan were identified as the top ten countries with the largest numbers with diabetes based on the latest NCD-RisC survey. Android and iOS apps in the ten national languages were extracted with a search strategy. App titles and descriptions were systematically screened by trained reviewers, including apps specific for diabetes self-management and excluding apps for healthcare providers, general well-being, health and product promotion and traditional cure. Eighteen apps in the above languages were then downloaded based on availability and popularity, and assessed for clinical relevance to diabetes self-management with reference to current clinical guidelines.

RESULTS: The diabetes-related search terms identified 3374 Android and 4477 iOS apps, where 1019 Android and 1303 iOS apps were screened as being relevant for diabetes self-management. Chinese and English language apps constitute above 80% of the diabetes apps, have more downloads, and more comprehensive clinically relevant functions compared to other languages. None of the apps assessed met all criteria for information provision and app functionalities, nor provided information cited from accredited sources.

CONCLUSIONS: English and Chinese literate patients with diabetes will have more app choices to self-manage diabetes, but may face challenges in selecting the most suitable app. Moving forward, apps should preferably be regulated, context specific and more tailored to users’ needs to best complement multifaceted diabetes care.

4243

4445

464748

495051525354555657

585960616263

64656667

68

INTRODUCTION

Diabetes is a complex, chronic disease characterized by the body’s failure to properly use and store glucose.1 The global age-standardized prevalence of diabetes is estimated to be 9.0% in males and 7.9% in females.2 Over 400 million adults currently have diabetes, and this number is expected to increase by 50% in the next 25 years if the rise is not tackled.3,4 Poorly-managed diabetes leads to severe complications such as stroke, myocardial infarction (MI, “heart attack”), kidney failure, nerve damage, blindness and leg amputation. Diabetes attributed complications not only decrease the quality of life of the patient, but are also costly to the patients, their families and the economy.5 The NCD Risk Factor collaboration (NCD-RisC) estimates the global direct annual cost of diabetes to be $825 billion in 2014.2 Diagnosed diabetes cost the United States US$245 billion in 2012, attributable to direct medical costs and loss in work productivity6. While interventions such as drugs and lifestyle changes can facilitate good glycemic control, there is no fixed approach to diabetes management as individual responses to the disease is unique to the duration and stage of diabetes, life expectancy and predisposition to diabetes complications.7 With the large number of people with diabetes (PWDs) and a shortage of diabetes health professionals,8 innovative solutions utilizing digital interventions can enable better diabetes care as part of multifaceted intervention following validated treatment guidelines.

Health-related smartphone applications (apps) started gaining prominence in 2010, and studies have demonstrated the utility of smartphone apps in complementing diabetes care. Meta-analyses have found that apps can help to improve blood glucose control and strengthen the perception of self-care in PWDs by enhancing health knowledge. 9,10

The number of smartphone users have surpassed 2 billion in 2016 and is expected to increase to 2.86 billion by 2020.11 Currently, the Android and Apple markets represent 80% and 15% of the apps market share respectively. The number of health apps worldwide have more than doubled to 318,000 in 2017 in a span of 2 years.12 According to the latest report from the IQVIA Institute for Human Data Science, strong clinical evidence in app efficacy exists for diabetes care, and may be considered for incorporation into standard of care recommendations.12 With increasing positive evidence, the number of diabetes management apps are expected to further increase in magnitude and capabilities with positive proof-of-concept work.

While health app developments are advancing rapidly, the number of apps intended for diabetes self-management is unknown. Studies of health apps arfe also generally limited to English language apps. Disparities in health apps access and usage exist within and between health systems.13,14 With the high number of non-English speaking people with diabetes worldwide, it is imperative to understand the global diabetes apps landscape and appreciate the applicability of high quality mobile health apps for different populations and contexts. To get an overview of the diabetes apps landscape, we assessed the number and proportion of diabetes self-management apps in major languages spoken by the top ten countries with the greatest number of PWDs, and assessed the clinical relevance of selected apps in each language.

