John B. Kellogg, BA1, Jessica D. Lee2, Daniel R. Murphy, MD, MBA1,3, Monisha Arya, MD, MPH1,3
1Baylor College of Medicine, One Baylor Plaza, Houston, TX 77030, USA;
2Rice University, 6100 Main St, Houston, TX 77005, USA;
3Center for Innovation in Quality, Effectiveness and Safety, Michael E. DeBakey VA Medical Center, 2002 Holcombe Blvd, Houston, TX 77030, USA;
Corresponding Author: Kellogg@bcm.edu
Journal MTM 8:1:20–28, 2019
Background: Despite the prevalence of hepatitis C virus (HCV) and the availability of effective treatments, HCV screening remains suboptimal, in part due to primary care physicians’ (PCPs) unawareness of and discomfort discussing HCV risk factors. Patient-facing text message campaigns may overcome these barriers by empowering patients to initiate screening discussions with their PCPs.
Aims: The objectives were to evaluate a patient-facing text message campaign in terms of (1) feasibility, (2) acceptability, and (3) impact on patient-PCP discussions about HCV screening.
Methods: Primary care patients were recruited to receive either an HCV text message, which contained HCV information and a prompt to discuss HCV with their PCPs, or a calcium control text message. Forty minutes before their appointments, participants were sent their assigned text message. Participants were then called for an evaluation of the text message campaign.
Results: Of 185 patients called, 38 enrolled and completed the study. Participants who were sent an HCV text message (n=25) were significantly more likely to initiate a conversation with their PCPs about HCV screening than participants sent a calcium control text message (n=13) (p=0.008). Thirty-two (82%) participants liked receiving a health-related text message (88% in the HCV group; 70% in the control group).
Conclusions: A patient-facing HCV text message campaign shows promise as a novel method to activate primary care patients to initiate HCV screening discussions with their PCPs. This campaign may help educate patients about the importance of HCV screening, overcome physician barriers to screening, and, ultimately, help control the HCV epidemic.
Keywords: hepatitis C, text messaging, physician-patient relations, preventive health services, primary health care
Karen Hong1, Sean Collon2, David Chang3, Sunil Thakalli4, John Welling4, Matthew Oliva4, Esteban Peralta5, Reeta Gurung6, Sanduk Ruit6, Geoffrey Tabin1,4, David Myung1,7, Suman Thapa6
1Byers Eye Institute, Stanford University School of Medicine
2Vanderbilt University School of Medicine
3Los Altos Eye Physicians, Los Alto, CA
4Himalayan Cataract Project
5University of Florida School of Medicine
6Tilganga Institute of Ophthalmology, Kathmandu, Nepal
7Division of Ophthalmology, Veterans Affairs Palo Alto Health Care System, Palo Alto, CA
Corresponding Author: email@example.com
Journal MTM 8:1:1–10, 2019
Background: To compare screening referral recommendations made by remotely located ophthalmic technicians with those of an ophthalmologist examining digital photos obtained by a portable ophthalmic camera system powered by an iOS handheld mobile device (iPod Touch).
Methods: Dilated screening eye exams were performed by ophthalmic technicians in four remote districts of Nepal. Anterior and posterior segment photographs captured with a Paxos Scope ophthalmic camera system attached to an iPod Touch 6th generation device were uploaded to a secure cloud database for review by an ophthalmologist in Kathmandu. The ophthalmic technicians’ referral decisions based on slit-lamp exam were compared to the ophthalmologist’s recommendation based on the transmitted images.
Results: Using the transmitted images, the ophthalmologist recommended referral for an additional 20% of the 346 total subjects screened who would not have been referred by the ophthalmic technician. Of those subjects, 34% were referred to the retina clinic. Conversely, among the 101 patients referred by the technician, the ophthalmologist concurred with the appropriateness of referral in more than 97% of cases but thought eight (2.8%) of those patients had variants of normal eye pathology.
Conclusion: An ophthalmologist who reviewed data and photos gathered with the mobile device teleophthalmology system identified a significant number of patients whose need for referral was not identified by the screening technician. Posterior segment pathology was most frequently found by the remote reader and not by the technician performing dilated slit lamp examinations. These results are promising for further clinical implementation of handheld mobile devices as tools for teleophthalmic screening in resource-limited settings.
Keywords: telemedicine, rural population, ophthalmology, referral and consultation, global health
Professional Involvement in Ophthalmology iPhone Application Development: An Update
Louis Stevenson, MBBS1
1 The Royal Victorian Eye and Ear Hospital, East Melbourne, Victoria, Australia 3002
The author is not the recipient of a research scholarship.
Corresponding author: Louis.Stevenson@eyeandear.org.au
Journal MTM 7:2:1–6, 2018
Background: Smartphone technology and related applications are increasingly prevalent in the field of medicine and ophthalmology, offering a wide range of hand-held capabilities not previously available. While these technologies have enormous potential, many apps are developed without the involvement of qualified professionals leading to concerns about their quality and validity.
Aims: To assess iPhone® applications aimed at eye care professionals for qualified professional involvement in their development.
Methods: Applications were identified by searching the Apple® (Cupertino, CA) iTunes® Store using the terms ‘ophthalmology’ and ‘ophthalmologist’ in addition to a number of common eye conditions outlined by the Centers for Disease Control and Prevention. Applications were then assessed for category of application, intended audience, documented involvement of medical professionals in application development, price, user rating and date of publication.
Results: In total, 152 applications were identified across 12 categories. Applications were found to target eye-care professionals (ophthalmologists and non-ophthalmologists) (32.3%), ophthalmologists specifically (32.3%), non-ophthalmology eye-care professionals (3.3%) and patients (34.2%). Overall, 36 (23.7%) applications had clearly documented professional involvement in their development.
Conclusions: There continues to be a low level of professional involvement in the development of ophthalmology based iPhone® applications. This is concerning given the growing prevalence of these technologies and their enormous potential. It is therefore incumbent on clinicians to be informed about the applications they use and promote high quality applications developed with professional expertise.
Disclosures: All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: no support from any organisation for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.
Keywords: ‘Ophthalmology’, ‘smartphone’, ‘technology’, ‘ophthalmologist’, ‘telemedicine’.
Smartphones are mobile, handheld devices, with functional capabilities similar to those of laptop computers.1 These devices represent a significant technological milestone and provide users with a diverse range of easily accessible, handheld capabilities not previously available.
The field of medicine has increasingly been subject to the influence of smartphones and their applications (apps). Between 2001 and 2013, smartphone use among health professionals rose from 30%2 to 86%3, highlighting the dramatic uptake of this technology. Accompanying the rise in smartphone use has been a similarly dramatic increase in the number of health related apps.4, 5 The potential benefits of the technology are extensive and range from increased patient compliance, data management, displacement of old expensive technologies and increased communication capabilities.2
The specialty of ophthalmology has also been impacted by the development of smartphones, with a diverse range of ophthalmology themed apps now available.6 These apps target a wide range of audiences from ophthalmologists and non-ophthalmology eye care professionals, through to medical researchers and healthcare consumers.6, 7 The range of functions typically performed by these apps fall into several broad categories including; clinical examination and assessment tools, medical administration, professional and patient education, and clinical calculators.2, 5–8 The areas of telemedicine and teleophthalmology have also been impacted by smartphones9 with a Brazilian study finding smartphone based photography to be both sensitive and specific in the diagnosis of emergency eye conditions when used as a teleconsultation tool.10 These results are highly promising and have the potential to improve access to healthcare in isolated populations.
While these capabilities are unprecedented and offer enormous potential, there is concern that their utility is undermined by low quality app development and a lack of evidence supporting their use.2 A 2014 report by Cheng et al. found that less than one third of ophthalmology iPhone® apps available on the Apple® iTunes® Store, had documented medical input in their development.7
Such concerns are not limited to ophthalmology with similar findings having been made across other disciplines. A 2011 review of smoking cessation apps found many apps deviated significantly from relevant clinical guidelines11, while a 2016 review of health related smartphone apps found that the majority of apps targeting clinical practice lacked scientific evidence underpinning their use.12 Furthermore, apps are difficult to regulate and a number of privacy issues related to their use have been raised.13
Given these findings, a major challenge now facing clinicians, researchers, and consumers is harnessing the potential of this technology while avoiding its shortcomings. This report provides an update on the quality of ophthalmology apps available on the Apple® iTunes® Store with specific attention paid to the involvement of medical professionals in app development. These findings will be compared to those previously published to determine whether there has been a change in the quality of app development over recent years. This is a topic of significance given the recognised importance of evidence-based practice, and the potential for poor quality, or unproven technologies to undermine this.14
Materials and methods
The Apple® iTunes® Store was searched on Saturday 17 March 2018 to identify ophthalmology related apps. Where appropriate, this study has adopted some of the methodology used previously by Cheng et al.7 in order to facilitate comparisons between the two reports.
Apps were considered for inclusion provided they were returned using the search terms ‘ophthalmology,’ and ‘ophthalmologist’. Additionally, the Apple iTunes Store was searched using terms derived from common eye disorders as outlined by the Centers for Disease Control and Prevention15 including ‘refractive’, ‘macular degeneration’, ‘cataract’, ‘diabetic retinopathy’, ‘glaucoma’, ‘amblyopia’, and ‘strabismus.’
Apps relating to fields other than human ophthalmology such as veterinary ophthalmology or the physics of light were excluded. Apps were also excluded where the primary purpose was to promote a particular practitioner, conference, institution or product. Similarly, ‘demo’ or ‘lite’ versions of apps were excluded.
Data collection and statistical analysis
The following data were collected within the Apple® iTunes® Store based on information provided in the app product description; app name, developer, target audience, description of app function, category of app, cost, iTunes® rating, documentation of medical involvement in app development, year first published or copyright year, and size of the app in megabytes (MB). Professional involvement was defined as clear documentation of medical professional involvement or reputable institutional involvement such as that by a university or hospital in app development. Individual medical practitioners did not need to be identified in order to satisfy this criterion. Information was not sought from external websites, via contacting app developers or within the apps themselves. The data were analysed using descriptive statistics in Excel (Microsoft, Redmond WA).
This research was conducted in accordance with the relevant ethical guidelines. No human participants were involved.
A total of 152 ophthalmology themed apps were identified with the above search terms. Apps were categorised into 11 categories according to their primary function in addition to a miscellaneous category (Fig. 1).
Figure 1: Categories of ophthalmology applications and their relative distributions
Eleven apps (7.2%) were aimed at providing educational material to all eye-care professionals (ophthalmologists and non-ophthalmologists) and included atlases and written reference material. Five (3.3%) apps were aimed at providing educational material to non-ophthalmology healthcare professionals and included flash cards and quizzes. Eleven (7.2%) apps provided educational material aimed at ophthalmologists specifically and primarily included written reference material. Six (3.9%) apps were electronic versions of academic journals. Eleven (7.2%) apps provided patient education including information delivered in written and multimedia formats. Furthermore, a number of apps provided ophthalmologists with images to be used for patient education such as in the setting of pre-operative counselling. Thirty-seven (24.3%) apps were clinical examination tools and included visual acuity charts, Amsler grids and colour vision testing plates. Twenty (13.2%) apps were clinical calculators of various types including toric and non-toric intraocular lens calculators, glaucoma risk calculators, and visual acuity converters.
