Systematic review of the effectiveness of health-related behavioral interventions using portable activity sensing devices (PASDs)

Hamed Abedtash; Richard J Holden

doi:10.1093/jamia/ocx006

. 2017 Feb 11;24(5):1002–1013. doi: 10.1093/jamia/ocx006

Systematic review of the effectiveness of health-related behavioral interventions using portable activity sensing devices (PASDs)

Hamed Abedtash ^*, Richard J Holden ^*,^*

PMCID: PMC6259669 PMID: 28339620

Abstract

Background: Portable activity sensing devices (PASDs) have received significant interest as tools for objectively measuring activity-related parameters and promoting health-related outcomes. Studies of PASDs suggest the potential value of integrating them with behavioral interventions to improve intermediate and downstream clinical outcomes.

Objectives: This systematic review describes and evaluates evidence from controlled studies of interventions using PASDs on their effectiveness in health-related outcomes. Study quality was also assessed.

Methods: A systematic literature search was performed of MEDLINE, Cochrane Central Register of Controlled Trials, PsycINFO, EMBASE, and CINAHL databases. We included English-language papers of controlled trials through 2015 reporting the effectiveness of PASDs in improving health-related outcomes in any population. We extracted and analyzed data on study characteristics including design, target population, interventions, and findings.

Results: Seventeen trials met the inclusion criteria from a total of 9553 unique records. Study objectives varied greatly, but most sought to increase physical activity. Studies with a “passive” intervention arm using a PASD with minimal behavioral support generally did not demonstrate effectiveness in improving health-related outcomes. Interventions integrating PASDs with multiple behavioral change techniques were more likely to be effective, particularly for intermediate outcomes such as physical activity and weight loss. Trials had small sample sizes but were generally free of bias, except for blinding and selection bias.

Conclusion: There is insufficient evidence to draw a conclusion about the general health-related benefits of PASD interventions. PASD interventions may improve intermediate outcomes when coupled with multiple behavioral change techniques. Devices alone or with minimal behavioral change support are insufficient to change health-related outcomes.

Keywords: physical activity tracker, activity sensor, portable sensor, wearable device, mobile health (mHealth) technology

INTRODUCTION

Portable activity sensing devices (PASDs), including wearable accelerometers and pedometers, have generated considerable interest from health care researchers.¹^,² These devices objectively measure and estimate physical activity, balance control, exercise adherence, activity intensity, and energy expenditure more easily and accurately than self-reporting questionnaires or diaries.^3–6 Mobile information technology (IT) advances – including wireless connectivity, real-time messaging, advanced visualization, and context awareness – also permit these devices to motivate and inform users.^7–9 With decreasing technology costs, PASDs have also penetrated the consumer marketplace with activity trackers (eg, Fitbit devices) and smartphones/smartwatches with sensor-based health applications (eg, Apple Health, Google Fit).^10–12

PASDs can play 2 important roles in health care delivery and health promotion. First, they provide increasingly powerful measurement, storage, and communication of health-related variables, such as the total amount, duration, frequency, timing, and intensity of physical activity, body postures, and body movements. These are interpreted into activity levels, step counts, fall risk estimates, and other constructs associated with cardiovascular disease,¹³ diabetes,^14–16 cancer,¹⁷^,¹⁸ hypertension,¹³ neurological disorders,¹⁹ gait disturbance,²⁰ and balance impairment and falls.²¹^,²² Regular assessment or remote monitoring of these measures could result in more timely, personalized, and appropriate therapeutic interventions or preventive strategies, implemented by clinicians or patients themselves.¹⁰^,²³ Individual-level PASD data can also be aggregated for population health surveillance and subgroup comparisons, as demonstrated by the use of pedometers in US and Japanese national health studies.²³

The second role PASDs can play in health and health care is in “behavioral informatics” interventions.²⁴ This involves integrating PASDs into behavioral interventions to improve physical activity or decrease health-related risks.²⁴^,²⁵ For example, devices can help individuals self-monitor physical activity over time,²⁶ motivate individuals to reduce sedentary behaviors,²⁵ or provide coaching to improve body balance.^27–29 These behavioral change techniques can be and often are combined and delivered through mobile, web, or desktop IT software linked to PASDs.³⁰ Such software can facilitate longitudinal tracking, feedback, motivational communication, goal setting, exercise planning, time management, and social media support.²³^,^31–35

Enough literature has accumulated to assess whether patients indeed experience better health-related outcomes when exposed to behavioral interventions using PASDs. Prior reviews demonstrated the validity,²⁶^,³⁶^,³⁷ feasibility,³⁷^,³⁸ reproducibility,³⁷ potential efficiency,³^,²⁶^,³⁰^,³⁸ and acceptance³⁹ of PASDs. However, the present study is, to our knowledge, the first to systematically review and synthesize evidence from controlled trials testing the health-related effectiveness of PASD-enabled behavioral interventions.

Objectives

The primary aim of this systematic review is to describe and evaluate controlled studies of interventions using PASDs. The review examines the effect of device-based interventions on intermediate outcomes such as physical activity levels and downstream clinical outcomes such as organ function. We hypothesize that intervention effectiveness varies depending on the nature of the behavioral change techniques used. A secondary objective was to evaluate the quality of reviewed trials.

METHODS

We conducted this systematic review of the English-language scholarly literature in accordance with Preferred Reporting Items for Systematic Reviews and Meta-analysis guidelines. One reviewer (HA) performed study selection, data extraction, and quality assessment and another (RJH) independently monitored these processes for completeness and accuracy.

Data sources and search queries

We searched 5 online databases from inception to December 30, 2015: MEDLINE (PubMed), Cochrane Central Register of Controlled Trials (CENTRAL), PsycINFO (EBSCO), EMBASE, and CINAHL (EBSCO). Queries covered 3 domains: (1) sensing devices, (2) portability or wearability, and (3) physical (body) activities (for details, see Supplementary Table S1). In addition, reference lists of relevant reviews and empirical studies were checked for eligible studies.

Study selection

Records were downloaded into an EndNote X7.7 (Clarivate Analytics, Philadelphia, PA, USA) library. After removing duplicate records, we screened titles and abstracts for inclusion and exclusion criteria (for details, see Supplementary Table S2). We used a “topic-collated” approach to accelerate screening: instead of sequentially reviewing publications in author or chronological order, we collated records by topics using keyword searches in the EndNote library prior to full-text review.

Studies were included in the systematic review if they were in English, randomized, studied humans of any age, and compared the health-related effectiveness of PASDs in the target population against a control group. Journal and peer-reviewed conference articles were included.

Data extraction

Data were extracted from eligible articles into a spreadsheet. Data included study design, objectives, target population, participant characteristics, interventions, device type, device data capture and processing, principal findings, and adverse events.

Quality assessment

Articles included in the final review were assessed for quality and risk of bias using updated criteria from the Cochrane Consumers and Communication Review Group.⁴⁰ Studies were graded as low, unclear, or high for each Cochrane bias domain: selection, performance, detection, attrition, reporting, and other (Cochrane Handbook, Table 8.5.d).⁴¹

RESULTS

We identified 9771 search records and 4 additional papers through cited reference search (Figure 1 ). After removing duplicates, we reviewed titles and abstracts for 9553 records. Of these, 920 underwent full-text review. Most articles were excluded because the PASD was used not in the main intervention but to detect objective physical activity levels. A final total of 17 articles met inclusion and exclusion criteria, each using a PASD in the main intervention (Table 1; for more details, see Supplementary Table S3). All 17 met minimal quality standards for review inclusion.

Table 1.