6970717273747576777879808182838485868788899091

9293949596979899

100101102103104105106107108

109

MATERIALS AND METHODS

Selection of apps

The method of app selection and identification is similar to the process of a systematic literature review.15 In brief, China, India, United States of America (USA), Brazil, Russian Federation, Mexico, Indonesia, Egypt, Japan and Pakistan were identified as the top ten countries with the largest number of people with diabetes based on the latest NCD-RisC publication.4 App users are assumed to prefer to use apps in their mother tongue. Hence, the languages spoken and used most frequently by majority of the population in these countries were included in this study. They are Chinese for China, Hindi for India, English for USA, Portuguese for Brazil, Russian for Russian Federation, Spanish for Mexico, Bahasa Indonesia for Indonesia, Arabic for Egypt, Urdu for Pakistan and Japanese for Japan. Hindi and English are the official languages in India. Although only approximately 12% of the Indian population speak English, India has the second largest English speaking population after USA. App searches in Indian vernacular languages returned close to zero results, and were identified from screening the English search results. These apps would either have partial descriptions in a vernacular language or English descriptions indicating that the app is in a vernacular language”. In addition, apps in German, French, Tamil, Bengali, Danish, Korean, Norwegian, Polish, Malay, Filipino and Swedish were searched to ensure that major international languages are covered.

The titles and descriptions of Android and iOS smartphone apps were extracted in ten languages with a search strategy. All extracted app titles and descriptions were checked for language relevance on google translate and sorted by the identified languages. Other unspecified languages that were identified were dropped from the analysis. Duplicated apps in the same language and on the same app platform were then removed before being screened for relevance to diabetes self-management by trained reviewers.

Search strategy

Diabetes related search terms such as “Diabetes”, “Glucose”, “Insulin” and variations of these words were searched on the Android and iOS platforms in June 2017 in ten languages (See exhibit 1). Chinese language apps were additionally searched on third-party platforms such as “Baidu”, “Wandoujia” and “360 zhushou” due to Google restrictions in China. Native speakers translated the terms from English to Chinese, Arabic, Spanish, Portuguese, Russian, Japanese, Hindi, Urdu and Bahasa Indonesia.

<Insert Table 1>

110

111

112

113114115116117118119120121122123124125126127128

129130131132133134

135

136

137138139140141142

143

144

145

App screening

App titles and descriptions were screened in English, Chinese, Arabic, Spanish, Portuguese, Russian, Japanese, Bahasa Indonesia and Urdu. The purpose of app screening was to identify the number of apps relevant for diabetes self-management from the search terms for a more accurate assessment of the proportion of apps in different languages. Self-management in this context refers to behavior or actions an individual can partake without the presence of a healthcare professional to manage diabetes. Apps in all languages were correctly classified by the language detector except for English apps, as a mixture of English and other languages are mentioned in some app descriptions. These apps were checked and reclassified into the appropriate language. Approximately 1% of the apps were translated to multiple languages and were duplicated across the languages covered by this study. All apps were screened by native speakers.

Prior to the full screening, 100 English apps were selected randomly for adjustments of the inclusion and exclusion criteria by six reviewers independently. Fleiss' generalized kappa coefficient, an adaptation of Scott’s pi was used to calculate the inter-rater agreement among all reviewers.16 An agreement of between 60% - 80% represents a reasonably good agreement between the reviewers.17 Disagreements were discussed, and the apps were rescreened until a good agreement was achieved between the reviewers. The final inclusion and exclusion criteria are shown below:

Inclusion criteria

Targeted at people with all types of diabetes For self-management or informal care givers of people with diabetes Recipes-specifically for diabetes

Exclusion criteria

Apps intended solely for healthcare provider’s use Medical dictionaries for doctors or patients Apps for diabetes prevention or prediction of diabetes risk Apps promoting pharmaceutical products without any diabetes self-management

components Apps for general well-being without specific focus on diabetes Apps intended for other chronic diseases other than diabetes Apps espousing traditional cure Demonstration/trial apps

Apps assessment

146

147148149150151152153154155156157

158159160161162163164

165

166167168

169

170171172173174175176177178

179

180

181

182

Selected apps were downloaded and assessed against a checklist (See supplementary material) developed with reference to the 2017 American Diabetes Association (ADA) clinical guidelines. 18 Up to three apps in each language were selected from the list of apps screened to be suitable for diabetes self-management based on availability and downloads and popularity. Selected apps must be suitable for adults with type 2 diabetes and at least have information or a functionality supporting blood glucose monitoring, as optimal blood glucose control is often the primary outcome of diabetes management.

Data analysis

Apps were profiled according to app platform and language. Descriptive summary statistics was used to collate the number and proportion of apps according to languages. Excel 2016 was used to conduct all analyses.

