This is an open-access article distributed under the terms of the Creative Commons Attribution License (https://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work, first published in JMIR mHealth and uHealth, is properly cited. The complete bibliographic information, a link to the original publication on https://mhealth.jmir.org/, as well as this copyright and license information must be included.
The world is aging. The number of older patients is on the rise, and along with it comes the burden of noncommunicable diseases, both clinical and economic. Attempts with mobile health (mHealth) have been made to remedy the situation with promising outcomes. Researchers have adopted human-centered design (HCD) in mHealth creation to ensure those promises become a reality.
This systematic review aims to explore existing literature on relevant primary research and case studies to (1) illustrate how HCD can be used to create mHealth solutions for older adults and (2) summarize the overall process with recommendations specific to the older population.
We conducted a systematic review to address the study objectives. IEEE Xplore, Medline via Ovid, PubMed, and Scopus were searched for HCD research of mHealth solutions for older adults. Two independent reviewers then included the papers if they (1) were written in English, (2) included participants equal to or older than 60 years old, (3) were primary research, and (4) reported about mHealth apps and their HCD developments from start to finish. The 2 reviewers continued to assess the included studies’ qualities using the Mixed Methods Appraisal Tool (MMAT). A narrative synthesis was then carried out and completed.
Eight studies passed the eligibility criteria: 5 were mixed methods studies and 3 were case studies. Some studies were about the same mHealth projects with a total of 5 mHealth apps. The included studies differed in HCD goals, target groups, and details of their HCD methodologies. The HCD process was explored through narrative synthesis in 4 steps according to the International Standardization Organization (ISO) standard 9241-210: (1) understand and specify the context of use, (2) specify the user requirements, (3) produce design solutions to meet these requirements, and (4) evaluate the designs against requirements. The overall process and recommendations unique to older adults are summarized logically with structural order and time order based on the Minto pyramid principle and ISO 9241-210.
Findings show that HCD can be used to create mHealth solutions for older adults with positive outcomes. This review has also summarized practical HCD steps and additional suggestions based on existing literature in the subfield. However, evidence-based results are still limited because most included studies lacked details about their sampling methods and did not set objective and quantifiable goals, leading to failure to draw significant conclusions. More studies of HCD application on mHealth for older adults with measurable design goals and rigorous research strategy are warranted.
The word “mHealth,” or “mobile health,” has been rising in popularity. A search of the term in an academic research database bears tens of thousands of results in 2020 alone. It is being studied as a medical intervention for arthritis [
mHealth and telehealth are the means to achieve timely and accurate health management; they help enable a seamless sharing of medical information between all those involved, creating the so-called connected health environment that the current trend strives for [
Although mHealth has remarkable potential, most projects cannot scale to their own target population and fail to achieve the intended results. This can be attributed to (1) poor understanding of the end users and (2) failing business models [
These issues get even more complicated with older adults. The United Nations defines older persons as those aged over or equal to 60 or 65 years; now, over 703 million people are aged over 65 years, and that number is projected to double by 2050 [
In searching for the best methodology to create the most usable mHealth, many have put the said value at the core of their work: having the humans at the center of focus. This review aims to explain how HCD can be applied to create mHealth suitable for older adults and to summarize the overall process with recommendations from relevant primary research studies of mHealth design and development.
The research question of this review is the following: How can HCD be used to create mHealth solutions for older adults? This issue was formed during the first author’s attempt to develop an mHealth app for older adults to solve their current pain points in a geriatric wellness clinic. Despite the constant mentioning of HCD, previous scoping searches of literature bear a heterogeneous group of research studies differing in interpretation, execution, and the extent of evaluation. The need for further clarification on the procedural details is identified.
A systematic approach following Siddaway et al’s guide [
Community, primary, secondary, or tertiary care.
Any qualitative, quantitative, or mixed methods study of original primary research.
Participants must include, but not limited to, older adults (aged ≥60 years).
Design goals must focus on mobile health (mHealth) solutions in the form of mobile apps intended for older adults.
Study procedures must be in line with the human-centered design (HCD) philosophy.
Studies must include details of mHealth apps and their development process, participants, design goals, and some implementation data.