Twenty-one (13.8%) apps provided treatment for amblyopia and strabismus through games or visual tasks. Seven (4.6%) apps functioned as low vision aids by providing screen magnification or allowing users to change the display colour scheme to better suit those with colour vision deficits. Five (3.3%) apps were aimed at medication compliance, all of which were eye drop reminders. Sixteen (10.5%) apps were categorised under miscellaneous. Apps in this category included an operator simulator, ophthalmology media, clinical administration, social media and telehealth.
The apps identified on the iTunes® Store were aimed at 4 different audiences. Forty-nine (32.3%) apps were targeted at ophthalmologists, while 5 (3.3%) and 46 (30.3%) were targeted at non-ophthalmology eye care professionals and eye care professionals (ophthalmologists and non-ophthalmologists), respectively. Fifty-two (34.2%) apps were targeted at patients.
Forty-two (27.6%) apps had five or more user ratings resulting in them having an overall rating on the iTunes® Store. The mean rating amongst these apps was 3.4 stars. In contrast, 110 (72.4%) apps did not have an average user rating provided as they had been rated less than 5 times.
Medical professional involvement
In total, 36 (23.7%) apps had clearly documented medical involvement, or were developed by a reputable organisation such as a hospital or university. There were 34 (22.4%) apps that were assumed to have medical or professional input into their development because of their content or presentation, however they did not explicitly state this. A total of 82 (53.6%) apps had no professional involvement in their app development (Tab. 1).
Table 1: Professional involvement in app development by category of app
Year of publication
The number of apps produced each year from 2009 to 2018 (inclusive) was; 3 (2.0%), 8 (5.3%), 15 (9.9%), 10 (6.6%), 15 (9.9%), 18 (11.8%), 29 (19.1%), 20 (13.2%), 23 (15.1%) and 11 (7.2%) (Fig. 2). The 2018 figure of 11 represents the number of apps produced until 17 March 2018 when the search was conducted.
Figure 2: Number of ophthalmology themed apps by year of publication or copyright
The mean and median price of apps was USD$5.25 and $USD0.00, respectively. The price of apps ranged from USD$0.00 – USD$99.99. Eighty-nine (58.6%) apps were free while 8 (5.3%), 11 (7.2%), 7 (4.6%), 19 (12.5%) and 19 (12.5%) apps were USD$0.99, USD$1.99, USD$2.99, USD$3.00 – USD$10.00 and >USD$10.00 respectively.
This report provides an update on the involvement of medical professionals in the development of ophthalmology apps available through the Apple® iTunes® Store. A similar report published in 2014 identified a total of 182 ophthalmology themed apps compared to 152 identified in this report,7 however significantly broader search terms were used in this previous study. As such, these results likely confirm that there are an increasing number of ophthalmology apps available. This is supported by the significant increase in number of ophthalmology apps released each year. While three apps were released in all of 2009, 11 apps have already been released as of 17 March 2018 highlighting the rapid growth in ophthalmology themed iPhone® apps over the past decade.
A broader number of app categories were identified in this report compared to those identified in previous studies.7 Examples of these new categories include low vision aids, those targeting medication compliance in the form of eye drop reminders and apps that provide treatment for strabismus and amblyopia.
Clearly documented professional involvement or clear documentation of reputable institution involvement in app development was low, with only 23.7% of apps fulfilling these criteria. These findings are considerably lower than figures quoted in previous studies.7 This is especially concerning given the very low standard required to meet these criteria. For example, merely stating that there was doctor involvement was sufficient with no requirement to provide evidence for these claims. Additionally, simply having medical involvement in app development does not in any way validate an app against non-iPhone® standards. It is not within the scope of this report to assess the validity of all available apps however. When combined with those apps that were assumed to have professional involvement this figure rose to 46.1%.
While a lack of professional involvement in some categories such as eye drop alarms may be benign, in other categories of apps it may be dangerous. For example, almost 86% of apps that aimed to provide ‘lazy eye training’ in the setting of amblyopia and strabismus had no professional involvement in their development. These apps primarily consisted of games and superficially they may appear harmless, however they may represent an opportunity cost and deny patients of time spent performing proven treatments. This highlights the need to validate these apps to avoid adverse outcomes. Additionally, apps that provide visual assessment tools such as visual acuity or colour vision testing may provide clinicians and patients with false information if improperly designed.
Such limitations have already been identified within the literature. A 2015 report by Perera et al. investigating the validity of iPhone® based visual acuity charts failed to identify a single app of sufficient accuracy for clinical use16 while discrepancies between iPhone® based, and standard Ishihara charts have also been documented.17 These findings are particularly concerning given that clinical examination apps account for approximately one-quarter of the app market and have the potential to directly lead to clinical errors and adverse outcomes.
While the large number of free apps available may seem beneficial, it may have the effect of promoting the development of poor quality apps due to lower standards expected and tolerated by users. Additionally, the large number of apps aimed at non-ophthalmologists may also promote poor development given that this target audience is likely to lack the expertise needed to critically appraise the technology.
These findings are a significant concern given the growing prevalence and influence of smartphone technology in medicine and ophthalmology.2–5 With the exception of a few, current smartphone apps are poorly designed, lack evidence to support their use and as such, the technology is not reaching its full clinical potential. Despite this, it seems highly unlikely that ophthalmology based smartphone technology will not continue to grow and offer an ever increasing range of functions. As such, it is incumbent on clinicians to promote high-quality, evidence-based apps where possible in order to derive maximum benefit from this technology, whilst avoiding the inherent pitfalls.
This report provides and an up to date review on the quality of ophthalmology iPhone® app development with the results highlighting the low level of professional involvement in this process. This is concerning given the growing prevalence of these technologies and their enormous potential. It is therefore incumbent on clinicians to be informed about the applications they use and promote high quality applications developed with professional expertise.
1. BinDhim NF, Freeman B, Trevena L. Pro-smoking apps for smartphones: the latest vehicle for the tobacco industry? British Medical Journal. 2012;23(1):e4[1–8].
2. Mosa AS, Yoo I, Sheets L. A systematic review of healthcare applications for smart phones. BMC Med Inform Decis Mak. 2012;12(67).
3. Epocrates. 2013 mobile trends report [Available from: http://www.epocrates.com/oldsite/statistics/2013 Epocrates Mobile Trends Report_FINAL.pdf.
4. Kulendran M, Lim M, Laws G, Chow A, Nehme J, Darzi A, et al. Surgical Smartphone Applications Across Different Platforms: Their Evolution, Uses, and Users. Surgical Innovation. 2014;21(4):427–40.
5. BinDhim NF, Trevena L. There’s an App for That: A Guide for Healthcare Practitioners and Researchers on Smartphone Technology. Online Journal of Public Health Informatics. 2015;7(2):e218–5.
6. Zvornicanin E, Zvornicanin J, Hadziefendic B. The Use of Smart Phones in Ophthalmology. Acta Inform Med. 2014;22(3):206–9.
7. Cheng NM, Chakrabarti R, Kam JK. iPhone applications for eye care professionals: a review of current capabilities and concerns. Telemed J E Health. 2014;20(4):385–7.
8. Bastawrous A, Cheeseman RC, Kumar A. iPhones for eye surgeons. Eye. 2012;26:343–54.
9. Grisolia ABD, Abalem MF, Lu Y, Aoki L, Matayoshi S. Teleophthalmology: where are we now? Arq Bras Oftalmol. 2017;80(6):401–5.
10. Ribeiro AG, Rodrigues RA, Guerreiro AM, Regatieri CV. A teleophthalmology system for the diagnosis of ocular urgency in remote areas of Brazil. Arq Bras Oftalmol. 2014;77(4):214–8.
11. Abroms LC, Padmanabhan N, Thaweethai L, Phillips T. iPhone Apps for Smoking Cessation: A Content Analysis. American Journal of Preventative Medicine. 2010;40:279–85.
12. Gan SK-E, Koshy C, Nguyen P-V, Haw Y-X. An overview of clinically and healthcare related apps in Google and Apple app stores: connecting patients, drugs, and clinicians. Scientific Phone Apps and Mobile Devices. 2016;2(8).
13. BinDhim NF, Trevena L. Health-related smartphone apps: regulations, safety, privacy and quality. BMJ Innovations. 2015;1(2).
14. Birbeck GL, Wiysonge CS, Mills EJ, Frenk JJ, Zhou X-N, Jha P. Global health: the importance of evidence-based medicine. BMC Medicine. 2013;11(223).
15. Centers for Disease Control and Prevention. Common Eye Disorders United States of America: CDC; 2015 [Available from: http://www.cdc.gov/visionhealth/basics/ced/index.html.
16. Perera C, Chakrabarti R, Islam FMA, Crowston J. The Eye Phone Study: reliability and accuracy of assessing Snellen visual acuity using smartphone technology. Eye (Lond). 2015;29(7):888–94.
17. Sorkin N, Rosenblatt A, Cohen E, Ohana O, Stolovitch C, Dotan G. Comparison of Ishihara Booklet with Color Vision Smartphone Applications. Optometry and Vision Science. 2016;93(7):667–72.
Designing a WIC App to Improve Health Behaviors: A Latent Class Analysis
Sylvia H. Crixell PhD, RD1, Brittany Reese Markides MS, RD2, Lesli Biediger-Friedman PhD, MPH, RD3, Amanda Reat MS, RD2, Nicholas Bishop PhD4
1Nutrition and Foods Professor, School of Family and Consumer Sciences, Texas State University, San Marcos, Texas; 2Nutrition and Foods Lecturer, School of Family and Consumer Sciences, Texas State University, San Marcos, Texas; 3Nutrition and Foods Assistant Professor, School of Family and Consumer Sciences, Texas State University, San Marcos, Texas; 4Family and Child Development Assistant Professor, School of Family and Consumer Sciences, Texas State University, San Marcos, Texas
Corresponding Author: firstname.lastname@example.org
Journal MTM 7:2:7–16, 2018
Background: Smartphone apps have potential to effectively deliver health education and improve health behaviors among at-risk populations. To be successful, apps should include user input during stages of development. Previously, a prototype app designed for participants in the Texas Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) was developed based on input from focus groups.
Aims: This research aimed to continue app design by soliciting user input via a survey from a state-wide sample.
Methods: Texas WIC clients were asked about physical activity, healthy eating, and breastfeeding behaviors, stage of change regarding health behaviors, current use of health-related apps, and perceptions of app prototype features. Latent class analysis (n=942) was used to identify mutually exclusive groups based on the strength of participants’ agreement that prototype features would help them exercise more or consume more fruits and vegetables. Logistic regression examined health-related characteristics and sociodemographic differences between classes.
Results: Response to app prototype features was positive. A 2-class model best described latent classes. Class members that strongly agreed that prototype features would help them improve health behaviors were younger (< 35 years), not pregnant, already using health-related apps, and in the contemplation, preparation, or action stages of change regarding physical activity.
Conclusion: Refinement of the Texas WIC app should incorporate input from individuals who are pregnant, older than 35 years, or in pre-contemplation regarding physical activity. The iterative process of user-centered design applied in this research may serve as a useful framework for development of other public health apps.