RESULTS

Overall, the diabetes-related search terms identified 3374 Android and 4477 iOS apps from the respective app markets. After screening, 1019 Android and 1303 iOS apps were screened as being relevant for diabetes self-management. Of the included apps, 193 apps were duplicated across the Android and iOS platform. Although the English language search terms returned the highest number of results compared to all other languages, approximately half to a third were re-classified as other languages. Reclassified apps in languages that were not in the scope of this study were excluded from the analysis.

Proportion of apps by languages

The app description screening revealed 27.6% more iOS apps relevant to diabetes self-management compared to Android apps. English and Chinese language apps constitute over 80% of apps for diabetes self-management among the ten languages spoken by countries with large number of people with diabetes on both app platforms (See Figure 2). The inclusion of eleven other languages and dialects prior to screening did not alter the proportion of English and Chinese apps drastically. Chinese language apps have the largest proportion of relevant diabetes self-management apps on the Android platform, possibly due to the large app user base and low barriers of entry to the many third-party android app platforms. The iOS market is dominated by English language diabetes apps.

<Insert table 2>

183184185186187188189

190

191

192193194

195

196

197198199200201202203

204

205

206207208209210211212213214

215

216

217

Relevance of apps from diabetes-related search terms to diabetes self-management

The proportion of apps relevant to diabetes self-management was determined by computing the percentage of apps included from screening the list of apps obtained from the searches. Overall, the searches made by the diabetes-related search terms returned apps that were of low relevance to diabetes self-management (See Figure 3).

The iOS apps inHindustani language (Urdu and Hindi) were all identified from the English search terms, as the search terms in these languages did not return any results. There were five or less apps identified for diabetes self-management in these two languages on each platform, and the apps were mainly focused on diet and education. Despite the high number of apps returned by the English search terms, only approximately a quarter of the apps were determined to be relevant for diabetes self-management. The Chinese Android apps have the highest proportion (55.6%) of apps relevant for diabetes self-management, possibly due to additional searches conducted on third-party platforms.

<Insert Figure 3>

Clinical relevance of diabetes apps

The profile of apps checked against the list of criteria drawn from the ADA guidelines is shown in figure 4. The criteria are grouped according to information provision and app functionality. Apps are colour coded according to language, and criteria met by each app are shaded in the colour which corresponds to the language. The row height of each column shows the logarithm of the number of Android downloads of the app. Majority of the apps assessed were available in both Android and iOS platforms, but Android apps were assessed due to the existence of third party Android platforms in China. Eighteen apps were assessed across all languages. Except for English, Chinese and Japanese apps, we could not find more than two apps with at least information provision or recording functionality of blood glucose level in other languages. Only one app available in various Indian vernacular languages met the criteria for apps assessment. Therefore, the Hindi and Urdu app assessed are the same app.

Apps in English and Chinese language had many more users and more comprehensive diabetes management functionalities in general, compared to other languages. None of the apps assessed met all criteria for information provision and app functionalities. Although the English apps were very close to meeting all app functionalities criteria, none of these top downloaded apps have any form of information provision. Two of three Chinese apps assessed had full information provision, but both apps contained online retail services, with dubious information sources. Of the apps with information provision, none of the apps assessed had information cited from accredited sources.

218

219220221222

223224225226227228229230

231

232

233

234

235236237238239240241242243244245246247248249250251252253254255

As blood glucose level information or functionality was set as the basis of selecting apps for assessment, all but one app had blood glucose monitoring functionality. Insulin, food and oral medication logging were found in two thirds of the apps assessed. Only five and three apps had blood glucose level goal setting and mood tracker functions respectively. Less than a third of the apps assessed provided information on diabetes management, which may imply either a lack of demand for diabetes education or perceived low importance of diabetes education by app developers.

<Insert Figure 4>

DISCUSSION

Our systematic screening identified more than 2000 commercially available apps for diabetes self-management across ten languages, covering more than half of the world’s population. These identified apps are not distributed proportionately across app operating platforms and the languages assessed. English and Chinese speaking populations have a greater variety, and more clinically relevant apps to choose from, while other populations may not have an app of clinical relevance. This may be explained by the connection between demand and innovation, where the demand for health apps stimulated the development of apps with better features, which in turn encouraged demand. Despite the large variety of apps for diabetes self-management, none of the apps assessed for clinical relevance covered all the important app functionalities and provided information from accredited sources. Several observations can be derived from our unpreceded findings.