Study characteristics of reviewed randomized clinical trials

Reference	Outcome	Intervention duration	Group size	Intervention and comparison group characterization
Paschali et al. (2005)⁴²	Exercise adherence	3 months	13	Intervention: accelerometer; recorded daily activity in a paper diary, received individualized exercise plan
Paschali et al. (2005)⁴²	Exercise adherence	3 months	13	Comparison: accelerometer with no access to activity data, received individualized exercise plan
Butler et al. (2009)⁴³	Physical activity adherence	6 months	44	Intervention: pedometer; recorded daily steps, received motivating phone calls, set activity goals based on step counts, given brochure on physical activity
Butler et al. (2009)⁴³	Physical activity adherence	6 months	46	Comparison: given brochure about physical activity
Newton et al. (2009)⁴⁴	Physical activity level	12 weeks	38	Intervention: unblinded pedometer; 10 000 steps/day goal, recorded daily step counts, received reminder text messages
Newton et al. (2009)⁴⁴	Physical activity level	12 weeks	40	Comparison: closed pedometer; received standard care
McMurdo et al. (2010)⁴⁵	Physical activity	6 months	60	Pedometer plus behavioral intervention: pedometer; recorded daily pedometer counts in diaries, attended counseling sessions to increase physical activity, received individualized activity plan and barrier coping plans, received motivational phone calls
			53	Behavioral intervention: no pedometer; recorded walking time outdoors, received motivational phone calls
			66	Comparison: usual care; no pedometer and behavioral intervention
Maturi et al. (2011)⁴⁶	Physical activity and body measures	12 weeks	32	Intervention: pedometer; 10 000 steps/week goal, recorded step counts on a calendar, received individualized motivational consultations, received pamphlet on weight loss
Maturi et al. (2011)⁴⁶	Physical activity and body measures	12 weeks	32	Comparison: usual care; no pedometer
Shuger et al. (2011)³⁴	Weight loss and waist circumference	9 months	26	SenseWear armband alone (SWA): SenseWear; had access to real-time web-based personalized weight management, received regular feedback on energy expenditure through web, received weight loss manual, instructed to monitor diet, physical activity
			28	Group-based behavioral weight loss education (GWL): attended weight loss counseling sessions, received weight loss manual, instructed to monitor diet, physical activity
			37	Combined GWL and SWA group (GWL + SWA)
			32	Comparison: weight loss manual; instructed to monitor nutrition and physical activity
Kovelis et al. (2012)⁴⁷	Daily physical activity	1 month	23	Intervention: pedometer; 10 000 steps/day goal, recorded daily step counts
Kovelis et al. (2012)⁴⁷	Daily physical activity	1 month	17	Comparison: booklet on advantages of having an active life, disadvantages of smoking
Adams et al. (2013)⁴⁸	Physical activity	6 months	10	Intervention: open pedometer; set daily goals based on actual step count data, received motivational emails and short text messages, given motivational brochures and motivating feedback, given financial incentives for achieving daily goals
Adams et al. (2013)⁴⁸	Physical activity	6 months	10	Comparison: pedometer; 10 000 steps/day goal, received motivational emails and short text messages, given brochures to motivate physical activity, given feedback and incentives for submitting pedometer data (but not for meeting daily goals)
Bickmore et al. (2013)⁴⁹	Step counts	12 months	132	Intervention: pedometer; interacted daily with embodied conversational agent software, which provided 5-minconversation consisting of motivational feedback, troubleshooting discussion (if any), and individualized exercise counseling
Bickmore et al. (2013)⁴⁹	Step counts	12 months	131	Comparison: pedometer; recorded step counts daily
Wijsman et al. (2013)³⁵	Physical activity level and metabolism	3 months	114	Intervention: accelerometer with access to DirectLife web-based physical activity program; received individualized motivational feedback, set daily goals based on current activity
Wijsman et al. (2013)³⁵	Physical activity level and metabolism	3 months	112	Comparison: no activity monitor; given no instructions about daily physical activity
Choi et al. (2015)⁵⁰	Physical activity	12 weeks	15	Intervention: open accelerometer with access to activity data; received motivational messages or short videos from an application, set individualized weekly goals
Choi et al. (2015)⁵⁰	Physical activity	12 weeks	15	Comparison: open accelerometer with access to data; set goal to reach 8500 steps/day
Grewal et al. (2015)²⁸	Postural stability and daily physical activity	4 weeks	18	Intervention: LegSys; did balance training exercises, visualized movement errors during exercises, received visual and auditory (positive sound) feedback
Grewal et al. (2015)²⁸	Postural stability and daily physical activity	4 weeks	16	Comparison: no intervention; no training program
Mansfield et al. (2015)³²	Walking activity and gait performance	3–26 days	29	Intervention: accelerometer with access to data; physiotherapist and participant collaboratively assigned walking activity goals based on analyzed activity data
Mansfield et al. (2015)³²	Walking activity and gait performance	3–26 days	28	Comparison: accelerometer with access to data; physiotherapist discussed daily walking goals based on participant’s self-reported walking activity
Martin et al. (2015)³³	Physical activity	Phase 1: 2 weeks	32	Unblinded, no text message: accelerometer with access to data through Fitbug application
		Phase 1: 2 weeks	16	Blinded, with no text messages: accelerometer with no access to activity information
		Phase 2: 2 weeks	16	Unblinded, personalized messages: accelerometer; had access to data through Fitbug, received personalized texts from physician, set 10 000 steps/day goal, received motivational messages
			16	Unblinded, no text messages: accelerometer with access to data through Fitbug
			16	Blinded, no text message: accelerometer with no access to activity information
Oh et al. (2015)⁵¹	Weight loss	24 weeks	118	Intervention: pedometer with access to data; entered step counts into mobile application; received clinical decision support system algorithm-based feedback, received health education and nutrition consults over the phone based on activity levels and lifestyle
Oh et al. (2015)⁵¹	Weight loss	24 weeks	210	Comparison: pedometer; kept body weight journals, recorded daily step counts on a log sheet, visited hospitals to meet physicians and get consultation about nutrition and exercise
Schwenk et al. (2016)²⁹	Balance and gait performance	4 weeks	9	Intervention: LegSys; did balance training exercises, visualized movement errors during exercises, received visual and auditory (positive sound) feedback
Schwenk et al. (2016)²⁹	Balance and gait performance	4 weeks	10	Comparison: no intervention; no formal exercise program
Ginis et al. (2016)⁵²	Body balance and quality of life	6 weeks	20	Intervention: CuPiD system; used auditory feedback and freeze-of-gait training smartphone applications with 2 inertial measurement units, received positive verbal feedback on optimal gait performance, cued continuously during walking, received booklet of personalized instructions, recorded frequency and duration of training in diary
Ginis et al. (2016)⁵²	Body balance and quality of life	6 weeks	18	Comparison: no application or device; received training and recommendations, recorded frequency and duration of training sessions

Open in a new tab

Study characteristics

All 17 reviewed studies were randomized clinical trials with a PASD incorporated into the main intervention (Table 1). A majority (76%) were published after 2010. Eight studies (47%) were conducted in the United States,²⁸^,²⁹^,³³^,³⁴^,⁴²^,^48–50 1 was a collaboration between researchers in Belgium and Israel,⁵² and others were reported by investigators in Australia,⁴³ New Zealand,⁴⁴ Scotland,⁴⁵ Iran,⁴⁶ Brazil,⁴⁷ the Netherlands,³⁵ Canada,³² and South Korea.⁵¹ Sample sizes of the trials ranged from 19 to 328, with a median of 57 participants.

The reviewed studies mainly aimed to improve physical activity levels, metabolism, physiologic measures, body measures, balance control, and quality of life (Figure 2 ).

Figure 2. — Objectives and studied interventions of the 17 reviewed trials. ^aFor example, self-monitoring, motivational messages, goal setting, and exercise planning.

Three trials (18%) studied multiple intervention groups, comparing the effectiveness of PASDs integrating different behavioral change techniques.³³^,³⁴^,⁴⁵ In 9 studies (53%), intervention participants wore PASDs and had access to device data, while controls did not use the device.²⁸^,²⁹^,³⁴^,³⁵^,⁴³^,^45–47^,⁵² In 3 trials (18%), both intervention and control participants used PASDs, but only the intervention group could access device data.³³^,⁴²^,⁴⁴ Six additional studies (35%) provided PASDs and access to device data in both trial arms, but intervention participants received individualized exercise plans based on device-assessed activity data, while controls were provided static or no exercise planning.³³^,³⁴^,^48–51

Target population characteristics

Included studies evaluated the effectiveness of accelerometer-based activity programs among 1665 total participants. Thirteen trials (76%) included men and women,²⁸^,²⁹^,^32–35^,^42–44^,^47–49^,⁵¹ 3 (18%) studied women only,⁴⁵^,⁴⁶^,⁵⁰ and 1 did not specify gender distribution.⁵²

Five studies (29%) enrolled younger and older adults,²⁸^,³²^,⁴³^,⁴⁷^,⁵¹ 4 trials (24%) enrolled older adults only,²⁹^,³⁵^,⁴⁵^,⁴⁹ 6 trials (35%) included younger adults only,³³^,³⁴^,⁴²^,⁴⁶^,⁴⁸^,⁵⁰ 1 enrolled adolescents,⁴⁴ and 1 did not specify age group.⁵² Age distribution was poorly reported, with only 4 trials providing complete age information.²⁸^,²⁹^,³²^,⁴² Limited information was also reported on race and ethnicity of participants, with only 4 studies reporting distributions of white vs non-white participants,³³^,^48–50 and 3 reporting Hispanic vs non-Hispanic distributions.^48–50

Reviewed studies recruited individuals with different diseases or health-related conditions, including type 1 diabetes,⁴⁴ type 2 diabetes,²⁸^,⁴² cardiovascular disease,³³^,⁴³ sedentary lifestyle,³⁴^,³⁵^,⁴⁵^,⁴⁸^,⁴⁹ postpartum care,⁴⁶ pregnancy,⁵⁰ smoking,⁴⁷ stroke,³² metabolic syndrome,⁵¹ chemotherapy-induced peripheral neuropathy (CIPN),²⁹ and Parkinson’s disease.⁵²

Utilized portable activity sensing devices

All interventions involved 1 form of PASD to objectively track participants’ activity (Table 2). Eight studies (47%) used pedometers to record daily step counts; devices included Omron⁴⁵^,⁴⁶^,⁴⁸^,⁴⁹ and Yamax Digiwalker.⁴³^,⁴⁷ Six trials (35%) used accelerometers to measure motion along 3 axes; devices included BioTrainer,⁴² SenseWear,³⁴ Fitbit Ultra,⁵⁰ Fitbug Orb,³³ Tracmor,³⁵ and X6‐2mini.³² Three studies (18%) used inertial measurement unit (IMU) sensors to measure kinematic parameters for balance assessment; these included LegSys²⁸^,²⁹ and EXLs3.⁵² Two trials did not disclose the PASD model used.⁴⁴^,⁵¹ Two studies used a separate PASD to measure physical activity outcomes (eg, step counts) independent of the device that was part of the intervention.³⁵^,⁴⁵

Table 2.

Technological characteristics of reviewed studies

Technology	No. of studies	References
Portable activity sensing devices
Pedometer	8	^43–49^,⁵¹
Accelerometer	6	^32–35^,⁴²^,⁵⁰
Inertial measurement unit	3	²⁸^,²⁹^,⁵²
Behavioral informatics software
No software	6	³²^,^43–48
Desktop application	3	²⁸^,²⁹^,⁴²
Web-based application	4	^33–35^,⁵⁰
Smartphone application	4	³³^,^50–52
Tablet application	1	⁴⁹
Real-time physical activity feedback	5	^33–35^,⁵⁰^,⁵¹
Interactive virtual coaching	1	⁴⁹
Audiovisual training program	3	²⁸^,²⁹^,⁵²
Sensor data retrieval and storage
Manually entered into diary	5	^43–47
Manually entered into software	2	⁵⁰^,⁵¹
Manually downloaded from sensor to software	5	³²^,³⁴^,³⁵^,⁴²^,⁴⁸^,⁴⁹
Real-time transfer from sensor to software	5	²⁸^,²⁹^,³³^,⁵²
Sensor data processing and display
To measure physical activity level, eg, step counts, energy expenditure, and activity duration	15	^32–35^,^42–52
To provide graphical feedback on physical activity	4	³³^,⁴²^,⁵⁰^,⁵¹
To provide audio feedback on gait performance and individualized gait advice	1	⁵²
To provide visual feedback on joint position	2	²⁸^,²⁹
To set individualized activity goals	6	³²^,⁴³^,⁴⁵^,^48–50
Used EHR-integrated CDSS algorithm for individualized feedback	1	⁵¹

Open in a new tab

EHR = electronic health record; CDSS = clinical decision support system.