Keywords: health promotion, technology, vegetables, smartphone, exercise
In the United States, poverty affects women and children disproportionately, as they make up approximately 70% of the low-income population.1 Poverty is associated with deleterious health behaviors, such as consuming a low-quality diet and being physically inactive, particularly among vulnerable populations such as women and children.2 These behaviors contribute to serious health concerns, including poor birth outcomes, obesity, heart disease, type 2 diabetes, and certain cancers.2,3 Limited access to evidence-based information related to health, physical activity, nutrition, and infant care is a likely contributor to poor health behaviors and outcomes among low-income individuals and may be an important barrier that contributes to ongoing health disparities.2
Launched in 1972, the Special Supplemental Nutrition Program for Women, Infants, and Children (WIC) is a federal grant program that serves low-income pregnant, postpartum, and breastfeeding women, infants, and children up to age 5 who are at nutritional risk, with the goal of improving health behaviors and outcomes during critical periods of development.4 Annually, 8 million women, infants, and children are enrolled in WIC, with approximately 886 thousand participating in the state of Texas.5 WIC provides a number of resources, including vouchers for healthful foods to support pregnancy, lactation, and growth, and referrals to health care services.6 While they are enrolled in the program, WIC clients are expected to regularly participate in education that focuses on promoting healthy behaviors such as breastfeeding, exercise, and healthy eating (e.g. eating fruits and vegetables, cooking meals at home, eating meals as a family).6 Historically, WIC clinics have worked to impart evidence-based information related to health behaviors through education offered at clinics via face-to-face education. This education modality presents barriers to an already taxed population, which may lack reliable transportation to clinics and childcare during education sessions.7 In an attempt to mitigate these barriers, many WIC state agencies now offer online client education.8 However, reliable access to a computer with internet connectivity is not ubiquitous among Americans, and low-income smartphone owners are more likely to rely on their smartphones as a primary way of connecting to the internet.9 Indeed, Texas WIC clients have expressed a desire to receive education and services delivered via phone.7,10 Thus, smartphone apps may offer a viable alternative interface for providing innovative, accessible, and customizable health education to the WIC population. Research has supported the use of smartphones as a behavioral modification tool, with a number of applications developed to improve diet and physical activity.11
Smartphone apps have unique characteristics that may make them particularly ideal for delivering health education and supporting health behavior change. For example, smartphone apps can be used to access clients in real time, offer continual assessment of identified treatment goals, and deliver meaningful information and support to reinforce behavior change.11 Despite the promise of smartphone apps, individuals tend to discontinue app use after three months of downloading.12 Therefore, app developers should consider, a priori, the expressed needs of intended users. Indeed, all approaches to developing technological tools to improve health outcomes should engage people first.13 One approach to developing apps that prioritizes individuals is user-centered design (UCD), an evidence-based, iterative process prioritizing user input and engagement in designing products and services.14 Because UCD has been previously used to develop appealing smartphone apps that target health behaviors, such as physical activity,15 this process shows promise for developing an effective app for WIC clients. In 2014, based on strong interest among Texas WIC clients for delivery of nutrition education and services via their smartphones,16 the Texas Department of State Health Services WIC program commissioned us to develop a smartphone app prototype. To do this, we began the UCD process by conducting focus groups with a diverse sample of female WIC participants in south central Texas to explore current smartphone app use and preferences.17 Based on this initial user-input and tenets of the Social Cognitive Theory,18 we developed an app prototype, designed to provide customizable, interactive, and user-centered health education to the Texas WIC population.17 The prototype included features to support physical activity (i.e., activity calendar, activity tracker, exercise videos, resource library), healthy eating (i.e., meal calendar, healthy eating tracker, cooking videos, resource library, shopping list, fruit and vegetable game, farmer’s market locator), and breastfeeding (i.e., breastfeeding timer, growth chart, live assistant, resource library).17
The aim of this research was to continue the UCD process of developing an app for Texas WIC clients by disseminating a statewide survey seeking input regarding the app prototype features. Analysis of clients’ perceptions of prototype features designed to support physical activity and healthy eating are included in this report. Our approach was to use latent class analysis to identify subgroups of respondents based on the extent to which they agreed that the features would help them increase physical activity and intake of fruits and vegetables, and logistic regression to examine how membership in the latent classes were associated with sociodemographic and health-related characteristics. Recommendations for continued user-centered design of the WIC app were informed by characteristics of respondents in latent classes.
The survey was posted on the Texas WIC website from September 9, 2014 through November 6, 2014. Clients visiting the website were greeted with a pop-up window presenting an offer to take the survey in English or Spanish. Additionally, clinics in central Texas who had access to client email addresses sent invitations to clients to take the survey. Of the 606 emails sent, 88 addresses were invalid, 102 began taking the survey, and 63 finished. Overall, 1,019 WIC clients completed the survey. Participants who were younger than 18 (n=50), male (n=14), and had an implausible reported height (shorter than 4 feet or taller than 7 feet, n=27)19 were removed from the analytic sample, leaving a total sample of 942 respondents. The Institutional Review Boards of Texas State University and the Texas Department of State Health Services approved this study.
The survey, developed in English in collaboration with Texas WIC staff, included approximately 130 questions, depending on responses to logic-driven branches. To develop a version of the client survey in Spanish, the English survey was translated to Spanish, back-translated, and discrepancies were reconciled. The survey was implemented using Qualtrics software (2014, Provo, UT). The welcome page briefly described the survey, provided assurances of privacy, and described incentives for survey completion, which included credit for taking a WIC nutrition class and receiving a t-shirt. After giving informed consent, participants were asked if they owned a smartphone. Those who responded with ‘no’ were routed to a thank you page and the survey was discontinued.
The survey was divided into 3 major sections corresponding to health behaviors addressed by the WIC app prototype, including physical activity, healthy eating, and breastfeeding, followed by a set of demographics questions. Each health behavior section asked about current health practices, stage of change, facilitators and barriers to performing the health behavior, belief that app features would help to improve health behaviors, and current use of apps regarding that health behavior. Facilitators and barriers to health behaviors were drawn from focus groups held during the initial phase of the UCD of this prototype app.17 The current study is an analysis of participant response to the physical activity and healthy eating features and does not include breastfeeding.
Current practices regarding physical activity were measured using the Godin leisure-time exercise questionnaire, which creates a physical activity score based on questions about intensity and duration of exercise.20 This score classifies participant activity as insufficiently active, moderately active, or active. For analysis, categories were collapsed into a binary variable (0 = insufficient activity, 1 = moderately active or active). Stage of change for physical activity behaviors was assessed using a 4-question system adapted from Wolf et al. (1 = pre-contemplation, 2 = contemplation, 3 = preparation, 4 = action).21 Survey respondents were asked to indicate on a 5-item Likert scale to what extent they agreed that specific barriers and facilitators to physical activity applied to them personally (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree); these data were not included in this analysis. Finally, after each of the four app prototype features addressing exercise was displayed (an activity calendar, activity tracker, exercise videos, and resource library), clients were asked whether they agreed that the feature would help them exercise more often (1 = strongly disagree, 2 = disagree, 3 = neither agree nor disagree, 4 = agree, 5 = strongly agree).
Intake of fruits and vegetables was used as an indicator of healthy eating practices and was measured using a brief screener employed by Wolf et al. (fruit and vegetable servings consumed on the previous day, excluding servings of white potatoes, were summed for the final count and included in analysis as a continuous variable).21 Participants were also asked how many family meals they had each week, which was included as a continuous variable. Assessments of stage of change and whether the seven prototype features (a meal calendar, healthy eating tracker, cooking videos, a resource library, shopping list, fruit and vegetable game, and a farmer’s market locator) that addressed healthy eating would help them eat more fruits and vegetables were conducted in the same manner as described for physical activity.
The survey included questions about demographic characteristics. Household size was measured as a continuous variable. Age of participants (0 = 35 years or older, 1 = younger than 35 years), education (0 = high school or less, 1 = post-secondary education), race/ethnicity (White = reference, Black, Hispanic, Other), language used to complete the survey (0 = English, 1 = Spanish), employment status (0 = unemployed, 1 = employed), location of residence (0 = urban, 1 = rural), and body mass index (BMI; < 18.5 = underweight, 18.5 – 24.9 = normal weight, 25 – 29.9 = overweight, > 30 = obese) were coded as categorical variables. Due to few participants having a BMI identifying them as underweight, participants identified as underweight and normal weight were combined and used as the BMI reference group. BMI was not calculated for women who were pregnant (1 = pregnant, 0 = not pregnant). Food security was assessed with the U.S. Household Food Security Survey Module: Six-Item Short Form.22 Food security status was coded as a dichotomous variable (0 = very low or low food security, 1 = marginal or high food security). Current use of physical activity or healthy eating apps were coded as dichotomous variables (0 = never or almost never use, 1 = sometimes or daily use).
Latent class analysis, a form of mixture modeling allowing for the classification of unobserved heterogeneity in responses to multiple variables, was used to identify homogenous, mutually exclusive groups of WIC clients based on the extent to which they agreed that prototype features would help them exercise more often or eat more fruits and vegetables.23 To identify the number of classes that best represented the underlying response groups, a series of model fit tests were conducted starting with a single-class model. Model fit indices, including Akaike Information Criterion (AIC), Bayesian Information Criterion (BIC), and sample-size adjusted AIC (SSA-AIC), were used to determine whether the inclusion of each additional class provided improved model fit. Model entropy, representing the accuracy of assigning individuals to classes, was also considered. Finally, the Vuong-Lo-Mendell-Ruben (VLMR) likelihood ratio provided a statistical test of whether the estimated model significantly improved model fit compared to a model with one less class. Once the optimal number of latent classes was determined, logistic regression was used to examine differences between classes based on sociodemographic and health-related characteristics. The latent class analysis and logistic regression was conducted with Mplus 7.324 using maximum likelihood estimation with robust standard errors, providing treatment of missing data with maximum likelihood and estimation of standard errors robust to non-normality. All other analyses were conducted using IBM SPSS Statistics for Windows, version 24 (IBM Corp., Armonk, N.Y., USA).
Table 1 presents model fit indices used to identify the optimal number of latent classes based on WIC participants’ responses to the survey questions “this app feature would help me exercise more/eat more fruits and vegetables.” Compared to the 1-class model, the 2-class model had improved AIC, BIC, and SSA-BIC model fit indices; the rate of model fit improvement decreased with the 3-class model. The VLMR test indicated that the 2-class model was a significant improvement on the 1-class model (p < .001), but the 3-class model was not a significantly better fit than the 2-class model (p = 0.8). Thus, based on model fit tests and the necessity of parsimony, the 2-class model was identified as the best description of latent classes.
Table 1: Goodness of fit indices for determining number of latent classes among Texas WIC survey respondents.
Class 1 (strongly agree; 32.9% of the sample) was identified as the group that strongly agreed that app features would help them improve targeted health behaviors; those in class 2 (neutral, agree; 64.8% of the sample) agreed or were neutral regarding whether the app features would help improve health behaviors. Figure 1 shows the distribution of responses to questions asking if using the app features would help respondents improve targeted health behaviors.
Figure 1: WIC clients’ agreement that app features would help improve targeted health behaviors. a) Depicts Class 1 (strongly agree) and Class 2 (neutral, agree) feedback regarding physical activity features. b) Depicts Class 1 (strongly agree) and Class 2 (neutral, agree) feedback regarding healthy eating features.
Table 2 includes descriptive statistics for the complete analytic sample as well as by latent class. On average, respondents in the complete sample had approximately 5 household members and consumed 7.4 family meals per week, including 3.5 servings of fruits and vegetables per day. Approximately three out of four respondents were younger than 35 years of age at the time of the survey, which categorizes them as millenials.25 Slightly more than half of participants were Hispanic and the vast majority took the survey in English. Approximately 45% of respondents were employed and the majority were urban-dwellers. Fifty-eight percent were overweight or obese. Sixteen percent of the sample was pregnant. Approximately a third of the sample had completed post-secondary education and a third had marginal or high food security. Two-thirds engaged in at least 150 minutes of moderate or intense physical activity each week. Almost two-thirds of participants used apps for exercise and three-quarters used healthy eating apps. The vast majority recognized the benefits of being physically active and eating fruits and vegetables. Likewise, the majority were in contemplation, preparation, or active stages of change regarding being physically active and eating fruits and vegetables.