mHealth adoption is important to benefit from its potential

There is potential for mHealth to improve the quality of healthcare in all countries, but one problem many countries face is the lack of mHealth adoption,19 which undermines any benefit mHealth can bring to population health. The lack of mHealth adoption impedes the growth in demand for health apps and thus the exploration of mHealth as a solution for diabetes care or healthcare.20 The lack of government support, funding, quality assessment, regulation and interoperability are commonly cited reasons for the lack of mHealth adoption, which should be addressed to advance mHealth.21

mHealth is complementary to a multidisciplinary approach

mHealth seeks to complement rather than to replace conventional care. While smartphone apps can facilitate diabetes self-management, a multidisciplinary healthcare team is key to achieving

256257258259260261262263264265266267268269270271272273274275276277278279280281

282

283

284285286287288289290

291

292

293294

satisfactory patient outcomes.22 Health policy makers should engage the right stakeholders early in the process of the intervention. For example, involving both patients and clinicians in the design of an app can increases the possibility of creating a more clinically relevant app. The success of a mHealth intervention is also dependent on partnerships between multiple partners and stakeholders. Local-international and public-private partnerships should bring expertise together to learn from successful cases, avoid past mistakes and to support the administrative, financial and operational aspects of the initiative.

Recommendations on health apps should be incorporated as clinical guidanceThere is a lack of mHealth regulation regardless of the number of apps in a language. This problem is intensified in countries with a gamut of health apps for consumers to choose from.23 Support from an authoritative body and providing clinicians with more guidance on app selection and communication with patients will enhance the incorporation of app use into the clinical pathway. Recommendations should be made based on at least an available, accredited and clinically relevant app. App functionalities and content should be assessed to identify potential ‘useful’ attributes for diabetes self-management, and this information should be used to create a trusted third-party platform to guide app selection. For example, the US Food and Drug Administration (FDA) has issued a consultation on draft guidance for industry and FDA staff,24 and a digital innovation action plan to commit to the regulation of digital health products. The United Kingdom National Health Service launched a digital apps library in April 2017 to host approved (from a technical perspective) healthcare apps that can be trusted by the public. 25 Apps that are listed on the NHS platform are assessed using a set of tools to ensure its safety and validity.

Populations most in need often do not have access to technology

The inverse care law, which states that “the availability of good medical care tends to vary inversely with the need for it in the population served” is observed in the use of mHealth.14,26 Rural populations have lower access to medical resources and are more in need of remote care, but access to technology is often much poorer in rural areas with lower income and poorer ICT infrastructure. Although studies have demonstrated the ability of technology to improve access to care, inequality is still high between younger and older and more and less deprived populations.14 App developers should consider venturing into developing apps with elderly friendly features,27 as the elderly population is capable of adopting mHealth. While e- and m-commerce has grown tremendously in China, mHealth presents more challenges in terms of pricing and regulations.28

Under-developed app markets can focus on creating a few good apps through partnerships, learning from examples, and translation of good apps to context specific content and languages for the population of interest. Only 12% of the Indian population speak English currently. Most of the older and rural Indian population rely on vernacular language for communication, but most Indian apps for diabetes self-management only have an English description. This means that

295296297298299300301

302303304305306307308309310311312313314315316317

318

319

320321322323324325326327328329

330331332333334

the app user has to be proficient in English to use an app in a vernacular language. This presents a major gap as future internet and smartphone populations in India will be older, more rural, more gender balanced and more vernacular. There are a few apps that are translated into various Indian vernacular languages, but no single or a combination of apps covered the full spectrum of diabetes self-management. An example is the ‘Humrahi’ app which educates patients and carers on diabetes care. Translation of app languages will improve the adoption of apps with larger population coverage.

As diabetes is a complex disease, a stand-alone app is less likely to be effective in helping the patient to achieve good diabetes treatment outcomes. A good app should be tailored to users’ requirements and supported by healthcare-affiliated systems. In addition, governments or insurers can consider reimbursing health apps that have shown to be effective in disease management to better integrate apps into the overall care path.

Moving forward, frameworks should be in place to control the quality of apps used for chronic disease management. The reliability of the educational content or advice given by the app is of paramount importance, as such information can potentially elicit behavioral change29. For example, diabetes apps can be checked for the advice given on acute complications surrounding blood glucose fluctuations, and the prevention and management of complications and multimorbidities. These frameworks should also be regularly updated as per accepted diabetes guidelines to keep up with the evidence base. We have since developed a tool to systematically assess the content, functionality and usability of all diabetes apps.