Studied interventions

We identified 6 possible categories of PASD interventions, depicted in Figure 3 , based on the nature of behavioral change techniques used and whether IT software was used separate from the device itself. The 3 categories of behavior change techniques were: (1) passive self-monitoring, where participants wore a device and its data were available for self-monitoring; (2) goal-based self-monitoring, where participants self-monitored device data with a specific goal in mind; and (3) integrative interventions, involving a PASD and at least 3 behavioral change techniques, including goal-based self-monitoring, motivational messages, coaching or training programs, group-based education, planning, incentives, and combinations thereof.

Technology characteristics

Table 2 summarizes the hardware, software, data retrieval and storage, data processing, and display characteristics of the technology used in reviewed studies.

The IT software involved in behavioral informatics interventions varied in terms of platform (eg, desktop vs smartphone) and type (eg, real-time feedback vs virtual counseling agent). One study⁴² used a desktop application to provide graphical feedback of weekly physical activity to participants based on sensor data. In 3 trials, sensor data were stored in a web-based data-management application that provided real-time feedback to participants.^33–35^,⁵⁰ Smartphone applications were also used to collect sensor data and give feedback to participants.³³^,⁵⁰^,⁵¹ Two studies provided feedback through both smartphone and web applications.³³^,⁵⁰ In another trial, sensor data were downloaded to a tablet-based interactive counseling application, where an animated character delivered personalized exercise coaching.⁴⁹ In one study, patients with Parkinson’s disease received gait-related audiovisual feedback from smartphone-based training applications.⁵² Another interactive training program provided real-time visual feedback on joint positions and angles through processing of inertial sensor-based kinematic data, helping patients understand motor errors and better control balance.²⁸^,²⁹ Other studies did not specify the use of software.³²^,^43–48

In all studies, retrieval of data from the device was important to assess study outcomes or deliver the interventions. There were 4 ways to retrieve or transfer data from devices (Table 2). In 5 trials (29%), participants manually entered step counts from a pedometer into daily, weekly, or monthly paper diaries.^43–47 In other studies, sensor data were manually downloaded from the pedometer or accelerometer into desktop software or a web application by participants or investigators,³²^,³⁴^,³⁵^,⁴²^,⁴⁸^,⁴⁹ or participants entered step counts into a smartphone application.⁵⁰^,⁵¹ In fewer cases, sensor data were automatically transferred in real time to a smartphone application³³^,⁵² or a balance training program²⁸^,²⁹ for further processing.

Approaches to processing sensor data (Table 2) included applying algorithms to estimate step counts,³²^,³³^,³⁵^,^42–52 energy expenditure,³⁴ or physical activity duration.^32–35^,⁵⁰^,⁵² Mansfield et al.³² also processed accelerometer data using a customized, trained machine-learning algorithm to detect step counts, walking bouts, and activity duration. Other studies converted data into graphical feedback on physical activity level³³^,⁴²^,⁵⁰^,⁵¹ or audio feedback with individualized gait advice.⁵² Kinematic data in 1 study were used for visual feedback on joint positions and angles.²⁸^,²⁹ Moreover, collected data from sensors were often used to set individualized activity goals.³²^,⁴³^,⁴⁵^,^48–50 In one study, sensor data were integrated with the health care center’s server to generate individualized feedback based on a clinical decision support system algorithm.⁵¹

Effects of interventions on health-related outcomes

Tables 3 and 4 summarize the effects of reviewed interventions on physical activity (Table 3) and other intermediate and downstream health-related outcomes (Table 4). To test the hypothesis that PASD intervention effectiveness depends on the nature of the intervention, these tables organize interventions by the type of behavioral change techniques employed.

Table 3.

Reported effectiveness of portable activity sensing devices (PASDs) on physical activity, organized by category of employed behavioral change techniques

Outcomes and Behavioral Interventions	PASD with Passive Self-Monitoring		PASD with Goal-Based Self-Monitoring		PASD with Integrative Intervention
Outcomes and Behavioral Interventions	Improved	No change	Improved	No change	Improved	No change
Physical activity level
Self-monitoring		⁴²
Self-monitoring, goal setting				⁴⁷
Self-monitoring, motivational messages, goal setting
a. Under 6 months					⁴⁶	³² ^, ⁴⁴
b. 6 months or more					⁴⁵ ^, ⁴⁸
Self-monitoring via computer software		³³
Computerized goal setting, motivational message, self-monitoring
a. Under 6 months					³³ ^, ³⁵ ^, ⁴⁹	⁵⁰
b. 6 months or more						⁴⁹
Computerized balance training program, self-monitoring		²⁸
Physical activity duration
Self-monitoring		⁴²
Self-monitoring, motivational messages, goal setting
a. Under 6 months					⁴³	³² ^, ⁴⁴
b. 6 months or more					⁴³	⁴⁵
Self-monitoring via computer software		³³
Computerized goal setting, motivational message, self-monitoring					³³ ^, ³⁵
Number of physical activity sessions
Self-monitoring, motivational messages, goal setting
a. Under 6 months					⁴³
b. 6 months or more					⁴³
Metabolic equivalents
Self-monitoring, motivational messages, goal setting						⁴³

Open in a new tab

Effectiveness refers to statistically significant differences between intervention and comparison groups.

Table 4.

Reported effectiveness of portable activity sensing devices (PASDs) on outcomes besides physical activity, organized by category of employed behavioral change techniques

Outcomes and Behavioral Interventions	PASD with Passive Self-Monitoring		PASD with Goal-Based Self-Monitoring		PASD with Integrative Intervention
Outcomes and Behavioral Interventions	Improved	No change	Improved	No change	Improved	No change
Body mass index
Self-monitoring, motivational messages, goal setting					⁴⁶	⁴³^,⁴⁴
Self-monitoring via computer software		³⁴
Computerized goal setting, motivational message, self-monitoring					⁵¹	³⁴
Hip circumference
Self-monitoring, motivational messages, goal setting					⁴⁶
Computerized goal setting, motivational message, self-monitoring						³⁵
Waist circumference
Self-monitoring, motivational messages, goal setting					⁴⁶	⁴³
Self-monitoring via computer software		³⁴
Computerized goal setting, motivational message, self-monitoring
a. Under 6 months					³⁵
b. 6 months or more						³⁴
Body fat percentage
Computerized goal setting, motivational message, self-monitoring					³⁵	³⁴
Systolic and diastolic blood pressure
Self-monitoring, motivational messages, goal setting						⁴⁴
Computerized goal setting, motivational message, self-monitoring						³⁵^,⁵¹
HbA1c
Self-monitoring, motivational messages, goal setting						⁴⁴
Computerized goal setting, motivational message, self-monitoring					³⁵
Serum lipid profile
Computerized goal setting, motivational message, self-monitoring						³⁵^,⁵¹
Framingham 10-year risk
Computerized goal setting, motivational message, self-monitoring						³⁵
Fasting insulin level
Computerized goal setting, motivational message, self-monitoring					³⁵
Fasting glucose serum level
Computerized goal setting, motivational message, self-monitoring						³⁵
Weight loss
Self-monitoring, motivational messages, goal setting					⁴⁶
Self-monitoring via computer software		³⁴
Computerized goal setting, motivational message, self-monitoring
a. Under 6 months					³⁵	³⁴
b. 6 months or more					³⁴^,⁵¹
Insulin total daily dose reduction
Self-monitoring, motivational messages, goal setting						⁴⁴
Balance control, postural stability, and gait performance
Computerized balance training program, self-monitoring	²⁸^,²⁹^,⁵²
Quality of life
Self-monitoring, motivational messages, goal setting						⁴⁴^,⁴⁵
Computerized balance training program, self-monitoring	²⁸	²⁹^,⁵²
Smoking behavior
Self-monitoring, goal setting				⁴⁷
Computerized goal setting, motivational message, self-monitoring						⁵¹
Lung function
Self-monitoring, goal setting				⁴⁷
Lower extremities function
Self-monitoring, motivational messages, goal setting						⁴⁵

Open in a new tab

Effectiveness refers to statistically significant differences between intervention and comparison groups.

Physical activity

Merely providing objective feedback of step counts and activity level to participants for self-monitoring purposes did not effectively increase physical activity level and duration (Table 3).³³^,⁴² Kovelis et al.⁴⁷ reported similar results when adding goal setting along with objective step counts for physically inactive smokers, concluding that the interventions were not sufficient to motivate physical activity.

Use of 3 or more behavioral change techniques in the integrative interventions was associated with more promising, though still mixed, evidence for improved physical activity levels. Six of 10 trials reported such improvements with a combination of PASD, motivational messages, exercise plans, and objective activity feedback³³^,³⁵^,⁴⁵^,⁴⁶^,⁴⁸^,⁴⁹; 4 reported no effect.³²^,⁴⁴^,⁴⁹^,⁵⁰ In a study by Bickmore et al.⁴⁹ on the effect of a virtual coaching computer program with individualized exercise goals and motivational messages, the intervention was successful at 2 months for increasing daily step counts, but the effect did not last after 1 year. Martin et al.³³ also found no significant effect on activity level and duration when participants received only device feedback; however, there was a significant increase in step counts per day and duration of exercise after the group received motivational text messages, planned exercise goals, and activity-level feedback compared to an attention control group using a device with no access to device data. In contrast, Newton et al.⁴⁴ reported that weekly text messages to participants reminding them to wear the pedometer and stay active did not improve either activity level or amount of exercise. McMurdo et al.⁴⁵ found an increase in physical activity at 6 months among sedentary elderly women by providing individualized activity plans and motivational messages but no accelerometer.