Table 2: Description of the overall sample of Texas WIC clients and the 2 latent classes.
The results of the multinomial logistic regression are shown in Table 3; class 2 (neutral, agree) was used as the reference group. Age, pregnancy, current app use, and stage of change regarding exercise were significant predictors of class membership. Specifically, respondents were more likely to be in class 1 (strongly agree) if they were younger than 35 years old (OR = 1.40; CI = 1.05, 1.86), were in contemplation, preparation, or active stages of change regarding exercise (OR = 2.28, CI = 1.26, 4.11), or were currently using apps for exercise (OR = 1.33; CI = 1.02, 1.78) or healthy eating (OR = 1.72, CI = 1.22, 2.42). Respondents were significantly less likely to be in class 1 (strongly agree) if they were pregnant (OR = 0.55, CI = 0.37, 0.82).
Table 3: Predictors of Texas WIC clients’ perceptions of WIC app prototype features.
This paper describes an intermediate stage of UCD of an app designed for Texas WIC participants. By investigating variation in responses to the app prototype features, our aim was to identify characteristics of survey respondents associated with the strength of their agreement that the physical activity and healthy eating features would help them improve targeted health behaviors. Importantly, survey respondents’ reactions to the app prototype features were positive. Participants who strongly agreed that the app features would help support behavior changes were more likely to be younger than 35 years of age, in the contemplation, preparation, or action stages of change for the targeted health behaviors, and currently using apps to foster these health behaviors.
The influence of age on class membership is not unexpected. Indeed, millennials, or those born after 1980, are the age group most likely to be continually engaged with smartphones26 and use them for a variety of activities such as searching for jobs and accessing health information.27 One potential avenue for increasing the acceptability of this app among older WIC participants would be to link to or otherwise leverage platforms that have cross-generational appeal.28 In the context of the transtheoretical model, it is also not surprising that participants in contemplation, preparation, or action stages of change were more likely to view the proposed app features as supportive, as they were already interested in improving the targeted health behaviors.29 Similarly, individuals currently using apps are likely more receptive to prototype app features, in general. A surprising finding was that pregnancy was a significant predictor of class membership, with pregnant women being half as likely to be in class 1 (strongly agree). Previous research has reported that, in general, pregnant women are interested in health-related apps with features that are provide information specific to pregnancy, such as pregnancy-related risk factors, gestational weight gain, diet and lifestyle, postpartum depression, social support, and early infant feeding.30–32 In light of this, one explanation for the relatively tepid response of pregnant women in this study could be that the features were not specific to dietary and exercise recommendations for pregnancy. Additionally, mobile health interventions targeting pregnant women often suffer from low enrollment and high attrition, suggesting that, in general, pregnancy may be a challenging time to address health behavior change.33 However, firm conclusions about the allure of mobile apps to address health behaviors during pregnancy cannot be drawn due to a dearth of relevant studies.33 Given that health behaviors during pregnancy can have a profound impact on maternal and child health, it is important for an app designed for WIC clients to specifically address the needs of pregnant women to support a healthful pregnancy.
A strength of this study was the use of a large sample of Texas WIC clients who have experience with technology. Sample demographics were somewhat comparable to Texas WIC, with 56% of the sample being Hispanic, compared to 68% in Texas WIC,34 and 58% being overweight or obese, compared to 52% in Texas WIC.35 Limitations include electronic recruitment of individuals who owned a smartphone, resulting in a sample biased towards technology use.
Given the enthusiastic response to the Texas WIC app prototype, it would be tempting to finalize app development by simply incorporating the features described in this study. However, while many factors may impact the ultimate success of public health apps, perhaps a central piece revolves around the needs and preferences of the intended user, framed within the context of his or her specific life’s challenges.13 A user-centered approach to developing technology-based tools, such as the UCD process, is a critical step in ensuring that public health interventions reach their target audiences and elicit desired health outcomes.14 Given the health status of vulnerable populations, such as low income women and children participating in WIC,2 developing and implementing efficacious, evidence-based, population-specific tools and technologies that meet the needs of participants is of paramount importance, and may provide a catalyst to improve health equity by removing barriers to accessing information. In the case of the development of a Texas WIC app, engaging clients who are older, in a pre-contemplation stage of change, or pregnant should occur next, so that their specific needs and preferences can be incorporated into the final version. The iterative process of UCD used in this research may serve as a useful framework for development of public health apps.
This research was funded by a grant from the Texas Department of State Health Services (contract number 2014-045584). WIC staff provided input on prototype feature design,17 reviewed survey content, and facilitated survey dissemination via their website.
The Institutional Review Boards of Texas State University (2013E3835) and the Texas Department of State Health Services (14-014) approved this study.
Declaration of Competing Interests
All authors have completed the Unified Competing Interest form at www.icmje.org/coi_disclosure.pdf (available on request from the corresponding author) and declare: all authors had financial support from the Texas Department of State Health Services for the submitted work; no financial relationships with any organisations that might have an interest in the submitted work in the previous 3 years; no other relationships or activities that could appear to have influenced the submitted work.
1. Legal Momentum. Women and Poverty in America. Available from: https://www.legalmomentum.org/women-and-poverty-america. Accessed December 2016.
2. Adler NE, Stewart J. Health disparities across the lifespan: Meaning, methods, and mechanisms. Ann N Y Acad Sci. 2010;1186:5-23.
3. Eidelman AI, Schanler RJ, Johnston M, et al. Breastfeeding and the use of human milk. Pediatrics. 2012;129(3):e827-e841.
4. Oliveira V, Racine E, Olmsted J, Ghelfi LM. The WIC Program: Background, trends and issues (No. 33847). United States Department of Agriculture, Economic Research Service; 2002.
5. United States Department of Agriculture Food and Nutrition Service. WIC Program: Total Participation. Available from: https://www.fns.usda.gov/sites/default/files/pd/26wifypart.pdf. Accessed July 2017.
6. Special Supplemental Nutrition Program for Women, Infants, and Children. 7 C.F.R. §246.11 1985. Available from: http://www.ecfr.gov/cgi-bin/text-idx?SID=a42889f84f99d56ec18d77c9b463c613&node=7:184.108.40.206.10&rgn=div5#se7.4.246_111. Accessed July 2017.
7. Greenblatt Y, Gomez S, Alleman G, et al. Optimizing nutrition education in WIC: Findings from focus groups with Arizona clients and staff. J Nutr Educ Behav. 2016;48(4):289-294.
8. Cates S, Capogrossi K, Sallack L. WIC Nutrition Education Study: Phase I Report. Alexandria, VA: United States Department of Agriculture, Food and Nutrition Service, Office of Policy Support; 2016. Available from: https://fns-prod.azureedge.net/sites/default/files/ops/WICNutEd-PhaseI.pdf. Accessed June 2017.
9. McHenry G. Evolving technologies change the nature of Internet use. National Telecommunications and Information Administration. Available from: https://www.ntia.doc.gov/blog/2016/evolving-technologies-change-nature-internet-use. Accessed July 2017.
10. Deehy K, Hoger FS, Kallio J, et al. Participant-centered education: Building a new WIC nutrition education model. J Nutr Educ Behav. 2010;42(3S):S39-S46.
11. Schoeppe S, Alley S, Van Lippevelde W, et al. Efficacy of interventions that use apps to improve diet, physical activity and sedentary behaviour: A systematic review. Int J Behav Nutr Phys Act. 2016;13(1):127.
12. Localytics. Helping marketers better understand app user trends. Available from: http://www.localytics.com/resources/app-stickiness-index-q1-2015. Accessed July 2017.
13. Barclay G, Sabina A, Graham G. Population health and technology: Placing people first. Am J Public Health. 2014;104(12):2246-2247.
14. McCurdie T, Taneva S, Casselman M, et al. mHealth Consumer apps: The case for user-centered design. Biomed Instrum Technol Mob Heal. 2012;Suppl:49-56.
15. Vorrink SN, Kort HS, Troosters T, et al. A mobile phone app to stimulate daily physical activity in patients with chronic obstructive pulmonary disease: Development, feasibility, and pilot studies. JMIR mHealth uHealth. 2016;4(1):e11.
16. Texas WIC Nutrition Education Survey – Statewide Report. Texas Department of State Health Services; 2016. Available from: https://www.dshs.texas.gov/wichd/bf/surveysreports.aspx. Accessed July 2017.
17. Biediger-Friedman L, Crixell SH, Silva M, et al. User-centered design of a Texas WIC app: A focus group investigation. Am J Health Behav. 2016;40(4):461-71.
18. Bandura A. Self-efficacy: Toward a unifying theory of behavioral change. Adv Behav Res Ther. 1978;1(4):139-161.
19. Koebnick C, Smith N, Huang K, et al. The prevalence of obesity and obesity-related health conditions in a large multiethnic cohort of young adults in California. Ann Epidemiol. 2012;22(9):609-616.
20. Godin G, Shephard R. Godin leisure-time exercise questionnaire. Med Sci Sport Exerc. 1997;29(6):S36-S38.
21. Wolf RL, Lepore SJ, Vandergrift JL, et al. Knowledge, barriers, and stage of change as correlates of fruit and vegetable consumption among urban and mostly immigrant black men. J Am Diet Assoc. 2008;108(8):1315-1322.
22. United States Department of Agriculture Economic Research Service. Household Food Security Survey Module: Six-Item Short Form. Available from: https://www.ers.usda.gov/media/8282/short2012.pdf. Accessed May 2014.
23. Hagenaars J, McCutcheon A, eds. Applied Latent Class Analysis. Cambridge, UK: Cambridge University Press; 2002.
24. Muthén L, Muthén B. Mplus User’s Guide. Vol 7th ed. Los Angeles, CA: Muthén & Muthén; 2015.
25. Experian. Millennials and Technology: The Natural Order of Things. Available from: http://www.experian.com/blogs/marketing-forward/2014/07/07/millennials-and-technology-the-natural-order-of-things. Accessed July 6, 2017.
26. Nielsen. Millennials are Top Smartphone Users. Available from: http://www.nielsen.com/us/en/insights/news/2016/millennials-are-top-smartphone-users.html. Accessed July 2017.
27. Pew Research Center. U.S. Smartphone Use in 2015. Available from: http://www.pewinternet.org/2015/04/01/us-smartphone-use-in-2015/. Accessed June 2015.
28. Pew Research Center. Social Media Update 2016. Available from: http://www.pewinternet.org/2016/11/11/social-media-update-2016/. Accessed May 2018.
29. Prochaska JO, Velicer WF. The transtheoretical model of health behavior change. Am J Heal Promot. 1997;12(1).
30. Osma J, Barrera AZ, Ramphos E. Are pregnant and postpartum women interested in health-related apps? Implications for the prevention of perinatal depression. Cyberpsychology, Behav Soc Netw. 2016;19(6):412-415.
31. Krishnamurti T, Davis A, Wong-Parodi G, et al. Development and testing of the MyHealthyPregnancy App: A behavioral decision research-based tool for assessing and communicating pregnancy risk. JMIR Mhealth Uhealth. 2017;5(4):e42.