Studies depicting different processes of the same product/project are included. For example, an mHealth project might have 2 separate papers such as 1 for design and 1 for evaluation; both are included in this review.
Trial and pilot studies are included.
Non-English language papers.
Any type of literature review, narrative review, or systematic review.
Studies with no relevant data or information that is of interest to the review question.
Systematic searches were conducted from the following 4 databases: IEEE Xplore, Medline via Ovid, PubMed, and Scopus. To best ensure comprehensive search results, search strings were compiled from keywords of the review question. Listed below are those strings with their corresponding similar terms:
“mHealth” OR “mobile health”, for the app to be reviewed;
“human centered” OR “human centered” OR “user centered” OR “user centered”, the approach in question;
“design” OR “development”, the process required;
“usability”, an outcome of HCD according to ISO 9241-210;
“elderly” OR “older adults” OR “geriatric”, the target population.
Each group of strings was put together with the “AND” Boolean operator in the search engines as all of the above key terms were required by the set eligibility criteria. No date range was set. Manual searches on Google Scholar and the references of the eligible papers were also conducted to identify possible additional relevant papers for screening. All searches were performed by a single reviewer (ZN) on the same day (November 12, 2020). The reason why the ACM Digital Library was not included is discussed in the “Limitations” section.
Microsoft Excel was used to record and manage the search results; duplications were removed. Two independent reviewers (SB and ZN) screened the deduplicated results by titles and abstracts. The full-text screening was done by the same reviewers using the eligibility criteria from
Quality appraisal of included studies.
Studies | Criteria from the Mixed Methods Appraisal Tool | ||||||||||||||
1.1a | 1.2b | 1.3c | 1.4d | 1.5e | 4.1f | 4.2g | 4.3h | 4.4i | 4.5j | 5.1k | 5.2l | 5.3m | 5.4n | 5.5o | |
Cornet et al [ |
1 | 1 | 1 | 1 | 1 | N/Ap | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
Cornet et al [ |
1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 | 1 | 0 |
Fortuna et al [ |
1 | 0 | 0 | 1 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 0 | 0 | 0 | 0 |
Harte et al [ |
1 | 1 | 1 | 1 | 1 | 0 | 1 | 1 | 0 | 1 | 1 | 1 | 1 | 1 | 1 |
Harte et al [ |
1 | 1 | 1 | 1 | 1 | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
Petersen et al [ |
1 | 1 | 1 | 1 | 1 | 0 | 0 | 1 | 0 | 0 | 1 | 1 | 1 | 1 | 0 |
Srinivas et al [ |
1 | 1 | 1 | 1 | 1 | 0 | 0 | 0 | 0 | 1 | 0 | 1 | 1 | 1 | 0 |
Stara et al [ |
1 | 1 | 1 | 1 | 1 | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A | N/A |
aIs the qualitative approach appropriate to answer the research question?
bAre the qualitative data collection methods adequate to address the research question?
cAre the findings adequately derived from the data?
dIs the interpretation of results sufficiently substantiated by data?
eIs there coherence between qualitative data sources, collection, analysis, and interpretation?
fIs the sampling strategy relevant to address the research question?
gIs the sample representative of the target population?
hAre the measurements appropriate?
iAre the confounders accounted for in the design and analysis?
jIs the statistical analysis appropriate to answer the research question?
kIs there an adequate rationale for using a mixed methods design to address the research question?
lAre the different components of the study effectively integrated to answer the research question?
mAre the outputs of the integration of qualitative and quantitative components adequately interpreted?
nAre divergences and inconsistencies between quantitative and qualitative results adequately addressed?
oDo the different components of the study adhere to the quality criteria of each tradition of the methods involved?
pN/A: not applicable.
One independent reviewer (ZN) performed data extraction from the 8 eligible papers. The information from 5 mixed methods studies included (on the data extraction form) the year of the study, the country of the study, the name of the project (if stated), study design, design goals, participants, study methods, quantitative or qualitative data used, results, and key discussions. The information from the other 3 case studies included the year of the study, the country of the study, goals, and results. All extracted texts were manually typed in Microsoft Excel.