There are some limitations to this study. First, description screening enables the identification of potential apps for diabetes self-management, but not the utility of the app, as the usability and content of the apps were not assessed. Second, the number of apps in the market increases very quickly, limiting any precise estimation of app numbers. Third, apps with multiple language versions may be counted multiple times. Apps intended for the Indian or Pakistani population may also be described in English, but the screening does not enable the identification of the country origin of the developer, thus limiting the classification of app by country. Lastly, relevance to diabetes self-management is context specific and dependent on individual needs, but this study covered a larger range of apps to suit different needs.

CONCLUSION

In conclusion, every patient with diabetes who has a smartphone, speaking one of the above ten languages can access an app for diabetes self-management. English and Chinese literate patients with diabetes will have more app choices to self-manage diabetes, but may face challenges in selecting the most suitable app. A larger variety of health apps can cater to more diverse user needs, but too many unregulated apps that do not meet users’ expectations may lead to app fatigue. In low resource settings, one good app for a specific population may suffice to complement diabetes care. While there is an app for everyone, not everyone can have an

335336337338339340341

342

343344345346347

348349350351352353354355

356357358359360361362363364

365

366

367368369370371372373

app. The future of mHealth is rosy with the rise in smartphone ownership and an increase in positive evidence. The onus is on governments to develop and regulate the industry and to drive mHealth adoption.

374375376

377

378

Acknowledgements

The authors acknowledge Tan Guan Zhong, Chua Kee Leng, Janis Lee, Wang Hua Xian, Martinengo Laura Beatriz, Tarig Osman, Geronimo Jimenez Larrain, Aspembitova Ayana, Elaine Lum Pooi Ming, Monika Semwal and Shoko Dauwels-Okutsu’s contribution in screening the apps list and assessing the apps. ZH would also like to acknowledge Elaine Lum Pooi Ming for her guidance on the manuscript. ZH conceived the idea, conducted the analysis, wrote and revised the manuscript. MS, BB and JC revised, commented on the manuscript and provided guidance to ZH. This paper represents independent work supported by the Lee Kong Chian School of Medicine, Nanyang Technological University (NTU), Singapore, and NTU Institute for Health Technologies (Health Tech NTU). The views expressed are those of the authors and not necessarily those of the NTU.

Conflicts of InterestNone declared

Figure legends

Table 1Caption: Diabetes search terms in ten languagesSource: Translations of diabetes search terms by native speakers

Figure 2 Caption: Number and proportion of diabetes apps for diabetes self-management after app description screening Source: Author’s analysis of data extracted from app meta-data platforms

Figure 3 Caption: Proportion of apps relevant to diabetes self-management by languageSource: Author’s analysis of data extracted from app meta-data platforms

Table 4Caption: Profile of apps assessed in ten languages against a list of ADA clinical guideline derived checklistSource: Author’s analysis of data from downloaded apps

379380

381382383384385386387388389390

391392393394395

396397398399400401402403404405406407408409410411412413414415

416

417

418

REFERENCES

1. American Diabetes Association. Diagnosis and Classification of Diabetes Mellitus. Diabetes Care. 2012;35(Supplement 1):S64-S71.

2. Collaboration NRF. Worldwide trends in diabetes since 1980: a pooled analysis of 751 population-based studies with 4· 4 million participants. The Lancet. 2016;387(10027):1513-1530.

3. Vos T, Barber RM, Bell B, et al. Global, regional, and national incidence, prevalence, and years lived with disability for 301 acute and chronic diseases and injuries in 188 countries, 1990–2013: a systematic analysis for the Global Burden of Disease Study 2013. The Lancet. 2016;386(9995):57.

4. Whiting DR, Guariguata L, Weil C, Shaw J. IDF Diabetes Atlas: Global estimates of the prevalence of diabetes for 2011 and 2030. Diabetes Research and Clinical Practice.94(3):311-321.

5. Funnell MM, Brown TL, Childs BP, et al. National Standards for Diabetes Self-Management Education. Diabetes Care. 2009;32(Supplement 1):S87-S94.

6. Economic Costs of Diabetes in the U.S. in 2012. Diabetes Care. 2013;36(4):1033-1046.

7. Prato SD, LaSalle J, Matthaei S, Bailey C. Tailoring treatment to the individual in type 2 diabetes practical guidance from the Global Partnership for Effective Diabetes Management. International journal of clinical practice. 2010;64(3):295-304.