32. Waring ME, Moore Simas TA, Xiao RS, et al. Pregnant women’s interest in a website or mobile application for healthy gestational weight gain. Sex Reprod Healthc. 2014;5(4):182-184.
33. O’Brien CM, Cramp C, Dodd JM. Delivery of dietary and lifestyle interventions in pregnancy: Is it time to promote the use of electronic and mobile health technologies? Semin Reprod Med. 2016;34(2):e22-e27.
34. United States Department of Agriculture Food and Nutrition Services. WIC Racial-Ethnic Group Enrollment Data 2012. Available from: https://www.fns.usda.gov/wic/wic-racial-ethnic-group-enrollment-data-2012. Accessed July 2017.
35. Johnson B, Thorn B, McGill B, et al. WIC Participant and Program Characteristics 2012. Alexandria, VA: United States Department of Agriculture, Food and Nutrition Service; 2013. Available at https://www.fns.usda.gov/wic/women-infants-and-children-wic-participant-and-program-characteristics-2012. Accessed June 2017.
Evaluation of Free Android Healthcare Apps Listed in appsanitarie.it Database: Technical Analysis, Survey Results and Suggestions for Developers
Dr. Lorenzo Di Matteo1, Dr. Carmela Pierri2, M.D. Sergio Pillon3, Eng. Giampiero Gasperini4, Eng. Paolo Preite1, Dr. Edoardo Limone4, Dr. Silvia Rongoni1
1Department of Training, Formit Foundation; 2Board of Directors, UNINT University/ Department of Training, Formit Foundation; 3Department of Cardiovascular Telemedicine, Azienda Ospedaliera San Camillo-Forlanini; 4Department of Strategy and Technologies, Formit Foundation
Corresponding Author: email@example.com
Journal MTM 7:2:17–26, 2018
Background: Health apps catalogued in dedicated databases are not scarce but still little is known about the situation concerning their technical aspects such as the general level of privacy and security.
Aims: This study aims to analyze android free health apps in a specific database.
Methods: A systematic technical analysis on a population of 275 android free app among the ones listed in the appsanitarie.it database (“Banca Dati delle app sanitarie”). Analysis has been carried out following a defined protocol with a survey as operative support tool to examine aspects such as the app rating in the store.
Results: The analysis concerned 275 health apps. Cardiology (38 apps) resulted to be the most populous medical branch. The overall app ratings average is 4,10. 18,54% of the apps required personal data at first launch. 84,36% of the apps allowed only manual data entry. Data sharing has been detected in 133 cases. 9,45% of the apps provides a backup option. 13% of the apps declare to be compliant to some kind of privacy regulation. Among this 13% of apps only 19% showed relevance to the EU privacy regulation. The 61,1% of the apps presented no reference for scientific background of the contents.
Conclusions: Manual data entry when redundant should be avoided by developers in favour of automatic calculation of derived parameters. Moreover a limited number of the analyzed apps adopt data protection mechanisms and declare privacy compliance. Security and Privacy are generally poor. Survey results suggest there is large room for improvement in app design.
Keywords: Telemedicine, eHealth, mHealth, Data Security, Baseline Survey
Apps on mobile devices such as smartphone offer a lot of perspectives of use in health and medical fields. App economy as the whole range of economic activity related to mobile applications evolve rapidly as the smartphone market. Other studies report that only the first 10 top mobile health apps generate up to 4 million free and 300.000 paid downloads per day1.
On the other side Healthcare researches find that vast majority of professionals is conscious of an interoperability lack for a better use of patient generated data2. Other researches show that more than half of the interviewed patients assert to have used a digital device including mobile apps to manage their health and almost two thirds think it would be helpful for their healthcare providers to have access to their patient generated data as part of their medical history3.
Studies showed that for patients with chronic diseases it is a comfortable solution sharing data with healthcare providers via online patient portal, mobile apps or message texts4. This could lead to some sort of benefits for both patients and healthcare providers but also expose to some risks, especially the first ones5,6. Unclear disclosures about data processing terms could lead to privacy risks for the user and insufficient security could bring to data breaches or loss risks, considering also that a smartphone loss could bring to a leakage7. Security or data protection could be not sufficient if the user is not fully capable to prevent the loss of data from the device or mechanisms as encryption or passwords are not available8.
On the other hand sharing patients health data with messaging and multimedia mobile applications as communication channels it’s handy for a professional but non completely compliant with health data protection standards a healthcare trust certainly adopt9. On the patient side new findings concluded that while less than half of the analyzed apps are useful to the targeted user, some apps seemed to sacrifice quality and safety to add more functionalities10.
The purpose of this study is to make a technical analysis of free android apps listed in a dedicated “healthcare apps” database, “Banca Dati delle app sanitarie” (at http://www.appsanitarie.it/banca-dati-app-sanitarie). The database has been developed as part of a Formit Foundation project financed by a grant of the General Directorate of Medical Devices and Pharmaceutical Service of the Italian Ministry of Health in 2015-2016. Launched in 2015 the database was created to list results of the apps census operated by the Observatory of the health apps established by Formit Foundation.
Apps in the database has been selected through a specific definition, “healthcare apps”, and selection workflow (see methods section for a full description). Database apps, both Android and iOS, have been selected through specific criteria in the stores (summoned in a workflow), and could be used in an healthcare context by patients and physicians. The database apps considered are 659 “healthcare apps”, divided in medical branches and 2% of them present a CE mark as medical device. The database has been chosen as starting point for the selection of listed apps because of a clear definition and a selection workflow.
The study which results will be here presented has not been conducted looking in the inner working mechanisms of the apps but with a highly technical analysis of the functionalities available to users. Analysis has been carried out facing four different groups of app characteristic: the app general details; the features as requested data, data entry, data access, connect-ability, online and sharing feature; password and backup security mechanisms; privacy terms and scientific references. Regulatory framework considered in matter of privacy is the General Data Protection Regulation, GDPR (Regulation EU 2016/679)11 due to its validity all over national member states legislations, and the Privacy Code of Conduct on mHealth apps for what concerns guidelines to enhance privacy in this field12.
This research project has been conducted with full compliance of research ethics norms. Research involved usage of mobile devices and apps. Survey development and data gathering involved part of the research team while survey fulfillment another one. Results analysis has been carried out by the whole team.
Apps has been selected among the list of free Android ones (Google play downloadable) in every medical branch composing the database (Banca Dati App sanitarie, BDA http://www.appsanitarie.it/banca-dati-app-sanitarie). It was also decided to exclude from the analysis apps requiring registration with medical credentials to dedicated platform or specific devices to work. Apps listed in the database has been chosen before this study following a specific definition and a workflow. In this sense not all the health apps could be listed in the database.
Apps defined as healthcare apps (App sanitarie) in the database are:
• CE marked Medical device apps (in order to achieve CE mark for their products in Europe, medical device manufacturers must comply with the appropriate medical device directive set forth by the EU Commission);
• Apps not developed with medical purpose by the producer but responding to one of this characteristics:
– receive data from medical devices;
– elaboration and transformation of healthcare and patient-related data;
– interaction with a non medical device that visualize, memorize, analyze and transmit data;
– receive health data by user with manual entry that are not only diet and fitness oriented.
According to the app definition this workflow was used:
Figure 1: App Selection Criteria
The apps selected from the database to be analyzed satisfy the following operative criteria:
– Available for Android (downloadable from Google play);
– With no mandatory registration to platform requiring medical credentials;
– Usable independently from connection with external devices.
Apps have been under a phase of technical analysis for 2 months, from March to April 2017.
The scope of the technical analysis is to examine some of the operating mechanisms of the selected apps. This has been done following a Technical Analysis Scheme characterized by different technical macro-area to identify diverse functional aspects and a metrical-statistical question-answer structure to ensure results measurability and repeatability.
To reach a technical analysis of the software, a survey has been designed and fulfilled. The analysis has been conceived to focus on the following elements:
- Information useful to identify the app;
- Operating characteristics of the app;
- Security related to password, back-up and data encryption;
- Presence of privacy and condition terms.
The technical analysis has been composed by the survey development, comprehensive of design and deployment, and a consequent phase of app analysis, then data gathering and results analysis.
To analyze selected apps a survey has been designed with different sections related to different type of data to collect about the four analytics aspects and organized following an answer-question structure. In this sense the sections which composed the survey are:
- App general characteristics, as name, version, developer name, rating on the store;
- App features, as requested personal data, modality of data entry, possibility to delete/change data, connect-ability, online platform registration, sharing on social media;
- App security, as password registration, password recovery, password security level, back-up possibility, backup destination, backup encryption;
- App privacy and reliability, as declaration of compliance to some privacy regulation, European privacy regulation compliance, scientific source or bibliography.
The online survey has been realized with the open source application Lime Survey.
Technical-functional analysis has been performed accessing the survey through authentication via username and password. Mobile devices with Android operative system have been used with the newest version of operative system available at the time. App search has been performed on the Google Play store. Download has followed if for free and if available in the country where the study took place (Italy). Once installation terminated, mandatory healthcare professional-only platform registration and necessary external device connection has been checked. If negative, the app has been analyzed through the survey fulfillment.
Data gathering and Results analysis
Through Lime survey data gathering has furnished the overall amount of data from the technical analysis. Results analysis instead has been realized on a compiled single dataset, using descriptive statistics to summarize and underline aspects of data collection. Collected data analysis has been accomplished through the data stored in a database managed directly by the application Lime survey, this allowed to export information in formats suitable for statistical work purposes.
App general characteristics
The analysis concerned 275 apps on the total amount of 659 in the database at the time the study took place, due to the existence of operative criteria described in the methods section. Most populous medical branch resulted cardiology (38 apps), oncology (22) and health & well-being (21), as shown in Figure 2.
Figure 2: Number of apps per medical branch
App rating in the store is expressed on a Likert scale from 1 to 5 by the user and shows the average of the overall amount of rating for an app on an incremental scale. Rating average of an app has been rounded down due to simplify data collection management. The average of the overall app rating averages resulted 4,10, where the lowest app rating average is 1 and the higher is 5. In the most populous medical branches, average of the app rating averages in cardiology is 3,66, while in oncology is 4,33 and in health & well-being is 4,60.
As Figure 3 shows average app rating in the store is high almost for every medical branch in line with the data of an average of the app rating averages of 4,10 on the Likert scale. In fact most of the analyzed apps resulted to be placed in the high ranks of the rating scale. Excluding 5 apps with no rating on the store, 105 on 270 apps, the 38,89% of the overall rated apps resulted having a rating average of 4 while 91 apps, the 33,70% is ranked with an average of 4,5 and 34 apps, the 12,60% showed a rating average of 5. However rating in the store could be subjected to distortive mechanisms such as comments directly or indirectly linked to the developers or an exiguous amount of them.
Figure 3: Average app rating per medical branch
Table 1: Rating average per number of apps
Generally health-ish apps need data input to perform one or more of their features. In this sense 18,54% of the analyzed apps showed to require personal data at their first launch in order to create a user profile. An app can request to the user one or more of the data listed in Table 2. The most frequently required data resulted to be gender for the 14,91% of the apps (41), followed by age the 11,62% (32) and weight the 8,62% (24).
Table 2: Personal Data required at first launch per number of apps
Table 3: Modality of data entry per number of apps
Modality of data entry followed the part of the survey section concerning personal data request. Data entry could happen through a possible synchronization with an external device in order to acquire data automatically, or at the contrary only manually or both. The great majority of the apps allowed only manual data entry, exactly 84,36% (232) of the apps. Only automatic and both data entry modality are allowed by the 8% and the 7,64% of the apps.