Because of the heterogeneous nature of the included studies, narrative synthesis was chosen. Following Popay et al’s guide [
Developing a theory of how the intervention works, why, and for whom
Previous studies were carried out under the same hypothesis that human-centered design (HCD) helps make a more usable system for its users. This review adopted that same assumption and aimed to elaborate on how HCD works, especially for older adults, in steps.
Developing a preliminary synthesis of findings of included studies
Textual descriptions together with tabulation were chosen to summarize and display the extracted data. A recurring concept was identified across the studies: the HCD process. To ensure transparency, suggested HCD activities from ISO 9241-210 were chosen to categorize these patterns into 4 steps as follows: (1) understand and specify the context of use, (2) specify the user requirements, (3) produce design solutions to meet these requirements, and (4) evaluate the designs against requirements [
Exploring relationships in the data
Qualitative case descriptions were used to explore details and findings among included studies that correlate with each theme/step. A conceptual diagram was then created to answer the review question. The diagram was structured according to the Minto pyramid principle, using the following rules: (1) ideas at any level in the pyramid must always be summaries of the ideas grouped below them, (2) ideas in each grouping must always be the same kind of idea, and (3) ideas in each grouping must always be logically ordered [
Assessing the robustness of the synthesis
All included studies were appraised by Mixed Methods Appraisal Tool (MMAT), and the synthesis process was reflected on critically.
PRISMA flow diagram. mHealth: mobile health; PRISMA: Preferred Reporting Items for Systematic Reviews and Meta-Analyses.
Five studies were mixed methods, employing both quantitative and qualitative techniques to design and evaluate mHealth apps. Three were qualitative case studies focusing on describing the methodology and problems found during the process with little or no actual quantitative or qualitative data from the research shown. One of the 3 case studies was about the same process from the same project of another included mixed methods study; it was included for its qualitative reflections on the design and development processes. All studies stated clear aims of the research and were deemed relevant to this review question.
To illustrate the overall quality of the included studies, their quality appraisal scores were reported in the MMAT-suggested format [
Summary of the included mixed methods studies.
Study | Project | Setting | Design goal | Participants | Methods |
Cornet et al [ |
Engage | Academic health center, the United States. | To evaluate and test the usability of a self-managing heart failure system for older adults developed in a study by Srinivas et al [ |
(1) 13 older adults and (2) 2 caregivers |
Study I: (1) A structured interview was used to assess participants’ daily self-management routines and technology familiarity. (2) The think-aloud method was employed as each participant completes 8 given tasks on the system. (3) Feedback from the patients after they finish was used. Study II: (1) The system was re-designed after Study I. (2) A structured interview was used. (3) The think-aloud method was employed as each participant completed a given scenario in which he/she was to act as if he/she were the assigned fictitious character. (4) Feedback from the patients after they finish was used. (5) SUSa was used after usability evaluations. (6) NASA-TLXb was used after usability evaluations. |
Fortuna et al [ |
—c | Specialized center, the United States. | To incorporate an existing psychosocial intervention into a selected mobile platform. | Phase I and Phase II: (1) authors; Phase III: (1) older adults and (2) experts; and Phase IV: (1) 10 middle-aged and older adults |
Phase I: (1) A literature review was done to identify requirements. Phase II: (1) A literature review was done to find a suitable existing mobile platform. Phase III: (1) The interdisciplinary panel of end users and experts work together to incorporate an existing psychosocial intervention into the chosen mobile platform. Phase IV: (1) The think-aloud method was employed as each older adult goes through task-based usability testing. (2) Feedback from patients was collected. (3) Surveys based on SUS; Post-Study System Usability Questionnaires; and USEd questionnaires were used after each usability testing. (4) The ability to perform tasks without help was recorded in percentage. |
Harte et al [ |
Wireless Insole for Independent and Safe Elderly Living | Academic health center, Ireland. | To develop, assess, and enhance usability and user experience of a mobile app of a connected health system designed for fall risk assessment and fall detection. | Phase I: (1) 10 experts and (2) 12 older adults; Phase II: (1) 10 experts from Phase I; and Phase III: (1) 10 older adults from Phase I |