8. Darzi A, Evans T. The global shortage of health workers--an opportunity to transform care. The Lancet. 2016;388(10060):2576.

9. Hou C, Carter B, Hewitt J, Francisa T, Mayor S. Do Mobile Phone Applications Improve Glycemic Control (HbA1c) in the Self-management of Diabetes? A Systematic Review, Meta-analysis, and GRADE of 14 Randomized Trials. Diabetes Care. 2016;39(11):2089-2095.

10.Bonoto BC, de Araújo VE, Godói IP, et al. Efficacy of Mobile Apps to Support the Care of Patients With Diabetes Mellitus: A Systematic Review and Meta-Analysis of Randomized Controlled Trials. JMIR mHealth and uHealth. 2017;5(3):e4.

11.Statista. Number of smartphone users worldwide from 2014 to 2020 (in billions). 2016; https://www.statista.com/statistics/330695/number-of-smartphone-users-worldwide/. Accessed 7th February, 2017.

12.science IIohd. The Growing Value of Digital Health: Evidence and Impact on Human Health and the Healthcare System. 7th November 2017 2017.

13.Singh K, Drouin K, Newmark LP, et al. Many Mobile Health Apps Target High-Need, High-Cost Populations, But Gaps Remain. Health Affairs. 2016;35(12):2310-2318.

14.Shen C, Wang PM, Chu TWJ, et al. Health App Possession Among Smartphone or Tablet Owners in Hong Kong: Population-Based Survey. JMIR Mhealth Uhealth. 2017;5(6):e77.

419

420421422423

424425

426427428

429430

431432

433

434435436

437438

439440441

442443444

445446447

448449

450451

452453

15.Cochrane Collaboration. Cochrane Handbook for Systematic Reviews of Interventions. In: Higgins JP, Green S, eds: Cochrane Training; 2017: http://training.cochrane.org/handbook.

16.Craig RT. Generalization of Scott's index of intercoder agreement. Public Opinion Quarterly. 1981;45(2):260-264.

17.McHugh ML. Interrater reliability: the kappa statistic. Biochemia medica. 2012;22(3):276-282.

18.Association AD. Standards of medical care in diabetes—2017 abridged for primary care providers. Clinical Diabetes. 2017;35(1):5-26.

19.Organization WH. Be He@lthy Be Mobile Report January 2013 to December 2014. 2015.

20.Eastwood B. Is Smartphone Use Encouraging Mobile Health Adoption? 2013; http://www.cio.com/article/2389377/healthcare/is-smartphone-use-encouraging-mobile-health-adoption-.html. Accessed 02/07/2017, 2017.

21.Payne JD. The state of standards and interoperability for mHealth. Washington, DC: mHealth Alliance. 2013.

22.Anzaldo-Campos MC, Contreras S, Vargas-Ojeda A, Menchaca-Díaz R, Fortmann A, Philis-Tsimikas A. Dulce Wireless Tijuana: A Randomized Control Trial Evaluating the Impact of Project Dulce and Short-Term Mobile Technology on Glycemic Control in a Family Medicine Clinic in Northern Mexico. Diabetes Technology & Therapeutics. 2016;18(4):240-251.

23.Xiaohui Y, Han H, Du Jiadong W, Liurong LC, Xueli Z, Haihua L. mHealth in China and the United States. Brookings Institution Center for Technology Innovation. 2014:9.

24.US Food and Drug Administration. Mobile Medical Applications: Guidance for Industry and Food and Drug Administration Staff. In: Administration UFaD, ed. Washington DC: US Food and Drug Administration; 2015.

25.Juliet Bauer RM. Apps library is advance for a digital NHS. NHS England. Vol 2017: NHS; 2017.

26.Hart JT. The inverse care law. The Lancet. 1971;297(7696):405-412.

27.Scheibe M, Reichelt J, Bellmann M, Kirch W. Acceptance factors of mobile apps for diabetes by patients aged 50 or older: a qualitative study. Medicine 20. 2015;4(1).

28.Sun J, Guo Y, Wang X, Zeng Q. mHealth For Aging China: Opportunities and Challenges. Aging and Disease. 2016;7(1):53-67.

29.Chavez S, Fedele D, Guo Y, et al. Mobile Apps for the Management of Diabetes. Diabetes Care. 2017:dc170853.