Similarly results about possibility to change and delete entered data showed that it was possible manual change for the 84% of the apps and manual deletion for the 72,72%. It has been noticed that it was not possible change manual entered data only for 18 apps and no possibility to delete for 47. Only 5,45% of the analyzed apps resulted to allow the modification of automatic entered data and 6,18% the deletion. In this sense results of N/A change and delete of automatic entered data and the possibility to change and delete manual entered data or vice versa almost match.
Table 4: Communication protocols per number of apps
Table 5: Communication protocols per number of apps
In line with the previous survey results, regarding communication protocols to exchange data with other systems or external medical devices, the most used resulted Wi-Fi for the 5,45% of the apps, followed by Bluetooth for the 5,1%. USB and a dedicated interface resulted to be used as communication protocols only by two of the analyzed apps.
In relation to communication and data exchange with online data storage services it was analyzed diffusion of mandatory registration to online platform in order to completely use the app. Generally apps requiring registration to an online platform permit backup of the data composing the user profile through a dedicated feature. In this sense it turned out to be only a 3,63% of the apps to require a mandatory registration to an online platform.
A considerable number of analyzed apps presented the feature “share” on different communication channels and social media. An app can allow more than one data sharing possibility. On the overall 133 times data sharing has been detected, the most frequent data sharing feature resulted to be e-mail (45 apps), almost doubling the second one that is SMS (23). Sharing on social media resulted to be possible only with 17 apps on Facebook and 14 on Twitter. Other channels not considered initially in the survey but of which it has been taken note in dedicated blank spaces, were hangouts resulting 12 times as social media and 10 times google drive as other sharing channel.
Table 6: Mandatory registration to online platforms per number of apps
Results regarding app security and data protection showed that only few of the analyzed apps provides password registration. The 5,1% of the apps shown to provide the creation of a password at the app start and only the 1.1% a secured password. On the overall amount of apps only 1,81% provides the password recovery generally known as “Forget Password?” button sending the new password to a previously saved email address.
For what concerns data storage option it resulted to be possible both locally, on the smartphone memory, that remotely with online storage services. Globally 9,45% of the overall analyzed apps provides a backup option. An online backup has been possible for the 5,1% of the apps, while a local memory back-up for the 4,36%. Regarding a clear-to-the-user encryption of the backup, 5,81% of the overall apps, more or less the all apps with backup option, showed no possibility to have clear information about encryption. Naturally it has been not feasible to check for encryption for the vast majority of the apps (93,81%) having no back-up.
Figure 4: Data sharing channels per number of apps
Table 7: App security and data protection
App privacy and reliability
App analysis concerning privacy showed that only 13% of the all apps declare to be compliant to any kind of privacy regulation for what concerns personal and health data about the user. The 87% of the apps showed no declaration of compliance to any kind of privacy regulation with any kind of message to the user, nor at launch nor in the menu.
Among this 13% of apps showing a declaration of privacy regulation compliance only 19% showed some sort of relevance to the EU privacy regulation. This is due to the fact that national regulation of Member States has been considered in relation to a wider European privacy regulation. In fact before General Data Protection Regulation, Directive 95/46/CE has been adopted by data protection and privacy national acts. The 81% of the remaining apps showed instead an international declaration related to an End-User License Agreement (EULA) model or some other type of generic declaration. For what concerns reliability it has been considered the presence of references quoted in the app regarding scientific sources. Generally scientific references and quotes has been found in the info or in the bibliography section of the app menu. The 61,1% of the analyzed apps presented no reference or quote regarding the scientific background of the contents, while 38,18% presented a bibliography or quoted studies in a dedicated part of the menu and 1,81% resulted to be not applicable to this check.
Figure 5: Percentage of Apps Showing Declaration of Privacy Regulation Compliance
Figure 6: Percentage of apps showing EU privacy regulation Compliance among apps declaring privacy regulation compliance
Table 8: Scientific references in the app
It is expected that Bring-Your-Own-Device connectivity will be preferred by select patient groups and will be used for the remote monitoring of 22.9 million patients in 202113. In this sense it is no surprise the number of health apps in the stores like Google Play, although the number of apps wears thin using a database based on a specific definition of mHealth app with a clearly defined selection workflow. Another boundary for the analysis has been represented by the possibility of free download and analyze functionalities without restrictions of use by necessary external device to operate or mandatory registration to platform requiring medical credentials.
On the other hand the strict database selection criteria and subsequently the limits of analysis operative criteria have brought to a homogeneity of apps population and uniformity of characteristics to analyze. Concerning the definition of app on which the database is based certainly the app analysis selection has been made among a population of apps that excludes low quality apps from the analysis spectrum. In this sense it has to be explained the small numbers of analyzed apps for some specialized medical branches and limits of findings for apps in those branches.
Likert scale is an important score to observe regarding adoption of an app by users. Although this data could be subjected to some distortions and elaboration of new assessing tools doesn’t miss14–18. Usability is generally measured with the perceived ease and enjoyment experiencing in using the app19, but the user could rate poorly an application with solid data security mechanisms but no catchy layout and the opposite with a more attractive but less secure app. Therefore rating of the apps is certainly important albeit partial.
The request of personal data opens up other considerations. Processing personal data would suggest the adoption of a database encryption mechanism, but once “compiled” in the form of “application package” can no longer be opened and verified without violating the copyright of the developer. It is also true that both Apple and Google have enabled encryption of the application database in the default mode.
It should also be noted that requiring to the user both age and date of birth impacts on data quality management. Most of the apps allow only manual data entry. It is an important factor due to its possible repercussion on data quality. Essentially more it is reduced input error, more data quality will be achieved. However data entry could be manual due to a developer lack or to a functional condition to respect in order to let the app run.
Regarding automatic data entry some applications use communication protocols with other systems or external devices as Bluetooth, wi-fi or USB. Regarding data exchange, Android allows the developer to easily implement communication to social networks or instant messaging systems with the possibility of data sharing with other users. Mail choice to share data is no surprise, the versatility of the medium is certainly more suitable to send ordinary messages to a doctor. Similarly SMSs are used by apps to share data since many applications elaborate a set of numeric values that can be easily included into a text message.
In matter of data protection the use of a secure password with a minimum of eight letters and alpha-numeric symbols seems to be rare. In this sense a personal identification number of five digits can be easily forced by a brute force attack trying all the possible combinations. It’s worth to mention also that a strong password is a condition that could be set up by design.
The apps including a backup function resulted to be limited too, but when the app not requires “persistent” data to trace, backup is useless. More than half of the analyzed apps has backup on online services (a cost to the app provider that local memory backup isn’t). The user could be not adequately aware of the technical and legal mechanisms that regulate the cloud computing service20, local backup instead allows complete data management. No sufficient information about data protection controls has been noticed. This is because the focus in the app description on the store generally seems on advertising, while neglecting privacy and security reliability.
The great majority of the apps showed no declaration of compliance to any kind of privacy regulation. Calculator apps or similar apps resets at every exit deleting all the entered data with no real data processing. Anyway it doesn’t really explain the absence of scientific references and quotes in almost 6 on 10 apps. In fact only a few apps presented clear scientific references. For some apps, especially scale calculator a professional may not need scientific references to identify or use a well-known tool in his or her medical branch, but for a user with no particular knowledge in medical science the lack of information could lead to misreading the outcome and to false negative self-diagnosis.
Considering that the analysis has been carried out on a limited number of apps, data-quality oriented approach should be used anyway by developers in order to realize a correct balance between manual data entry and automatic calculation. Manual data-entry should be reduced and the automation should be increased. Moreover format-control parameters or different controls (as sliders, or date-picker) should be used to reduce data-entry mistakes. Replacement of the classic text-field produces an increasing of speed during the filling process and reduces typing errors. In this sense could be useful to look for a better understanding of the perceived and desired usability by the user, rising research attention on this side. On the other hand it is necessary to examine feasibility of mHealth in the healthcare context, as effectiveness of mobile phone applications in healthcare services. In this sense a pathway could be an observational studies by experimenters with patient or physicians adopting mHealth solutions.
We would like to thank the General Directorate for Medical Devices and the Pharmaceutical Service of the Ministry of Health (Italy), especially Dir. Marcella Marletta, Eng. Pietro Calamea and Dr. Paola D’Alessandro for their support.
1. Mobile Health Market Report 2013-2017, research2guidance, March 2013:7.
2. Compton-Phillips A, What Data Can Really Do for Health Care, Care Redesign, Nejm Catalyst, March 2017:6-7.
3. Patient Expectations of Medical Information Sharing & Personalized Healthcare, Trascend Insights, Survey Report, February 2017.
4. The 2017 Patient Engagement Perspectives Study, CDW Healthcare, 2017.
5. Mitigate Cyberattacks with HIPAA-Compliant Communications, HIPAA White Paper, 2017.
6. Mobile Malware Evolution 2016, Kaspersky Lab, March 2016.
7. “What A Difference A Year Makes”, Healthcare Breach Report, Bitglass, January 2016.
8. “Is Your Data Security Due For a Physical?”, Healthcare Breach Report, Bitglass, January 2015.
9. DeepMind Health Indipendent Review Panel Annual Report, July 2017:14.
10. K. Singh, K. Drouin, L. P. Newmark et al., Developing a Framework for Evaluating the Patient Engagement, Quality, and Safety of Mobile Health Applications, The Commonwealth Fund, February 2016.
11. Regulation (EU) 2016/679 of the European Parliament and of the Council of 27 April 2016 on the protection of natural persons with regard to the processing of personal data and on the free movement of such data, and repealing Directive 95/46/EC (referred also as General Data Protection Regulation) http://eur-lex.europa.eu/legal-content/GA/TXT/?uri=uriserv:OJ.L_.2016.119.01.0001.01.ENG.
12. Privacy Code of Conduct on mHealth apps, https://ec.europa.eu/digital-single-market/en/privacy-code-conduct-mobile-health-apps.
13. mHealth and Home Monitoring – 8th Edition, Berg Insight, February 2017.
14. Stoyanov, Stoyan R et al. “Mobile App Rating Scale: A New Tool for Assessing the Quality of Health Mobile Apps.” Ed. Gunther Eysenbach. JMIR mHealth and uHealth 3.1 (2015): e27. PMC. Web, 26 July 2017.
15. Stoyanov, Stoyan R et al. “Development and Validation of the User Version of the Mobile Application Rating Scale (uMARS).” Ed. Gunther Eysenbach. JMIR mHealth and uHealth 4.2 (2016): e72. PMC. Web. 27 July 2017.
16. Domnich, Alexander et al. “Development and Validation of the Italian Version of the Mobile Application Rating Scale and Its Generalisability to Apps Targeting Primary Prevention.” BMC Medical Informatics and Decision Making 16 (2016): 83. PMC. Web. 27 July 2017.
17. Bradway, Meghan et al. “mHealth Assessment: Conceptualization of a Global Framework.” Ed. Mircea Focsa. JMIR mHealth and uHealth 5.5 (2017): e60. PMC. Web. 27 July 2017.
18. Baptista, Shaira, Brian Oldenburg, and Adrienne O’Neil. “Response to ‘Development and Validation of the User Version of the Mobile Application Rating Scale (uMARS).’” Ed. Gunther Eysenbach. JMIR mHealth and uHealth 5.6 (2017): e16. PMC. Web. 27 July 2017.