Phase I: (1) Likert scales were used to rate mock-ups called use cases. (2) The think-aloud method was employed during use case analysis. (3) Self-reported measures of the experts were collected. (4) Visual perception and cognitive processing metrics of older adults were collected. Phase II: (1) Likert scales were used to rate the paper prototypes based on use cases by experts. (2) ASQe and chosen usability metrics were used to rate the developed mobile working prototypes by experts after scenario-based usability testing. (3) The think-aloud method was employed during experts’ mobile working prototype runs. Phase III: (1) Likert scales were used to rate the mobile working prototypes by older adults. (2) ASQ, SUS, NASA-TLX, and chosen usability metrics were used to rate the mobile working prototypes by older adults after scenario-based usability testing. (3) The think-aloud method was employed during older adults’ working prototype runs. |
Petersen et al [ |
— | Academic health center, the United States. | To create a mobile app for older adults to monitor their use of a Bluetooth-connected resistance band for sarcopenia prevention. | Round 1: (1) 6 older adults; Round 2: (1) 3 clinicians and (2) 4 older adults; Round 3: (1) 3 clinicians and (2) 6 older adults |
Round 1: (1) Semistructured interviews gave information on how the app can be of use. Round 2: (1) The think-aloud method was employed as participants go through the wireframes. (2) A verbal prompting method was employed to encourage participants to give their thoughts. (3) Oral feedback from participants was recorded as they go through the video contents to be used in the prototype app. (4) The SUS was used after each participant finishes. (5) The USE score was used after each participant finishes. Round 3: (1) The think-aloud method was employed as participants go through the wireframes. (2) A verbal prompting method was employed to encourage participants to give their thoughts. (3) The SUS was used after each participant finishes. (4) The USE score was used after each participant finishes. |
|
|||||
Srinivas et al [ |
Engage | Specialized center, the United States. | To design, develop, and evaluate a consumer-facing health information technology system that supports heart failure self-care. | Phase I: (1) 63 older adults, (2) 35 caregivers, and (3) additionally data on 66 patients obtained from other literature; Phase II: (1) experts; Phase III: (1) 1 expert and (2) 5 older adults |
Phase I: Major themes of the app were synthesized from data gathered through direct observations at patient outpatient visits, standardized surveys on patient self-care, patients’ electronic medical record reviews, and semistructured interviews focused on patient self-care. Phase II: Core activities of the app were determined through educating, brainstorming, and design sessions of the research team. Phase III continues in Cornet et al (2017) [ |
aSUS: System Usability Scale.
bNASA-TLX: NASA-Task Load Index.
cNot stated.
dUSE: Usefulness, Satisfaction, and Ease of Use.
eASQ: After Scenario Questionnaire.
All 5 mHealth projects, from the included 5 mixed methods studies and 3 case studies, have the 4 key steps from ISO 9241-210 in their HCD processes, albeit described and mentioned to varying degrees. This section explores and illustrates these recurring steps across all included studies using the qualitative case description technique. All 8 studies are summarized and described in 4 HCD steps. Each step has 5 paragraphs representing a total of 5 mHealth projects: the first for patients with heart failure [
Understanding the context of use such as the end users, their current tasks, key activities, and working environment is essential to the design process; it helps guide how solutions should be tailored and set practical goals for the project [
Srinivas et al [
Fortuna et al [
Stara et al [
Petersen et al [
Cornet et al [
The second step of HCD focuses more on synthesizing further outputs from the first step. The goal is to derive what the users need to do and their objectives based on the gathered context and then set a clear statement of user requirements for the solution designs [
Srinivas et al [
Fortuna et al [
Stara et al [
Petersen et al [
Cornet et al [
This HCD step focuses on designing how the users interact with the system based on the requirements from the previous step [
Srinivas et al [
Fortuna et al [
Stara et al [
Petersen et al [
Cornet et al [
The human-centered evaluation activity is vital to HCD and is iterative by its nature [
Srinivas et al [
Fortuna et al [
Stara et al [
Petersen et al [
Cornet et al [
This systematic review has shown how HCD can be used to create mHealth for older adults, with additional recommendations reported. Eight studies are included in this review: 5 are mixed methods studies and 3 are case studies. All studies were published recently starting from 2017 onward, suggesting that the subfield is relatively new. All were conducted in developed countries and mostly in academic or specialized health care settings. Because of the diverse methodologies and details of the included studies, we used the Minto pyramid principle and the 4 HCD steps from ISO 9241-210 to guide the creation of 3 conceptual models:
Pyramid model of HCD project by structural order. HCD: human-centered design.