Three studies reported that interventions with 3 or more behavioral change techniques increased the duration of physical activity³³^,³⁵^,⁴³; 3 others reported no effect.³²^,⁴⁴^,⁴⁵ For example, Butler et al.⁴³ showed a long-term effect of integrative interventions among adult patients participating in cardiac rehabilitation programs, whereas McMurdo et al.⁴⁵ reported no effect in sedentary elderly women after 6 months. In the 1 study assessing reported exercise tolerance in metabolic equivalents, the intervention was not effective.⁴³

Body measures

Three trials concluded that an intervention using PASDs does not affect body mass index,³⁴^,⁴³^,⁴⁴ while 2 studies reported body mass index improvements in postpartum women⁴⁶ and obese adults (Table 4).⁵¹

Studies reported mixed outcomes of PASDs incorporating motivational messages, goal setting, and objective activity feedback on hip circumference. Maturi et al.⁴⁶ found the intervention effective among postpartum women, and Wijsman et al.³⁵ reported positive results in sedentary older adults. Two studies showed no effect on waist circumference.³⁴^,⁴³

Physiological measures

Few studies evaluated the impact of PASD interventions on physiological measures, and results were mixed (Table 4). Shuger et al.³⁴ reported no reduction in body fat percentage after providing objective activity feedback alongside motivational messages and exercise planning to sedentary obese adults, but Wijsman et al.³⁵ reported a positive change among sedentary older adults. Wijsman et al.³⁵ also found improvement in HbA1c, while Newton et al.⁴⁴ observed no change in HbA1c among type 1 diabetes patients. One study also found that a computerized intervention with motivational messages, when combined with self-monitoring and individualized goal setting, improved fasting serum insulin levels but had no effect on fasting serum glucose levels.³⁵ Other trials reported no effect on blood pressure,³⁵^,⁴⁴^,⁵¹ serum lipid profile,³⁵^,⁵¹ or Framingham 10-year risk of heart attack.³⁵

Weight loss

Four reports showed that use of PASDs could help patients lose weight (Table 4).³⁴^,³⁵^,⁴⁶^,⁵¹ The integrative interventions of PASD with self-monitoring, motivational messages, and goal setting were successful even without providing software.⁴⁶ Wijsman et al.³⁵ and Oh et al.⁵¹ studied the impact of computerized self-monitoring programs with individualized exercise planning and motivational feedback on weight in sedentary elders and obese adults with metabolic syndrome. Participants in the intervention groups lost, on average, 1.5 kg of weight after 3 months³⁵ and 2.2 kg after 6 months,⁵¹ significantly more than control group participants. Maturi et al.⁴⁶ conducted similar interventions, but without IT software, with postpartum women and found a significant mean weight decrease of 2.1 kg after 3 months. Shuger et al.³⁴ studied 2 types of interventions in the short and long term with physically inactive obese adults. They reported that providing activity-level feedback alone did not result in weight loss; however, adding computerized real-time self-monitoring along with group-based education resulted in an average 9-month weight decrease of 6.59 kg.

Balance control, postural stability, and gait performance

All 3 studies investigating the effect of IMUs on balance control reported positive change (Table 4).²⁸^,²⁹^,⁵² Grewal et al.²⁸ and Schwenk et al.²⁹ examined the same computerized system on patients with diabetic peripheral neuropathy and CIPN. By teaching patients better motor and posture control, the intervention resulted in significant reduction of hip, ankle, and center-of-mass sway in both groups. In another study by Ginis et al.,⁵² 2 smartphone applications provided real-time feedback on gait abnormalities and freezing-of-gait occurrence alerts to patients with Parkinson’s disease. They concluded that the intervention improved balance and gait performance compared to conventional gait training for people with Parkinson’s disease.

Quality of life

No improvement in quality of life was observed in 2 trials investigating the effect of pedometers and accelerometers within a behavioral intervention.⁴⁴^,⁴⁵ Grewal et al.²⁸ found that their computerized balance training program and IMU measuring device might improve quality of life in diabetic peripheral neuropathy patients, but not in CIPN patients.²⁹ Ginis et al.⁵² reported no effect of a gait-alerting smartphone-based application on quality of life among patients with Parkinson’s disease (Table 4).

Other outcomes

Only 1 study evaluated the effect of interventions on total daily insulin dose and found no significant impact.⁴⁴ No improvement in smoking behavior was reported by 2 studies.⁴⁷^,⁵¹ Another study evaluating the effect of self-monitoring plus goal-setting on lung function concluded that there was no improvement among smokers.⁴⁷ McMurdo et al.⁴⁵ found no improvement in lower extremity function among sedentary older women (Table 4).

Risk of bias in included studies

We evaluated study quality by assessing the risk of bias in the 17 included studies (Figures 4 and 5 ). Randomization was adequate in 14 studies (82%),²⁸^,²⁹^,^32–35^,⁴²^,⁴³^,^45–48^,⁵⁰^,⁵¹ but the randomization method was unclear in 3 others. Allocation was concealed in 10 studies (59%),²⁸^,²⁹^,³²^,³³^,^45–48^,⁵⁰^,⁵² but other studies did not provide enough information to assess allocation concealment. Most studies were not successful in blinding participants and research personnel to the allocated interventions.²⁸^,²⁹^,³²^,³⁴^,³⁵^,^42–51 Only 1 study³³ clearly stated that participants and research personnel were blinded; blinding was unclear in another.⁵² Blinding the assessors of outcomes was achieved in 7 studies (41%).²⁹^,^32–35^,⁴²^,⁴⁹ Seven trials (41%)^44–48^,^50–52 explicitly stated that they did not blind research personnel, and 2 (12%) did not clearly report this.²⁸^,⁴³ The completeness of outcome data was adequate in 16 trials (94%), whereas 1 study could not sufficiently collect outcomes due to participant nonadherence.⁴⁷ With respect to selective reporting, we identified 5 trials (29%) with available study protocols reporting all preselected outcomes^32–34^,⁴⁵^,⁴⁸; however, not enough information was available to judge selective reporting bias in other studies. In assessing other potential sources of bias, we identified 15 trials (88%) at risk of bias due to small sample size and lack of generalizability.²⁸^,²⁹^,^32–34^,^42–44^,^47–50^,⁵² Some of the trials were also at high risk of selection bias due to self-selected participants⁴²^,⁴³ or imbalanced education level between intervention and control groups.⁵¹ One study had adequate statistical power and was deemed to be free of other bias.⁴⁵

Figure 4. — Summary of authors’ consensus judgment about of risk of bias for each included study, by various sources of potential bias.

Figure 5. — Risk of bias graph based on authors’ judgment about risk of bias items, presented as percentages across all included studies.

DISCUSSION

Findings on the health-related impact of PASDs are highly diverse, but the majority of research supports a benefit from combining sensors with an integrative set of behavioral change techniques. In contrast to more passive interventions, where the burden is on participants to meaningfully use device data, integrative interventions support the active use of data through training or coaching, motivational feedback, and other behavioral change techniques alongside self-monitoring and goal setting.

The importance of behavioral change techniques

One implication of these findings is that passive interventions – ones where people receive a PASD with minimal further assistance, reinforcement, or additional behavioral change techniques – may not be effective. In other words, the device alone may be insufficient to effect a change in health. This echoes a general concern about the use of health IT as a stand-alone intervention and implies that there is a need to better explicate the role of health IT in behavioral theory.⁵³^,⁵⁴

In contrast, when paired with an appropriate array of behavioral change techniques, PASDs can produce benefits, particularly for intermediate outcomes such as physical activity or weight loss. Even so, it is unclear whether the PASD itself contributes to these outcomes beyond the behavioral change techniques. A recent study in which participants receiving an integrative weight-loss intervention were randomized to receive a PASD or continue the intervention with no PASD found no long-term differences between the 2 groups.⁵⁵ However, further research is needed to test the hypothesis that PASDs provide independent value or interact with behavioral change techniques to amplify their effectiveness. Leveraging interactive and adaptive mobile applications, in particular, might be an appropriate way to investigate the effect of health behavior interventions.⁵⁶

The role of behavioral informatics

In the reviewed studies, IT software often provided a convenient and efficient way of delivering interventions with techniques such as feedback and motivational messaging. Such behavioral informatics interventions were often, though not always, effective at increasing physical activity levels, activity duration, and weight loss. Given the small number of trials and the heterogeneity of software used, it is premature to make any conclusions about the value of IT software in addition to PASD hardware. Other studies report favorable but mixed findings in the literature concerning the effects of behavioral IT interventions, including mobile technology and text-messaging, on health outcomes.⁵⁷ While several studies have reported improved HbA1c in patients with type 2 diabetes,⁵⁸ adherence to blood glucose measurement and glycemic control in type 1 diabetic adults,⁵⁹^,⁶⁰ smoking cessation behavior,⁶¹^,⁶² nutrition education and healthy diet,⁶²^,⁶³ and psychological outcomes,⁶² other studies did not report benefits of such interventions for physical activity,⁶²^,⁶⁴ weight loss,⁶² and dietary behavior.⁶⁵ This could be the result of the health IT design approaches employed by those studies, with successful health IT being more likely to follow user-centered design principles and therefore yielding higher usability, user satisfaction, acceptability, and performance.⁶⁶ For example, systems produced through user-centered design have been effective at improving quality of life in HIV patients⁶⁷ and self-management behaviors among lung transplant recipients.⁶⁸ Further research should investigate specific software characteristics as well as software design approaches, eg, user-centered design,⁶⁶^,⁶⁹^,⁷⁰ that promote effectiveness of PASD interventions.

Health-related outcomes

Notably, few PASD intervention studies assessed downstream clinical outcomes such as organ function and quality of life, and none found an effect of PASD interventions. This generally conforms with the strength of evidence for intermediate as opposed to endpoint effectiveness in the overall literature on consumer-facing health IT.⁷¹^,⁷² More research is needed to assess downstream outcomes and specific PASD intervention approaches most likely to improve clinical outcomes, such as hospitalizations, disease onset, and health-related quality of life.