19. J. Nielsen, Usability 101: Introduction to Usability, January 4, 2012, https://www.nngroup.com/articles/usability-101-introduction-to-usability/.
20. Griebel, Lena et al. “A Scoping Review of Cloud Computing in Healthcare.” BMC Medical Informatics and Decision Making 15 (2015): 17. PMC. Web. 27 July 2017: 13.
Use of Personal Devices in Healthcare: Guidelines From A Roundtable Discussion
Laura Vearrier, MD1, Kyle Rosenberger, M.Ed2, Valerie Weber, DMD, MA3
1Assistant clinical professor, Department of Emergency Medicine, Drexel University College of Medicine, Philadelphia, PA; 2Instructional Designer, Ohio University Heritage College of Osteopathic Medicine and Ohio University’s Office of Instructional Innovation, Athens, OH; 3Assistant clinical professor, Department of General Dentistry and Oral Medicine, University of Louisville School of Dentistry, Louisville, KY.
Note: The corresponding author is not a recipient of a research scholarship.
Journal MTM 7:2:27–34, 2018
Background: In recent years, smartphone use in professional settings has been increasing, particularly with physicians. There are benefits and drawbacks that result from this increase. Despite this, there is relatively limited peer-reviewed medical literature on the subject. Thus, suitable guidelines for smartphone use in the health care setting is needed.
Aims: This article present guidelines for professional conduct related to the use of personal devices, such as smartphones, in the healthcare setting.
Methods: These guidelines were developed through an interdisciplinary roundtable discussion at the 2016 Academy for Professionalism in Health Care Conference in Philadelphia, PA.
Results: As a result of the roundtable discussions, several guidelines were developed. First, healthcare providers should be trained on the danger of distractions caused by personal devices and how to minimize them in a clinical setting. Second, the use of smartphones for personal use should be limited to specified use areas; however, if they are present during a patient encounter, they should be set to a mode that eliminates or minimizes interruptions. Third, providers should seek permission from patients prior to integrating smartphones into the provider-patient relationship. Finally, smartphone photography, while being a potential tool to improve patient care, should be used with caution concerning patient autonomy and privacy.
Conclusion: The guidelines serve as a foundation from which professionalism with regard to personal device use can be further developed.
Keywords: Professionalism, Smartphone, Physicians, Photography, Delivery of Health Care, Clinical Practice, Telemedicine
In the last few years, the use of personal devices such as smartphones has been rapidly increasing and smartphone ownership is highest among young adults of higher income and education level.1–2 This trend is being mirrored in the healthcare setting. Nearly all physicians and nurses own smartphones.3–4 Physicians’ usage of smartphones for professional purposes has been steadily increasing from 68% in 2012 to 84% in 2015.5 The Boston Consulting Group and Telenor Group remarked that the “smartphone is the most popular technology among doctors since the stethoscope”.6 A survey study of nurses reported that more than half of nurses have used their smartphone instead of asking a colleague for information.4
There are benefits and drawbacks of providers utilizing their smartphones in the healthcare setting for personal and professional purposes. With computing power and Internet connectivity, personal devices give providers access to textbooks, journal articles, practice guidelines, clinical calculators, and medical applications. Smartphones are improving the efficiency and accuracy of communication. Physicians and nurses are using short messaging services (SMS) to communicate patient information and smartphones have been reported to increase the connectedness of medical trainees’ with their supervisors.7–9 Smartphones are also improving communication between providers and patients. The use of videos on personal devices has been reported to be an efficient and effective way to educate patients on their disease that resulted in increased medication compliance and physicians are using smartphones to monitor patients remotely.10,11 Drawbacks of such constant connectivity include a risk for distraction from patient care. Providers may be interrupted for less acute clinical issues in addition to personal calls, texts, emails, social media, and applications. Personal devices also create a physical barrier between the user and the rest of the world. This barrier translates into cognitive and psychological barriers, and patients are often unaware of the clinical benefits of smartphones.12,13
Despite the ubiquity of smartphones in healthcare, there is limited peer-reviewed medical literature on issues with respect to the professionalism of smartphone use in the healthcare setting. An Ovid Medline keyword search of “professionalism” and the intersection of any of the following: “smartphone”, “smart phone”, “cell phone”, “mobile phone”, “tablet” or “personal device”, yielded only seven results (search performed May, 2016). There is a need for guidance and education regarding professional conduct and personal device use in the healthcare setting. In a survey study of medical students, the majority reported insufficient education from either their medical school curriculum or their senior residents or attendings on appropriate/inappropriate use of mobile devices to communicate patient information and how to conduct themselves professionally with mobile technology.14
A roundtable workshop exploring issues of professionalism and smartphone use in the healthcare setting was held at The Academy for Professionalism in Health Care 2016 conference in Philadelphia, PA. Participants included physicians, nurses, medical students, dentists and academic researchers. Participant occupational settings included medical education, academic clinical practice, and private practice. Four case-based scenarios (Figure 1) were discussed in small focus groups and then presented to the entire workshop for further analysis. The results of these discussions were compiled into the following guidelines.
Figure 1: Case-based scenarios and questions
Distraction and Smartphone Use
Smartphone use may result in provider distraction or so-called “distracted doctoring” and increase the risk for patient care errors. Smartphone use in the healthcare setting has the potential to result in distraction in a manner similar to driving. Use of smartphones may involve cognitive, visual, and/or manual tasks that divert the provider’s attention from their patient care responsibilities. As such, frequently repeated activities or procedures may increase the risk that providers will engage in distracting secondary tasks, such as smartphone use. “Bring Your Own Device” (BYOD) policies that require providers to utilize their personal devices for professional purposes may increase the risk of distraction due to non-professional phone calls, text messages, and app notifications. Further research is needed in the area of distracted doctoring and smartphone use.
Healthcare providers commonly hold the misperception that utilizing smartphones for multi-tasking in the healthcare setting improves efficiency and patient care. Professional and personal smartphone utilization in the healthcare setting increases the number of interruptions and the amount of information received and processed by providers while engaging in patient care. While multi-tasking is a necessary skill, it should be minimized when possible. Attentional shifts while multi-tasking interrupt the cognitive processing of information, situational awareness, and may increase the likelihood of patient care errors. Interruptions may be minimized by use of “silent”, “airplane”, and “do not disturb” modes, particularly during important patient care activities. The amount of information received by providers may be controlled through specialized ring or text message tones and disabling of app notifications.
Healthcare providers should be educated on the dangers of “distracted doctoring”. Individuals may overestimate their ability to engage in smartphone use without significant distraction. Education on distraction, generally, and specifically smartphone-associated distraction should be implemented at the undergraduate and graduate medical education levels and in patient-safety-related continuing education. Simulation exercises allow participants to experience the detrimental effects of distraction and develop skills for dynamic prioritization of incoming information.
Appropriate smartphone use may decrease distractions and should be encouraged. Smartphones may be utilized to reduce distractions from handheld pagers and overhead paging systems. Use of calendar functions, alarms, and notes apps may be used to reduce the cognitive load of “to-do lists”. Alarms for medication administration or other time-sensitive tasks may improve timeliness of administration. Smartphone information resources and clinical calculators reduce the need to interrupt current tasks to find a computer. Policies to utilize smartphones to reduce distractions should be considered at the institutional level.
Smartphone Photography in the Healthcare Setting
Smartphone photography is an advantageous learning and communication tool; however, respect for the patient and the patient’s privacy must be paramount. Smartphone photography has the potential to capture disease conditions and procedures that may otherwise be difficult or impossible to record and which may be used in educational materials or the peer-reviewed literature and therefore improve patient care on a global level. Smartphone photography may be used to transfer information about patients (e.g. lesion, electrocardiogram) to other providers, improving the clinical decision-making process and therefore improve patient care on the individual level. Pitfalls of photography in the healthcare setting include capture of patients while they are vulnerable or when they are unable to fully consent. Patients may perceive an element of coercion when asked to be photographed even if no direct coercive statement is made. The individual right to privacy, respect, and autonomy are paramount. Respect for privacy and autonomy as it pertains to smartphone photography should be taught at the undergraduate, graduate and continuing education levels. When institutional photography equipment is available, that equipment should be used in lieu of a smartphone.
Consent should be obtained at the time of image capture for the photograph, the intended use and any transmission. Consent should not be obtained at time of admission or triage for later photography. Consent for photography at that time contains an element of implied coercion and is too abstract to be considered informed consent. Informed consent should include the elements of the body part to be photographed, the intended use, and any transmission of the photograph. When possible, written consent should be obtained. If written consent is not possible, verbal consent should be witnessed and documented.
Photographs should be obtained in such a way as to minimize or eliminate the amount of protected health information that is captured. Photographs of the face are typically not necessary. Patient identifiers such as name, date of birth, and medical record numbers should not be included in photographs. Tattoos, piercings, skin conditions, and other unique identifiers compromise patient confidentiality and should be included only with explicit consent and when capture of those elements is required.
Smartphone photography in the healthcare setting for personal or entertainment purposes is inappropriate and should be avoided. Such photography contains too much potential for abuse to be acceptable. Unintended consequences include the inadvertent capture of protected health information or other patient identifiers. The content of seemingly innocuous photographs in the healthcare setting have the potential to distress patients, family members and others.
Healthcare providers have a duty to intervene in situations involving inappropriate smartphone photography. When possible, inappropriate photography should be prevented. If such photography has already occurred, appropriate interventions may include education, deletion of the photograph, or report at the institutional or law enforcement level depending on the scenario. Healthcare institutions should have protocols in place for reporting inappropriate smartphone photography with a well-defined chain of command and protections against retribution including the ability to anonymously report. Patients may similarly take inappropriate smartphone photographs in the healthcare setting and providers should intervene in those situations as well.
Smartphone etiquette and perceptions
Use of smartphones for personal calls, texting and social media apps is to be avoided in patient care areas. Personal use of smartphones in patient care areas may convey an informality or lack or professionalism to patients, their families, and other staff. Even when providers are not actively engaged in patient care activities, personal smartphone use should be avoided. Patients may perceive that aspects of their care are being adversely affected by personal smartphone use such as wait time, face time with providers, or attention to their complaints. Personal smartphone use may be permitted in staff lounge areas, provided that it does not adversely affect patient care.
Smartphones should be set to “silent”, “airplane”, or “do not disturb” modes during patient encounters. Vibrate modes are frequently audible and should not be utilized. App notifications should be turned off using one of the above-mentioned modes. Smartphone interruptions during sensitive discussions may be particularly distressing to patients. It is encouraged to remind colleagues to put their smartphone into one of the above-mentioned modes prior to such a discussion. Most healthcare providers do not need to be immediately available to colleagues. In the event that a healthcare provider must be immediately available during a patient encounter, the possibility of interruption should be communicated to the patient at the outset. The use of a “do not disturb” mode that permits calls from only pre-identified emergency contacts may reduce the risk of interruption and is recommended.
Patient permission should be obtained for professional smartphone utilization during patient care activities. As discussed above, smartphones are a resource for healthcare professionals, allowing increased communication with other providers, interface with EMR, clinical calculators, and immediate access to information resources (e.g. pharmacopeia). The inherent portability of smartphones over other electronic devices makes them particularly useful during patient care activities. However, explicit patient permission should be obtained prior to their utilization during patient care activities. Permission introduces the device into the patient-provider relationship and serves the dual purpose of informing the patient that the device is being used to facilitate their care and to confirm that use of the device will not be unduly distressing to the patient. Patients should be encouraged to ask their physicians what they are using their devices for to facilitate communication regarding this practice.