Pyramid model of HCD process with recommendations for older users by time order. HCD: human-centered design.
First, mHealth ideas, either novel or of existing concepts, should be based on what the users need, not what the creators want. As illustrated in the included studies, the authors, usually acting as the management team that oversees the project, identify and base their proposed mHealth solutions on real stakeholders both directly [
The included studies’ details and rationales for the number of participants and the sampling method were diverse and vague. For example, Fortuna et al [
Besides quantity, most included studies recommended that the sampling method include diverse groups of participants to ensure HCD solutions reflect real-world problems. Reaching out to older adults who are more physically inept or socially disadvantaged can be challenging. Fortuna et al [
Information on the context of use was mostly gathered through qualitative methods in the included studies. Observation is valued more than opinions in HCD: it shows how users currently pursue their goals from an unbiased perspective. Interviewing techniques that can be employed are (1) the critical incident technique, and (2) the think-aloud method of a fictitious scenario [
Second, mHealth solutions need to address the current pain points of the users and ensure they achieve their intended goals; a clear understanding of user needs and a concise list of user requirements help the HCD team accomplish that. The context of use plays a vital role in identifying the user problems from their current activities and what the users need to solve them. Then, user requirements based on these user needs are created to guide the HCD team on how the solutions should be designed. These requirements can be obtained through a literature review [
The included studies also pointed out that there was a set of requirements unique to older users; however, most had not listed these requirements at the beginning and dealt with them only after the users raised the problems in usability testing. These design considerations for older adults are well-established: Harte et al [
Specifying user requirements also means setting measurable goals for the mHealth system. This usually requires gathering quantitative data for the context of use, such as the duration to complete the conventional I-IMR, which is approximately 8-10 months [
Third, HCD seeks to create an ideal system through iterative prototyping together with the stakeholders, to make certain all user needs and requirements are accounted for. In the beginning, the functional team should design how the users will interact with the system and how the interfaces will be like. Harte et al [
It should be noted that high-fidelity prototypes are not open for major changes or costly to do so; the best approach would be to finalize user–system interaction and user interfaces before their creation. As this HCD step requires iteration, an agile approach is recommended [
Fourth, evaluation comes after the production of design solutions. The functional team should work closely with the technical team to evaluate the produced solutions. In HCD, evaluating the designs is done by usability experts and end users. A number of included studies recommended system evaluation by usability experts before end users. Expert evaluation can identify and classify usability problems early in the process where changes are less punishing; it is also much simpler to arrange compared with its user-based counterpart [
Evaluation by end users is critical to the process because getting feedback from the target users and improving accordingly would surely make the system usable for them. This result can be obtained by recruiting the right participants. Fortuna et al [
User-based evaluation methods range from giving the users specific tasks in a controlled environment to letting them use the system in the real world; the complexity also increased respectively so. Cornet et al [
The included studies concurred that both qualitative and quantitative data should be interpreted together for robust usability evaluation results [
ISO 9241-210 defines the components of usability as effectiveness, efficiency, and satisfaction [
Although limited, the positive outcomes from the studies in this review show that HCD can create usable mHealth systems for older adults. Stara et al [
The author again emphasize that user involvement in HCD is paramount to obtain such outcomes. Older adults are not extra design challenges to solve. Empathy toward users as individuals with pain points is essential to HCD; stereotypes and bias against older people could lead to design failures if left unchecked [
Hastened by the COVID-19 pandemic, the field of mHealth will only expand. Moving forward, digital health solutions are aiming further than empowering patients and enhancing delivery. They are going for “digital therapeutics.” These evidence-based interventions aim to prevent and manage medical conditions through digital platforms and mobile devices; one of its focuses is to deliver lifestyle therapy to combat chronic diseases such as type 2 diabetes [
Limitations of this systematic review are acknowledged. First, the ACM Digital Library was not included in this review despite being in the relevant field. We did search the database on the same day as the others: no studies from the ACM Digital Library passed our criteria. We then failed to mention this once we proceeded with the review. However, we ran another search with the same strategy on the database in May 2021 to recheck; 14 studies found did not pass our abstract screening according to our established eligibility criteria.