Quality of reviewed randomized controlled trials

Most examined trials did not follow CONSORT guidelines,⁷³ the globally accepted standard for efficiently and accurately reporting randomized controlled trial results. Although most reviewed trials reported randomization and measured outcomes, reporting quality for other categories was not acceptable. Participant demographics were poorly reported; only 24% of trials reported age distribution,²⁸^,²⁹^,³²^,⁴² 1 study did not specify gender distribution,⁵² and race/ethnicity information was minimal in 24% of trials.³³^,^48–50 Poor reporting of key methodological elements may not directly indicate the quality of the trials but limits the ability to assess the validity and generalizability of findings.^73–75 We therefore urge researchers to adopt CONSORT guidelines in reporting randomized controlled trials of PASD effectiveness.^76–78

Limitations

It is possible that eligible trials were inadvertently excluded from the review. The literature lacks accepted terminology for PASDs and PASD research. PASDs are also referred to as wearable devices, activity monitors, activity sensors, portable devices, wireless activity sensors, and wearable motion detectors. To mitigate the risk of excluding eligible trials, our search queries used these and other search terms, including the names of common PASDs (eg, Actigraph, Fitbit). We recommend using the term “PASD,” which refers to the device’s primary function (activity sensing) and includes portable devices that are not wearable (eg, phones or devices carried in a pocket or purse). Further, only 1 reviewer screened the publications, increasing the risk of selection bias. However, a second reviewer monitored the process for accuracy to mitigate this risk. Classifying interventions by the nature of behavioral change techniques and IT software integration (Figure 3) is novel to this review and may not accurately represent the taxonomy of PASD interventions. However, there was no other existing taxonomy, and ours was consistent with the intervention groups used in individual studies, which differed by the number of behavioral change techniques and use of software.²⁸^,²⁹^,^32–35^,^42–52 Lastly, the outcomes of reviewed studies were too heterogeneous to perform quantitative meta-analysis of the effectiveness of PASD interventions.

Future work

Future research should assess the impact of specific PASD characteristics – including hardware, software, and data transfer and processing elements – on outcomes. Study designs should permit disambiguation of the effect of the PASD itself vs other elements of the intervention, including behavioral change techniques. Studies can also test the hypothesis that integration of PASD with behavioral change techniques is more effective than use of the PASD or behavioral change techniques alone. An important future direction is the study of longitudinal effects, especially given evidence of a lack of lasting effects of PASD interventions.⁴⁹ This will also permit assessment of downstream outcomes such as disease onset or resolution.

CONCLUSION

There is insufficient volume or quality of evidence to conclude that PASD interventions are generally beneficial for health. However, PASD interventions might be effective in improving intermediate outcomes when coupled with multiple behavioral change techniques. Devices alone or with minimal behavioral change support appear to be insufficient to change health-related outcomes.

Funding

This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sector.

Competing Interests

Authors have no competing interests to declare.

Contributors

HA and RJH contributed to conceptualization, design, analysis, and writing the manuscript. HA performed study selection, data extraction, and quality assessment, monitored by RJH for completeness and accuracy.

SUPPLEMENTARY MATERIAL

Supplementary material is available at Journal of the American Medical Informatics Association online.