Transparency during professional smartphone use minimizes negative patient perceptions associated with provider smartphone utilization and is encouraged. Patients should be informed of the specific tasks being performed by the provider that are facilitating their care. When smartphones are being used to access EMR, clinical calculators, and information resources, sharing the smartphone screen with the patient provides transparency and empowers the patient to participate in their care. Visual information shared by smartphone device augments the verbal exchange of information between patient and provider and facilitates shared decision making.
Institutional policies on smartphone use
Institutions should have policies delineating appropriate and inappropriate smartphone use and notify employees of these policies. Considerations in the development of such policies must include patient care versus non-patient care areas, the rights of employees, patient safety and privacy, and professional versus personal use. Disciplinary actions may range from verbal warnings with documentation to dismissal depending on the offense. Administration should develop smartphone photography policies and outline the consequences for misuse. Institutional policies should also address smartphone use by patients and their families.
When personal smartphones are utilized for patient care activities, institutional policy should address device encryption and password protection. Institutions have a duty to patients and providers to reasonably protect their privacy against breaches. Difficulties associated with institutional oversight of personal devices may require that efforts are directed at provider education and institution-provider contracts on device parameters and use. Mandatory reporting of lost devices and remote wiping capabilities are appropriate policies. Multi-factor authentication should be considered for access to protected health information. Long-term storage of patient-related data or photographs on a personal device is inappropriate. While not all breaches in confidentiality are preventable, reasonable institutional oversight minimizes the risk and ramifications of such a breach.
When smartphone photography is used for patient care, the images should be integrated into the medical record. Institutions should have a mechanism by which smartphone photographs may be readily incorporated into the paper or electronic medical record and provide education to providers in this regard. Requiring integration into the medical record at the time that patient care is delivered minimizes the length of time that the image is retained on a personal device and risk that the provider fails to upload the photograph to the medical record.
Policies delineating the appropriateness of smartphones for telemedicine are encouraged. Such policies should be in accordance with state and federal laws. Video and conferencing apps enable remote care of patients and interdisciplinary cooperation, which enhance patient care when utilized appropriately. Telemedicine is not a replacement for bedside patient care when such care is reasonably available.
Professional and non-professional uses of smartphones can create distractions that can be detrimental to patient care. Healthcare providers should be educated on the dangers of distractions and be trained on how to minimize distractions related to personal devices. The use of smartphone functions, such as alarms, notes, and direct inter-provider communications, to decrease distraction is encouraged. Further research is needed in the area of distracted doctoring to determine the scope of this problem and the most effective intervention strategies.
Smartphone photography has the potential to improve patient care on the global and individual level but is associated with many pitfalls due to the ease and ubiquity of smartphone photography in general. Patient autonomy and privacy are paramount. Respect for the patient always trumps any potential benefit, global or individual, afforded by a photograph. Consent should be obtained just prior to obtaining a photograph and should be witnessed and documented. Photographs or videos for entertainment or personal uses is not appropriate. Providers have a duty to peer-regulate and intervene in the case of inappropriate smartphone use. Healthcare institutions should have protocols for reporting and addressing inappropriate use of smartphone photography.
The use of smartphones for personal applications should be limited to designated break or lounge areas. Smartphones should be set to modes that eliminate or minimize interruptions during patient encounters. Providers should seek permission from patients prior to integrating smartphones into the provider-patient relationship to improve communication and education. Providers should strive for transparency with regard to their professional use of smartphones explain and show to patients the clinical applications of smartphones.
These guidelines do not address legal aspects of restricting mobile phone use in clinical settings; however, they serve as a basis for the conversation to begin. While the Medicines and Healthcare Products Regulatory Agency (MHRA) does not support a blanket ban on the use of mobile phones in hospitals, some health systems are taking it upon themselves to implement regulations.15 For example, the Jewish General Hospital in Montreal, Quebec, has instituted a policy that addresses “the use of cell phones in the hospital for phone calls and for data usage (including text messages, browsing the internet or other). This policy applies to all cell phone users in the hospital, including staff, members of the CPDP, consultants, volunteers, visitors, and patients.”16 This policy looks to mitigate the use of mobile devices to respect the patients’ rights, with a focus on safety and confidentiality. Another hospital, Union Hospital in Eklton, Maryland, has issued a policy that includes advising employees that the “use of cell/camera phone during work, for other than hospital business should be avoided. Personal calls should be limited to break and meal breaks. All employees are required to silence their cell/camera phones while they are working.”17 Similarly, Greenville Hospital System in Greenville, South Carolina does not look to outright band mobile phone use, but advises that,
“During work time, employees are expected to exercise the same discretion with the use of personal communication devices as is expected with the use of any business phone. Personal phone calls (including text messaging) during the work day, regardless of the phone or device used, are not appropriate can interfere with productivity and be distracting to others.”18
While these are only a few examples of hospitals and health systems that have instituted policies around mobile devices, they do demonstrate a growing movement towards regulating such devices. This movement, while presumably controversial, is a natural occurrence as the prevalence of mobile technology increases. As such, hospitals and health systems will need to take a position on how strict their regulations will be, with the goal of improving patient care and safety.
As personal devices are becoming a technology that is integral to patient care, healthcare institutions should develop written policies regarding smartphone use. Policies should address appropriate use, consequences of misuse, mechanisms to protect patient information, integration of communication and images into medical records, and guidelines for telemedicine.
These guidelines establish a foundation for the professional use of smartphones in the healthcare setting. The very nature of smartphones, as personal devices, mandates that the use of smartphones be largely regulated on the individual and peer level. This self-regulation demands a strong internal locus of professionalism that must be developed, practiced, and assessed among all healthcare professionals.
The authors wish to acknowledge Nicholas DeVito, MPH; Lucy Hammond, PhD; Marguerite Heyns; Mark Kuczewski, PhD; Dennis Novack, MD; Steven Rosenzweig, MD; and Glen Solomon, MD for their significant contributions to the roundtable discussions.
1. Pew Research Center. Technology Device Ownership: 2015 [Internet]. 2015. [cited 23 January 2017]. Available from: http://www.pewinternet.org/2015/10/29/technology-device-ownership-2015/pi_2015-10-29_device-ownership_0-01/.
2. Anderson M. 6 facts about Americans and their smartphones [Internet]. 2015. [cited 23 January 2017] Avialable from: http://www.pewresearch.org/fact-tank/2015/04/01/6-facts-about-americans-and-their-smartphones/.
3. Wiechmann W, Kwan D, Bokarius A, et al. There’s an App for That? Highlighting the Difficulty in Finding Clinically Relevant Smartphone Applications. West J Emerg Med. 2016;17:2:191–4.
4. Kern C. Nurses Use Smartphones, Apps More Than Doctors. [Internet]. 2015 [cited 23 January 2017]. Available from: http://www.healthitoutcomes.com/doc/nurses-use-smartphones-apps-more-doctors-0001.
5. Statista. Physicians’ usage of smartphones for professional purposes in the U.S. from 2012 to 2015. [Internet]. 2015. [cited 23 January 2017]. Available from: http://www.statista.com/statistics/416951/smartphone-use-for-professional-purposes-among-us-physicians/.
6. Telenor Group. New study: The world is ready for mobile healthcare. [Internet]. 2012. [cited 23 Janaury 2017]. Available from: https://www.telenor.com/media/press-releases/2012/new-study-the-world-is-ready-for-mobile-healthcare/.
7. Prochaska M, Bird A, Chadaga A, Arora V. Resident Use of Text Messaging for Patient Care: Ease of Use or Breach of Privacy?. J Med Internet Res. 2015; 3(4):e37.
8. McCartney P. Texting protected health information in healthcare. MCN. Am J Matern Child Nurs. 2015; 40(1):61.
9. Wu R, Tzanetos K, Morra D, Quan S, Lo V, Wong B. Educational impact of using smartphones for clinical communication on general medicine: More global, less local. J Hosp Med. 2013; (7):365.
10. T, Smith S. Using digital videos displayed on personal digital assistants (PDAs) to enhance patient education in clinical settings. Int J Med Inform 2007;76:829–35.
11. Lei W, Pedersen P, Strong D, Tulu B, Agu E, Ignotz R. Smartphone-based wound assessment system for patients with diabetes. IEE Biomed Health Inform. 2015; (2): 477.
12. Miller K, Ziegler C, Greenberg R, Patel P, Carter M. Why Physicians Should Share PDA/Smartphone Findings With Their Patients: A Brief Report. J Health Commun. 2012;17:54–61.
13. Hsieh C, Yun D, Bhatia A, et al. Patient perception on the usage of smartphones for medical photography and for reference in dermatology. Dermatol Surg. 2015;41 (1):149–54.
14. Tran K, Lo V, Abrams H, Wu R, Morra D, Quan S. Medical students and personal smartphones in the clinical environment: The impact on confidentiality of personal health information and professionalism. J Med Internet Res. 2014;16(5): e132.
15. Muhammed R. Banning mobile phones in hospitals: is there enough evidence? Arch Dis Child. 2006;91(12):1043.
16. Jewish General Hospital. A new policy regulates the use of cell phones in hospital facilities, taking into consideration technical, ethical and practical issues [Internet]. n.d. [cited 2018 May 9]. Available from: http://jgh.ca/uploads/common/jgh%20cell%20 phone%20policy.pdf.
17. Union Hospital. Hospital policies and procedures: cell/camera phone use [Internet]. 2015 [updated 2015 Feb; cited 2018 May 9]. Available from: https://www.uhcc.com/wp-content/uploads/2016/03/HR-325-Cell-Camera-Phone-Policy.pdf.
18. Greenville Hospital System. Manual of policy directives: use of cell phones and other portable communication devices [Internet]. 2010 [cited 2018 May 9]. Available from: http://hsc.ghs.org/wp-content/uploads/2013/12/Use-of-Cell-Phones-and-Other-Portable-Communication-Devices.pdf.
19. Ross S, Forgie S. Distracted doctoring: smartphones before patients?. CMAJ [serial on the Internet]. (2012, Sep 4), [cited 29 January, 2017];184(12): 1440.
20. Corderoy A. Smartphones raise privacy issues in healthcare. Sydney Morning Herald. [Internet]. 2013. [cited 29 January 2017] Available from: http://www.smh.com.au/digital-life/digital-life-news/smartphones-raise-privacy-issue-in-healthcare-20130912-2tm51.html#ixzz45j4Btzqh.
21. Farooq A, White J. #Nomoretextbooks? The impact of rapid communications technologies on medical education. Can. J. Surg [serial on the Internet]. (2014, Jan 1), [cited 29 January 2017]; 57(4): E119-E120.
22. Miliard M. Nurses turning to smartphones for clinical advice. Healthcare IT News. [Internet]. 2015. [cited 29, January 2017] Available from: http://www.healthcareitnews.com/news/nurses-turning-smartphones-clinical-advice.
23. Engel H, Huang J, Tsao C, Lin C, Chou P, Cheng M, et al. Remote real-time monitoring of free flaps via smartphone photography and 3G wireless internet: A prospective study evidencing diagnostic accuracy. Microsurgery [serial on the Internet]. (2011, Nov), [cited 29 January 2017];31(8): 589.
24. Kaltman J, Best S, Kaltman S. Virtual rounding via videoconference-enabled smartphones: A case for multifacility rounds. Oral Surg Oral Med Oral Pathol Oral Radiol Endod [serial on the Internet]. (2012, June 1), [cited 29 January 2017];113(6): e15–e18.