454455

456457

458

459460

461

462463464

465

466467468469

470471

472473

474

475

476477

478479

480

Tables

Table 1: Diabetes search terms in ten languages

Figure 2: Number and proportion of diabetes apps for diabetes self-management after app description screening

English Chinese Spanish Portuguese Japanese Arabic RussianBahasa Indonesia

Urdu Hindi

Diabetes 糖尿病 Diabetes Diabetes 糖尿病 داءالسكرى Диабет Kencing manis/ diabet/ penyakit gula

ذیابیطس मधुमेह , डा यबिटी ज

Diabetic (noun) -Diabético/ Diabética

Diabético/ Diabética 糖尿 مريضبالسكر Диабетик - - मधुमेह रोगी

Diabetic (adj) - - - - - Диабетический -ذیابیتسکامریض -

Diabetics -Diabéticos/ Diabéticas

Diabéticos/ Diabéticas 糖尿病の مرضىالسكرى Penderita

diabetesذیابیطس मधुमेह रोगी

Glucose - Glucosa Glicose グルコースجلوكوز - شراب سكرى Глюкоза glukosa گلوکوز शर्क� रा

Glycaemic / Glycemic

-Glucémico/ Glucémica

Glicêmico / Glicêmica 血糖の نسبةالسكرفىالدم Гликемический glikemik گلیسکیم ग्लि��सी मिमर्क

Blood sugar 血糖 GlicemiaAçúcar sanguínea 血糖 سكرالدم Уровень

сахара в кровиgula darah خونکیشکر रक्त शर्क� रा

Insulin 胰岛素 Insulina Insulina インシュリンانسولي-ن инсулин Insulin انسولی-ن इन् सुलि�न

0

200

400

600

800

1000

1200

1400

Android apps for diabetes self-management iOS apps for diabetes self-management

Num

ber o

f app

s for

dia

bete

s sel

f-man

agem

ent

Number and proportion of diabetes apps relevant for diabetes self-management after app description screening

English Chinese Spanish Japanese Indonesian Urdu Russian Arabic Hindi

67%

26%

91%

45%

40%

85%

481

482

483

484

485

486

487

488

489

490

491

492

493

494

495496

497

498

499

500

501

502

503

504

505

506

507

508

509

510

511

Figure 3: Proportion of apps relevant to diabetes self-management by language

Table 4: Profile of apps assessed in ten languages against a ADA clinical guideline derived checklist

0.0%

10.0%

20.0%

30.0%

40.0%

50.0%

60.0%

70.0%

80.0%

90.0%

100.0%

English Chinese Spanish Portuguese Japanese Indonesian Urdu Russian Arabic Hindi

Prop

ortio

n of

app

s for

dia

bete

s sel

f-man

agem

ent

App language

Proportion of apps relevant to diabetes self-management

Android ios total

†The app in Hindi and Urdu are the same

Russian app 1

Russian app 2

Spanish app 1

Portuguese app 1

Portuguese app 2

Arabic app 1

Arabic app 2

Japanese app 1

Japanese app 2

Japanese app 3

Hindi/Urdu app†

Indonesian app 1

Chinese app 1

Chinese app 2

Chinese app 3

English app 1

English app 2

English app 3

Information provision App functionalities

App

s as

sess

ed in

ten

lang

uage

s ( R

ow h

eigh

t = lo

g(no

. of d

ownl

oads

))

Phys

ical a

ctivit

y

Nutri

tiont

hera

py

Blood

gluc

ose l

evel

monit

oring

Smok

ing ce

ssati

on

Weigh

t man

agem

ent

Medic

ation

Foot

care

Eye c

are

Phys

ical a

ctivit

y tra

cking

Food

logg

ing

Blood

gluc

ose l

evel

goal

settin

g

Blood

gluc

ose l

evel

monit

oring

Weigh

t man

agem

ent

Oral

medic

ation

logg

ing

Insuli

n log

ging

Blood

pres

sure

track

ing

Chole

stero

l trac

king

Mood

trac

king

>500,000

>100,000

>100,000

>100,000

>20,000

>50,000

>1,000

>80,000

>10,000

>1,000

>5,000

>10,000

>5,000

>500

>1,000

>1,000

>10,000

>5,000

No. of Android downloads

512

513

514

515

516

517

518

519

520

521

522

523

524

525

526

527

528

529

530

531

532

533

534

535

536

537

538

539

540

541