Supplementary Material

Supplementary Materials

Click here for additional data file.^{(46.6KB, docx)}

Supplementary Table 1

Click here for additional data file.^{(12.7KB, docx)}

Supplementary Table 2

Click here for additional data file.^{(14.5KB, docx)}

Supplementary Table 3

Click here for additional data file.^{(38KB, docx)}

References

1. Matthews CE, Hagstromer M, Pober DM et al. . Best practices for using physical activity monitors in population-based research. Med Sci Sports Exerc. 2012;44(1 Suppl 1):S68–76. [DOI] [PMC free article] [PubMed] [Google Scholar]
2. Case MA, Burwick HA, Volpp KG et al. . Accuracy of smartphone applications and wearable devices for tracking physical activity data. JAMA. 2015;3136:625–26. [DOI] [PubMed] [Google Scholar]
3. Mathie MJ, Coster AC, Lovell NH et al. . Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement. Physiol Meas. 2004;252:R1–20. [DOI] [PubMed] [Google Scholar]
4. Tudor-Locke C, Camhi SM, Troiano RP. A catalog of rules, variables, and definitions applied to accelerometer data in the National Health and Nutrition Examination Survey, 2003–2006. Prev Chronic Dis. 2012;9:E113. [DOI] [PMC free article] [PubMed] [Google Scholar]
5. Ainsworth B, Cahalin L, Buman M et al. . The current state of physical activity assessment tools. Prog Cardiovasc Dis. 2015;574:387–95. [DOI] [PubMed] [Google Scholar]
6. Prince SA, Adamo KB, Hamel ME et al. . A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008;51:56. [DOI] [PMC free article] [PubMed] [Google Scholar]
7. Kumar S, Nilsen WJ, Abernethy A et al. . Mobile health technology evaluation: the mHealth evidence workshop. Am J Prev Med. 2013;452:228–36. [DOI] [PMC free article] [PubMed] [Google Scholar]
8. Varshney U. Pervasive healthcare and wireless health monitoring. Mobile Networks Appl. 2007;12(2–3):113–27. [Google Scholar]
9. Loo Gee B, Griffiths KM, Gulliver A. Effectiveness of mobile technologies delivering Ecological Momentary Interventions for stress and anxiety: a systematic review. J Am Med Inform Assoc. 2016;231:221–29. [DOI] [PMC free article] [PubMed] [Google Scholar]
10. Piwek L, Ellis DA, Andrews S et al. . The Rise of Consumer Health Wearables: Promises and Barriers. PLoS Med. 2016;132:e1001953. [DOI] [PMC free article] [PubMed] [Google Scholar]
11. Gandhi M, Wang T. Digital Health Consumer Adoption. 2015. https://rockhealth.com/reports/digital-health-consumer-adoption-2015. Accessed October 9, 2016.
12. Krebs P, Duncan DT. Health app use among US mobile phone owners: a national survey. JMIR Mhealth Uhealth. 2015;34:e101. [DOI] [PMC free article] [PubMed] [Google Scholar]
13. American Heart Association Nutrition C, Lichtenstein AH, Appel LJ et al. . Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation. 2006;1141:82–96. [DOI] [PubMed] [Google Scholar]
14. Hu FB, Leitzmann MF, Stampfer MJ et al. . Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med. 2001;16112:1542–48. [DOI] [PubMed] [Google Scholar]
15. Wing RR, Goldstein MG, Acton KJ et al. . Behavioral science research in diabetes: lifestyle changes related to obesity, eating behavior, and physical activity. Diabetes Care. 2001;241:117–23. [DOI] [PubMed] [Google Scholar]
16. Despres J-P, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2006;4447121:881–87. [DOI] [PubMed] [Google Scholar]
17. Marchand LL, Wilkens LR, Kolonel LN et al. . Associations of sedentary lifestyle, obesity, smoking, alcohol use, and diabetes with the risk of colorectal cancer. Cancer Res. 1997;5721:4787–94. [PubMed] [Google Scholar]
18. Fader AN, Arriba LN, Frasure HE et al. . Endometrial cancer and obesity: epidemiology, biomarkers, prevention and survivorship. Gynecol Oncol. 2009;1141:121–27. [DOI] [PubMed] [Google Scholar]
19. Rovio S, Kåreholt I, Helkala E-L et al. . Leisure-time physical activity at midlife and the risk of dementia and Alzheimer’s disease. Lancet Neurol. 2005;411:705–11. [DOI] [PubMed] [Google Scholar]
20. Baezner H, Blahak C, Poggesi A et al. . Association of gait and balance disorders with age-related white matter changes: the LADIS study. Neurology. 2008;7012:935–42. [DOI] [PubMed] [Google Scholar]
21. Ensrud KE, Nevitt MC, Yunis C et al. . Correlates of impaired function in older women. J Am Geriatr Soc. 1994;425:481–89. [DOI] [PubMed] [Google Scholar]
22. Gregg EW, Pereira MA, Caspersen CJ. Physical activity, falls, and fractures among older adults: a review of the epidemiologic evidence. J Am Geriatr Soc. 2000;488:883–93. [DOI] [PubMed] [Google Scholar]
23. Bassett DR. Device-based monitoring in physical activity and public health research. Physiol Meas. 2012;3311:1769–83. [DOI] [PubMed] [Google Scholar]
24. Pavel M, Jimison HB, Korhonen I et al. . Behavioral informatics and computational modeling in support of proactive health management and care. IEEE Trans Biomed Eng. 2015;6212:2763–75. [DOI] [PMC free article] [PubMed] [Google Scholar]
25. Bravata DM, Smith-Spangler C, Sundaram V et al. . Using pedometers to increase physical activity and improve health: a systematic review. JAMA. 2007;29819:2296–304. [DOI] [PubMed] [Google Scholar]
26. Pitta F, Troosters T, Probst VS et al. . Quantifying physical activity in daily life with questionnaires and motion sensors in COPD. Eur Respir J. 2006;275:1040–55. [DOI] [PubMed] [Google Scholar]
27. Sveistrup H. Motor rehabilitation using virtual reality. J Neuroeng Rehabil. 2004;11:10. [DOI] [PMC free article] [PubMed] [Google Scholar]
28. Grewal GS, Schwenk M, Lee-Eng J et al. . Sensor-based interactive balance training with visual joint movement feedback for improving postural stability in diabetics with peripheral neuropathy: a randomized controlled trial. Gerontology. 2015;616:567–74. [DOI] [PubMed] [Google Scholar]
29. Schwenk M, Grewal GS, Holloway D et al. . Interactive sensor-based balance training in older cancer patients with chemotherapy-induced peripheral neuropathy: a randomized controlled trial. Gerontology. 2016;625: 553–63. [DOI] [PMC free article] [PubMed] [Google Scholar]
30. Mercer K, Li M, Giangregorio L et al. . Behavior change techniques present in wearable activity trackers: a critical analysis. JMIR Mhealth Uhealth. 2016;42:e40. [DOI] [PMC free article] [PubMed] [Google Scholar]
31. Lyons EJ, Lewis ZH, Mayrsohn BG et al. . Behavior change techniques implemented in electronic lifestyle activity monitors: a systematic content analysis. J Med Internet Res. 2014;168:e192. [DOI] [PMC free article] [PubMed] [Google Scholar]
32. Mansfield A, Wong JS, Bryce J et al. . Use of accelerometer-based feedback of walking activity for appraising progress with walking-related goals in inpatient stroke rehabilitation: a randomized controlled trial. Neurorehabil Neural Repair. 2015;299:847–57. [DOI] [PubMed] [Google Scholar]
33. Martin SS, Feldman DI, Blumenthal RS et al. . mActive: a randomized clinical trial of an automated mHealth intervention for physical activity promotion. J Am Heart Assoc. 2015;411:e002239. [DOI] [PMC free article] [PubMed] [Google Scholar]
34. Shuger SL, Barry VW, Sui X et al. . Electronic feedback in a diet- and physical activity-based lifestyle intervention for weight loss: a randomized controlled trial. Int J Behav Nutr Phys. Act 2011;8:41. [DOI] [PMC free article] [PubMed] [Google Scholar]
35. Wijsman CA, Westendorp RG, Verhagen EA et al. . Effects of a web-based intervention on physical activity and metabolism in older adults: randomized controlled trial. J Med Internet Res. 2013;1511:e233. [DOI] [PMC free article] [PubMed] [Google Scholar]
36. Tudor-Locke C, Williams JE, Reis JP et al. . Utility of pedometers for assessing physical activity: convergent validity. Sports Med. 2002;3212: 795–808. [DOI] [PubMed] [Google Scholar]
37. de Vries SI, Bakker I, Hopman-Rock M et al. . Clinimetric review of motion sensors in children and adolescents. J Clin Epidemiol. 2006;597: 670–80. [DOI] [PubMed] [Google Scholar]
38. de Bruin ED, Hartmann A, Uebelhart D et al. . Wearable systems for monitoring mobility-related activities in older people: a systematic review. Clin Rehabil. 2008;22(10–11):878–95. [DOI] [PubMed] [Google Scholar]
39. Schaefer SE, Van Loan M, German JB. A feasibility study of wearable activity monitors for pre-adolescent school-age children. Prev Chronic Dis. 2014;11:E85. [DOI] [PMC free article] [PubMed] [Google Scholar]
40. Higgins JP, Altman DG, Gotzsche PC et al. . The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. [DOI] [PMC free article] [PubMed] [Google Scholar]
41. Higgins J, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]: The Cochrane Collaboration; 2011. www.cochrane-handbook.org. Accessed February 25, 2016. [Google Scholar]
42. Paschali AA, Goodrick GK, Kalantzi-Azizi A et al. . Accelerometer feedback to promote physical activity in adults with type 2 diabetes: a pilot study. Percept Mot Skills. 2005;1001:61–68. [DOI] [PubMed] [Google Scholar]
43. Butler L, Furber S, Phongsavan P et al. . Effects of a pedometer-based intervention on physical activity levels after cardiac rehabilitation: a randomized controlled trial. J Cardiopulm Rehabil Prev. 2009;292:105–14. [DOI] [PubMed] [Google Scholar]
44. Newton KH, Wiltshire EJ, Elley CR. Pedometers and text messaging to increase physical activity: randomized controlled trial of adolescents with type 1 diabetes. Diabetes Care. 2009;325:813–15. [DOI] [PMC free article] [PubMed] [Google Scholar]
45. McMurdo MET, Sugden J, Argo I et al. . Do pedometers increase physical activity in sedentary older women? A randomized controlled trial. J Am Geriatr Soc. 2010;5811:2099–106. [DOI] [PubMed] [Google Scholar]
46. Maturi MS, Afshary P, Abedi P. Effect of physical activity intervention based on a pedometer on physical activity level and anthropometric measures after childbirth: a randomized controlled trial. BMC Pregnancy Childbirth. 2011;11:103. [DOI] [PMC free article] [PubMed] [Google Scholar]
47. Kovelis D, Zabatiero J, Furlanetto KC et al. . Short-term effects of using pedometers to increase daily physical activity in smokers: a randomized trial. Respir Care. 2012;577:1089–97. [DOI] [PubMed] [Google Scholar]
48. Adams MA, Sallis JF, Norman GJ et al. . An adaptive physical activity intervention for overweight adults: a randomized controlled trial. PLoS One. 2013;812:e82901. [DOI] [PMC free article] [PubMed] [Google Scholar]
49. Bickmore TW, Silliman RA, Nelson K et al. . A randomized controlled trial of an automated exercise coach for older adults. J Am Geriatr Soc. 2013;6110:1676–83. [DOI] [PubMed] [Google Scholar]
50. Choi J, Lee JH, Vittinghoff E et al. . mHealth physical activity intervention: a randomized pilot study in physically inactive pregnant women. Matern Child Health J. 2016;205:1091–101. [DOI] [PMC free article] [PubMed] [Google Scholar]
51. Oh B, Cho B, Han MK et al. . The effectiveness of mobile phone-based care for weight control in metabolic syndrome patients: randomized controlled trial. JMIR Mhealth Uhealth. 2015;33:e83. [DOI] [PMC free article] [PubMed] [Google Scholar]
52. Ginis P, Nieuwboer A, Dorfman M et al. . Feasibility and effects of home-based smartphone-delivered automated feedback training for gait in people with Parkinson’s disease: A pilot randomized controlled trial. Parkinsonism Relat Disord. 2016;22:28–34. [DOI] [PubMed] [Google Scholar]
53. Sawesi S, Rashrash M, Phalakornkule K et al. . The impact of information technology on patient engagement and health behavior change: a systematic review of the literature. JMIR Med Inform. 2016;41:e1. [DOI] [PMC free article] [PubMed] [Google Scholar]
54. Kukafka R, Johnson SB, Linfante A et al. . Grounding a new information technology implementation framework in behavioral science: a systematic analysis of the literature on IT use. J Biomed Inform. 2003;363:218–27. [DOI] [PubMed] [Google Scholar]
55. Jakicic JM, Davis KK, Rogers RJ et al. . Effect of wearable technology combined with a lifestyle intervention on long-term weight loss: The IDEA randomized clinical trial. JAMA. 2016;31611:1161–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
56. Riley WT, Rivera DE, Atienza AA et al. . Health behavior models in the age of mobile interventions: are our theories up to the task? Transl Behav Med. 2011;11:53–71. [DOI] [PMC free article] [PubMed] [Google Scholar]
57. Free C, Phillips G, Galli L et al. . The effectiveness of mobile-health technology-based health behaviour change or disease management interventions for health care consumers: a systematic review. PLoS Med. 2013;101:e1001362. [DOI] [PMC free article] [PubMed] [Google Scholar]
58. Quinn CC, Clough SS, Minor JM et al. . WellDoc mobile diabetes management randomized controlled trial: change in clinical and behavioral outcomes and patient and physician satisfaction. Diabetes Technol Ther. 2008;103:160–68. [DOI] [PubMed] [Google Scholar]
59. Kirwan M, Vandelanotte C, Fenning A et al. . Diabetes Self-Management Smartphone Application for Adults With Type 1 Diabetes: Randomized Controlled Trial. J Med Internet Res. 2013;1511:e235. [DOI] [PMC free article] [PubMed] [Google Scholar]
60. Cafazzo AJ, Casselman M, Hamming N et al. . Design of an mHealth app for the self-management of adolescent type 1 diabetes: a pilot study. J Med Internet Res. 2012;143:e70. [DOI] [PMC free article] [PubMed] [Google Scholar]
61. Krebs P, Prochaska JO, Rossi JS. A meta-analysis of computer-tailored interventions for health behavior change. Prev Med. 2010;51(3–4):214–21. [DOI] [PMC free article] [PubMed] [Google Scholar]
62. Portnoy DB, Scott-Sheldon LAJ, Johnson BT et al. . Computer-delivered interventions for health promotion and behavioral risk reduction: a meta-analysis of 75 randomized controlled trials, 1988–2007. Prev Med. 2008;471:3–16. [DOI] [PMC free article] [PubMed] [Google Scholar]
63. Brug J, Campbell M, van Assema P. The application and impact of computer-generated personalized nutrition education: a review of the literature. Patient Educ Couns. 1999;362:145–56. [DOI] [PubMed] [Google Scholar]
64. Neville LM, O’Hara B, Milat A. Computer-tailored physical activity behavior change interventions targeting adults: a systematic review. Int J Behav Nutr Phys Act. 2009;61:30. [DOI] [PMC free article] [PubMed] [Google Scholar]
65. Norman GJ, Zabinski MF, Adams MA et al. . A review of eHealth interventions for physical activity and dietary behavior change. Am J Prev Med. 2007;334:336–45. [DOI] [PMC free article] [PubMed] [Google Scholar]
66. Johnson CM, Johnson TR, Zhang J. A user-centered framework for redesigning health care interfaces. J Biomed Inform. 2005;381:75–87. [DOI] [PubMed] [Google Scholar]
67. Gustafson DH, Hawkins R, Boberg E et al. . Impact of a patient-centered, computer-based health information/support system. Am J Prev Med. 1999;161:1–9. [DOI] [PubMed] [Google Scholar]
68. De Vito Dabbs A, Song MK, Myers BA et al. . A randomized controlled trial of a mobile health intervention to promote self-management after lung transplantation. Am J Transplant. 2016;167: 2172–80. [DOI] [PMC free article] [PubMed] [Google Scholar]
69. Preece J, Rogers Y, Sharp H. Interaction Design: Beyond Human-Computer Interaction. New York: Wiley; 2015. [Google Scholar]
70. De Vito Dabbs A, Myers BA, Mc Curry KR et al. . User-centered design and interactive health technologies for patients. Comput Inform Nurs. 2009;273:175–83. [DOI] [PMC free article] [PubMed] [Google Scholar]
71. Finkelstein J, Knight A, Marinopoulos S et al. . Enabling patient-centered care through health information technology. Evid Rep Technol Assess (Full Rep). 2012;206:1–1531. [PMC free article] [PubMed] [Google Scholar]
72. Gibbons M, Samal RFWL et al. . Impact of Consumer Health Informatics Applications. Evidence Report/Technology Assessment No. 188. Rockville, MD: AHRQ Publication No. 09(10)-E019; 2009. [PMC free article] [PubMed] [Google Scholar]
73. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMC Med. 2010;81:18. [DOI] [PMC free article] [PubMed] [Google Scholar]
74. Soares HP, Daniels S, Kumar A et al. . Bad reporting does not mean bad methods for randomised trials: observational study of randomised controlled trials performed by the Radiation Therapy Oncology Group. BMJ. 2004;3287430:22. [DOI] [PMC free article] [PubMed] [Google Scholar]
75. Huwiler-Müntener K, Jüni P, Junker C et al. . Quality of reporting of randomized trials as a measure of methodologic quality. JAMA. 2002;28721:2801–04. [DOI] [PubMed] [Google Scholar]
76. Plint AC, Moher D, Morrison A et al. . Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic review. Med J Aust. 2006;1855:263–67. [DOI] [PubMed] [Google Scholar]
77. Moher D, Jones A, Lepage L. Use of the consort statement and quality of reports of randomized trials: a comparative before-and-after evaluation. JAMA. 2001;28515:1992–95. [DOI] [PubMed] [Google Scholar]
78. Kane RL, Wang J, Garrard J. Reporting in randomized clinical trials improved after adoption of the CONSORT statement. J Clin Epidemiol. 2007;603:241–49. [DOI] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Click here for additional data file.^{(46.6KB, docx)}