Second, the research question aims to address the whole HCD process, but an existing body of literature proves to be limited as the topic is an emergent subfield, especially with older adults as the target group. Although the criteria are forgiving, the search strategy and the inclusion criteria still demand that all steps of HCD are implemented in each app development. This excludes a large number of studies that feature only a part of HCD. For example, one study might focus on qualitative interviews without applying them, while another might test a newly developed system that is based solely on the authors’ vision, not actual user needs. Nevertheless, the included studies complement one another and thus can accommodate the research question as illustrated in this review.
Third, the highly diverse HCD goals and methodologies in the included mHealth apps restrict the means of analysis and synthesis of results. All studies relied heavily on various qualitative means for HCD such as literature reviews, interviews, and field notes from direct observations. Even the seemingly same approaches, such as interviews, still differ in detail such as the time, the duration, the focus, and the questions. Most studies also focus more on the process not the result, or in the case of case studies, the process itself is the result. This might be due to the fact that baselines of the existing activities are not established in the second step of HCD, specifying the user requirements, so comparisons for effectivity and efficiency of the newly developed mHealth interventions cannot be made with objective metrics. Because of that fact, the included studies have to be reviewed with qualitative techniques using narrative synthesis and guided by ISO 9241-210 together with logical ordering of the Minto pyramid principle [
This leads to the fourth limitation regarding the included studies: all but 1 of the 5 mixed methods studies are rated to be of inadequate quality by the MMAT. Their quantitative components lack clarity. They do not explain their sampling methods or have done so insufficiently, resulting in the inability to deem their samples representative of the target population and failure to address possible confounding factors in making the conclusion that HCD helps make a usable product. This issue of the sampling methodology is also raised by the authors of the included studies; future HCD research should note this point in their strategic planning accordingly.
Finally, the authors stress that the aim of this systematic review was not to assess the implementation of HCD in creating mHealth for older adults or the effectiveness of mHealth interventions. The objective was to explore existing literature and establish recommendations and pitfalls for subsequent HCD projects. The older adults might be a narrow target population, but being the more sensitive and vulnerable group, the insight gained could be applicable to a wider range of users and help make future mHealth solutions more inclusive as well.
This review concludes that HCD can be used to create mHealth solutions for older adults and has summarized the process based on the 4 HCD steps with additional recommendations. The findings of this review can help designers, developers, and researchers gain an overview of HCD for older adults and implement the framework in their projects. The growing body of literature is encouraging, but more evidence-based results of HCD on creating mHealth for older adults are still needed. Future research should also focus on applying artificial intelligence and machine learning in HCD and utilizing the framework to create novel mHealth solutions for the population.
Preferred Reporting Items for Systematic Reviews and Meta-Analyses 2009 checklist of this systematic review.
After Scenario Questionnaire
cardiac implantable electronic devices
human-centered design
Health Insurance Portability and Accountability Act of 1996
Integrated Illness Management and Recovery
International Organization for Standardization
mobile health
Mixed Methods Appraisal Tool
NASA-Task Load Index
Newest Vital Sign
Preferred Reporting Items for Systematic Reviews and Meta-Analyses
Post-Study System Usability Questionnaire
the Quebec User Evaluation of Satisfaction with Assistive Technology questionnaire
System Usability Scale
Usefulness, Satisfaction, and Ease of use
Wireless Insole for Independent and Safe Elderly Living
This research project was funded by the Ratchadapisek Sompoch Endowment Fund (2021) under Telehealth Cluster, Chulalongkorn University. The authors also thank the Center of Excellence in Hepatitis and Liver Cancer, Department of Biochemistry, Faculty of Medicine, Chulalongkorn University, and Thailand Research Fund (RTA6280004).
None declared.