Supplementary Table 1

Click here for additional data file.^{(12.7KB, docx)}

Supplementary Table 2

Click here for additional data file.^{(14.5KB, docx)}

Supplementary Table 3

Click here for additional data file.^{(38KB, docx)}

[ocx006-B1] 1. Matthews CE, Hagstromer M, Pober DM et al. . Best practices for using physical activity monitors in population-based research. Med Sci Sports Exerc. 2012;44(1 Suppl 1):S68–76. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B2] 2. Case MA, Burwick HA, Volpp KG et al. . Accuracy of smartphone applications and wearable devices for tracking physical activity data. JAMA. 2015;3136:625–26. [DOI] [PubMed] [Google Scholar]

[ocx006-B3] 3. Mathie MJ, Coster AC, Lovell NH et al. . Accelerometry: providing an integrated, practical method for long-term, ambulatory monitoring of human movement. Physiol Meas. 2004;252:R1–20. [DOI] [PubMed] [Google Scholar]

[ocx006-B4] 4. Tudor-Locke C, Camhi SM, Troiano RP. A catalog of rules, variables, and definitions applied to accelerometer data in the National Health and Nutrition Examination Survey, 2003–2006. Prev Chronic Dis. 2012;9:E113. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B5] 5. Ainsworth B, Cahalin L, Buman M et al. . The current state of physical activity assessment tools. Prog Cardiovasc Dis. 2015;574:387–95. [DOI] [PubMed] [Google Scholar]

[ocx006-B6] 6. Prince SA, Adamo KB, Hamel ME et al. . A comparison of direct versus self-report measures for assessing physical activity in adults: a systematic review. Int J Behav Nutr Phys Act. 2008;51:56. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B7] 7. Kumar S, Nilsen WJ, Abernethy A et al. . Mobile health technology evaluation: the mHealth evidence workshop. Am J Prev Med. 2013;452:228–36. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B8] 8. Varshney U. Pervasive healthcare and wireless health monitoring. Mobile Networks Appl. 2007;12(2–3):113–27. [Google Scholar]

[ocx006-B9] 9. Loo Gee B, Griffiths KM, Gulliver A. Effectiveness of mobile technologies delivering Ecological Momentary Interventions for stress and anxiety: a systematic review. J Am Med Inform Assoc. 2016;231:221–29. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B10] 10. Piwek L, Ellis DA, Andrews S et al. . The Rise of Consumer Health Wearables: Promises and Barriers. PLoS Med. 2016;132:e1001953. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B11] 11. Gandhi M, Wang T. Digital Health Consumer Adoption. 2015. https://rockhealth.com/reports/digital-health-consumer-adoption-2015. Accessed October 9, 2016.

[ocx006-B12] 12. Krebs P, Duncan DT. Health app use among US mobile phone owners: a national survey. JMIR Mhealth Uhealth. 2015;34:e101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B13] 13. American Heart Association Nutrition C, Lichtenstein AH, Appel LJ et al. . Diet and lifestyle recommendations revision 2006: a scientific statement from the American Heart Association Nutrition Committee. Circulation. 2006;1141:82–96. [DOI] [PubMed] [Google Scholar]

[ocx006-B14] 14. Hu FB, Leitzmann MF, Stampfer MJ et al. . Physical activity and television watching in relation to risk for type 2 diabetes mellitus in men. Arch Intern Med. 2001;16112:1542–48. [DOI] [PubMed] [Google Scholar]

[ocx006-B15] 15. Wing RR, Goldstein MG, Acton KJ et al. . Behavioral science research in diabetes: lifestyle changes related to obesity, eating behavior, and physical activity. Diabetes Care. 2001;241:117–23. [DOI] [PubMed] [Google Scholar]

[ocx006-B16] 16. Despres J-P, Lemieux I. Abdominal obesity and metabolic syndrome. Nature. 2006;4447121:881–87. [DOI] [PubMed] [Google Scholar]

[ocx006-B17] 17. Marchand LL, Wilkens LR, Kolonel LN et al. . Associations of sedentary lifestyle, obesity, smoking, alcohol use, and diabetes with the risk of colorectal cancer. Cancer Res. 1997;5721:4787–94. [PubMed] [Google Scholar]

[ocx006-B18] 18. Fader AN, Arriba LN, Frasure HE et al. . Endometrial cancer and obesity: epidemiology, biomarkers, prevention and survivorship. Gynecol Oncol. 2009;1141:121–27. [DOI] [PubMed] [Google Scholar]

[ocx006-B19] 19. Rovio S, Kåreholt I, Helkala E-L et al. . Leisure-time physical activity at midlife and the risk of dementia and Alzheimer’s disease. Lancet Neurol. 2005;411:705–11. [DOI] [PubMed] [Google Scholar]

[ocx006-B20] 20. Baezner H, Blahak C, Poggesi A et al. . Association of gait and balance disorders with age-related white matter changes: the LADIS study. Neurology. 2008;7012:935–42. [DOI] [PubMed] [Google Scholar]

[ocx006-B21] 21. Ensrud KE, Nevitt MC, Yunis C et al. . Correlates of impaired function in older women. J Am Geriatr Soc. 1994;425:481–89. [DOI] [PubMed] [Google Scholar]

[ocx006-B22] 22. Gregg EW, Pereira MA, Caspersen CJ. Physical activity, falls, and fractures among older adults: a review of the epidemiologic evidence. J Am Geriatr Soc. 2000;488:883–93. [DOI] [PubMed] [Google Scholar]

[ocx006-B23] 23. Bassett DR. Device-based monitoring in physical activity and public health research. Physiol Meas. 2012;3311:1769–83. [DOI] [PubMed] [Google Scholar]

[ocx006-B24] 24. Pavel M, Jimison HB, Korhonen I et al. . Behavioral informatics and computational modeling in support of proactive health management and care. IEEE Trans Biomed Eng. 2015;6212:2763–75. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B25] 25. Bravata DM, Smith-Spangler C, Sundaram V et al. . Using pedometers to increase physical activity and improve health: a systematic review. JAMA. 2007;29819:2296–304. [DOI] [PubMed] [Google Scholar]

[ocx006-B26] 26. Pitta F, Troosters T, Probst VS et al. . Quantifying physical activity in daily life with questionnaires and motion sensors in COPD. Eur Respir J. 2006;275:1040–55. [DOI] [PubMed] [Google Scholar]

[ocx006-B27] 27. Sveistrup H. Motor rehabilitation using virtual reality. J Neuroeng Rehabil. 2004;11:10. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B28] 28. Grewal GS, Schwenk M, Lee-Eng J et al. . Sensor-based interactive balance training with visual joint movement feedback for improving postural stability in diabetics with peripheral neuropathy: a randomized controlled trial. Gerontology. 2015;616:567–74. [DOI] [PubMed] [Google Scholar]

[ocx006-B29] 29. Schwenk M, Grewal GS, Holloway D et al. . Interactive sensor-based balance training in older cancer patients with chemotherapy-induced peripheral neuropathy: a randomized controlled trial. Gerontology. 2016;625: 553–63. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B30] 30. Mercer K, Li M, Giangregorio L et al. . Behavior change techniques present in wearable activity trackers: a critical analysis. JMIR Mhealth Uhealth. 2016;42:e40. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B31] 31. Lyons EJ, Lewis ZH, Mayrsohn BG et al. . Behavior change techniques implemented in electronic lifestyle activity monitors: a systematic content analysis. J Med Internet Res. 2014;168:e192. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B32] 32. Mansfield A, Wong JS, Bryce J et al. . Use of accelerometer-based feedback of walking activity for appraising progress with walking-related goals in inpatient stroke rehabilitation: a randomized controlled trial. Neurorehabil Neural Repair. 2015;299:847–57. [DOI] [PubMed] [Google Scholar]

[ocx006-B33] 33. Martin SS, Feldman DI, Blumenthal RS et al. . mActive: a randomized clinical trial of an automated mHealth intervention for physical activity promotion. J Am Heart Assoc. 2015;411:e002239. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B34] 34. Shuger SL, Barry VW, Sui X et al. . Electronic feedback in a diet- and physical activity-based lifestyle intervention for weight loss: a randomized controlled trial. Int J Behav Nutr Phys. Act 2011;8:41. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B35] 35. Wijsman CA, Westendorp RG, Verhagen EA et al. . Effects of a web-based intervention on physical activity and metabolism in older adults: randomized controlled trial. J Med Internet Res. 2013;1511:e233. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B36] 36. Tudor-Locke C, Williams JE, Reis JP et al. . Utility of pedometers for assessing physical activity: convergent validity. Sports Med. 2002;3212: 795–808. [DOI] [PubMed] [Google Scholar]

[ocx006-B37] 37. de Vries SI, Bakker I, Hopman-Rock M et al. . Clinimetric review of motion sensors in children and adolescents. J Clin Epidemiol. 2006;597: 670–80. [DOI] [PubMed] [Google Scholar]

[ocx006-B38] 38. de Bruin ED, Hartmann A, Uebelhart D et al. . Wearable systems for monitoring mobility-related activities in older people: a systematic review. Clin Rehabil. 2008;22(10–11):878–95. [DOI] [PubMed] [Google Scholar]

[ocx006-B39] 39. Schaefer SE, Van Loan M, German JB. A feasibility study of wearable activity monitors for pre-adolescent school-age children. Prev Chronic Dis. 2014;11:E85. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B40] 40. Higgins JP, Altman DG, Gotzsche PC et al. . The Cochrane Collaboration’s tool for assessing risk of bias in randomised trials. BMJ. 2011;343:d5928. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B41] 41. Higgins J, Green S, eds. Cochrane Handbook for Systematic Reviews of Interventions Version 5.1.0 [updated March 2011]: The Cochrane Collaboration; 2011. www.cochrane-handbook.org. Accessed February 25, 2016. [Google Scholar]

[ocx006-B42] 42. Paschali AA, Goodrick GK, Kalantzi-Azizi A et al. . Accelerometer feedback to promote physical activity in adults with type 2 diabetes: a pilot study. Percept Mot Skills. 2005;1001:61–68. [DOI] [PubMed] [Google Scholar]

[ocx006-B43] 43. Butler L, Furber S, Phongsavan P et al. . Effects of a pedometer-based intervention on physical activity levels after cardiac rehabilitation: a randomized controlled trial. J Cardiopulm Rehabil Prev. 2009;292:105–14. [DOI] [PubMed] [Google Scholar]

[ocx006-B44] 44. Newton KH, Wiltshire EJ, Elley CR. Pedometers and text messaging to increase physical activity: randomized controlled trial of adolescents with type 1 diabetes. Diabetes Care. 2009;325:813–15. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B45] 45. McMurdo MET, Sugden J, Argo I et al. . Do pedometers increase physical activity in sedentary older women? A randomized controlled trial. J Am Geriatr Soc. 2010;5811:2099–106. [DOI] [PubMed] [Google Scholar]

[ocx006-B46] 46. Maturi MS, Afshary P, Abedi P. Effect of physical activity intervention based on a pedometer on physical activity level and anthropometric measures after childbirth: a randomized controlled trial. BMC Pregnancy Childbirth. 2011;11:103. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B47] 47. Kovelis D, Zabatiero J, Furlanetto KC et al. . Short-term effects of using pedometers to increase daily physical activity in smokers: a randomized trial. Respir Care. 2012;577:1089–97. [DOI] [PubMed] [Google Scholar]

[ocx006-B48] 48. Adams MA, Sallis JF, Norman GJ et al. . An adaptive physical activity intervention for overweight adults: a randomized controlled trial. PLoS One. 2013;812:e82901. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B49] 49. Bickmore TW, Silliman RA, Nelson K et al. . A randomized controlled trial of an automated exercise coach for older adults. J Am Geriatr Soc. 2013;6110:1676–83. [DOI] [PubMed] [Google Scholar]

[ocx006-B50] 50. Choi J, Lee JH, Vittinghoff E et al. . mHealth physical activity intervention: a randomized pilot study in physically inactive pregnant women. Matern Child Health J. 2016;205:1091–101. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B51] 51. Oh B, Cho B, Han MK et al. . The effectiveness of mobile phone-based care for weight control in metabolic syndrome patients: randomized controlled trial. JMIR Mhealth Uhealth. 2015;33:e83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B52] 52. Ginis P, Nieuwboer A, Dorfman M et al. . Feasibility and effects of home-based smartphone-delivered automated feedback training for gait in people with Parkinson’s disease: A pilot randomized controlled trial. Parkinsonism Relat Disord. 2016;22:28–34. [DOI] [PubMed] [Google Scholar]

[ocx006-B53] 53. Sawesi S, Rashrash M, Phalakornkule K et al. . The impact of information technology on patient engagement and health behavior change: a systematic review of the literature. JMIR Med Inform. 2016;41:e1. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B54] 54. Kukafka R, Johnson SB, Linfante A et al. . Grounding a new information technology implementation framework in behavioral science: a systematic analysis of the literature on IT use. J Biomed Inform. 2003;363:218–27. [DOI] [PubMed] [Google Scholar]

[ocx006-B55] 55. Jakicic JM, Davis KK, Rogers RJ et al. . Effect of wearable technology combined with a lifestyle intervention on long-term weight loss: The IDEA randomized clinical trial. JAMA. 2016;31611:1161–71. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B56] 56. Riley WT, Rivera DE, Atienza AA et al. . Health behavior models in the age of mobile interventions: are our theories up to the task? Transl Behav Med. 2011;11:53–71. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B57] 57. Free C, Phillips G, Galli L et al. . The effectiveness of mobile-health technology-based health behaviour change or disease management interventions for health care consumers: a systematic review. PLoS Med. 2013;101:e1001362. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B58] 58. Quinn CC, Clough SS, Minor JM et al. . WellDoc mobile diabetes management randomized controlled trial: change in clinical and behavioral outcomes and patient and physician satisfaction. Diabetes Technol Ther. 2008;103:160–68. [DOI] [PubMed] [Google Scholar]

[ocx006-B59] 59. Kirwan M, Vandelanotte C, Fenning A et al. . Diabetes Self-Management Smartphone Application for Adults With Type 1 Diabetes: Randomized Controlled Trial. J Med Internet Res. 2013;1511:e235. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B60] 60. Cafazzo AJ, Casselman M, Hamming N et al. . Design of an mHealth app for the self-management of adolescent type 1 diabetes: a pilot study. J Med Internet Res. 2012;143:e70. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B61] 61. Krebs P, Prochaska JO, Rossi JS. A meta-analysis of computer-tailored interventions for health behavior change. Prev Med. 2010;51(3–4):214–21. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B62] 62. Portnoy DB, Scott-Sheldon LAJ, Johnson BT et al. . Computer-delivered interventions for health promotion and behavioral risk reduction: a meta-analysis of 75 randomized controlled trials, 1988–2007. Prev Med. 2008;471:3–16. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B63] 63. Brug J, Campbell M, van Assema P. The application and impact of computer-generated personalized nutrition education: a review of the literature. Patient Educ Couns. 1999;362:145–56. [DOI] [PubMed] [Google Scholar]

[ocx006-B64] 64. Neville LM, O’Hara B, Milat A. Computer-tailored physical activity behavior change interventions targeting adults: a systematic review. Int J Behav Nutr Phys Act. 2009;61:30. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B65] 65. Norman GJ, Zabinski MF, Adams MA et al. . A review of eHealth interventions for physical activity and dietary behavior change. Am J Prev Med. 2007;334:336–45. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B66] 66. Johnson CM, Johnson TR, Zhang J. A user-centered framework for redesigning health care interfaces. J Biomed Inform. 2005;381:75–87. [DOI] [PubMed] [Google Scholar]

[ocx006-B67] 67. Gustafson DH, Hawkins R, Boberg E et al. . Impact of a patient-centered, computer-based health information/support system. Am J Prev Med. 1999;161:1–9. [DOI] [PubMed] [Google Scholar]

[ocx006-B68] 68. De Vito Dabbs A, Song MK, Myers BA et al. . A randomized controlled trial of a mobile health intervention to promote self-management after lung transplantation. Am J Transplant. 2016;167: 2172–80. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B69] 69. Preece J, Rogers Y, Sharp H. Interaction Design: Beyond Human-Computer Interaction. New York: Wiley; 2015. [Google Scholar]

[ocx006-B70] 70. De Vito Dabbs A, Myers BA, Mc Curry KR et al. . User-centered design and interactive health technologies for patients. Comput Inform Nurs. 2009;273:175–83. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B71] 71. Finkelstein J, Knight A, Marinopoulos S et al. . Enabling patient-centered care through health information technology. Evid Rep Technol Assess (Full Rep). 2012;206:1–1531. [PMC free article] [PubMed] [Google Scholar]

[ocx006-B72] 72. Gibbons M, Samal RFWL et al. . Impact of Consumer Health Informatics Applications. Evidence Report/Technology Assessment No. 188. Rockville, MD: AHRQ Publication No. 09(10)-E019; 2009. [PMC free article] [PubMed] [Google Scholar]

[ocx006-B73] 73. Schulz KF, Altman DG, Moher D. CONSORT 2010 Statement: updated guidelines for reporting parallel group randomised trials. BMC Med. 2010;81:18. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B74] 74. Soares HP, Daniels S, Kumar A et al. . Bad reporting does not mean bad methods for randomised trials: observational study of randomised controlled trials performed by the Radiation Therapy Oncology Group. BMJ. 2004;3287430:22. [DOI] [PMC free article] [PubMed] [Google Scholar]

[ocx006-B75] 75. Huwiler-Müntener K, Jüni P, Junker C et al. . Quality of reporting of randomized trials as a measure of methodologic quality. JAMA. 2002;28721:2801–04. [DOI] [PubMed] [Google Scholar]

[ocx006-B76] 76. Plint AC, Moher D, Morrison A et al. . Does the CONSORT checklist improve the quality of reports of randomised controlled trials? A systematic review. Med J Aust. 2006;1855:263–67. [DOI] [PubMed] [Google Scholar]

[ocx006-B77] 77. Moher D, Jones A, Lepage L. Use of the consort statement and quality of reports of randomized trials: a comparative before-and-after evaluation. JAMA. 2001;28515:1992–95. [DOI] [PubMed] [Google Scholar]

[ocx006-B78] 78. Kane RL, Wang J, Garrard J. Reporting in randomized clinical trials improved after adoption of the CONSORT statement. J Clin Epidemiol. 2007;603:241–49. [DOI] [PubMed] [Google Scholar]

PERMALINK

Systematic review of the effectiveness of health-related behavioral interventions using portable activity sensing devices (PASDs)

Hamed Abedtash

Richard J Holden

Abstract

INTRODUCTION

Objectives

METHODS

Data sources and search queries

Study selection

Data extraction

Quality assessment

RESULTS

Figure 1.

Table 1.

Study characteristics

Figure 2.

Target population characteristics

Utilized portable activity sensing devices

Table 2.

Studied interventions

Figure 3.

Technology characteristics

Effects of interventions on health-related outcomes

Table 3.

Table 4.

Physical activity

Body measures

Physiological measures

Weight loss

Balance control, postural stability, and gait performance

Quality of life

Other outcomes

Risk of bias in included studies

Figure 4.

Figure 5.

DISCUSSION

The importance of behavioral change techniques

The role of behavioral informatics

Health-related outcomes

Quality of reviewed randomized controlled trials

Limitations

Future work

CONCLUSION

Funding

Competing Interests

Contributors

SUPPLEMENTARY MATERIAL

Supplementary Material

References

Associated Data

Supplementary Materials

ACTIONS

PERMALINK

RESOURCES

Similar articles

Cited by other articles

Links to NCBI Databases