Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events: The ENGAGE Randomized Clinical Trial | Cardiology | JAMA Cardiology | JAMA Network
[Skip to Navigation]
Visual Abstract. Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events
Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events
Figure 1.  Consolidated Standards of Reporting Trials Diagram
Consolidated Standards of Reporting Trials Diagram

Participants in all arms received a wearable device and established baseline measures. Participants in the control arm received a daily text message with feedback on the previous day’s step count but no other interventions. Participants in the gamification arms were randomly assigned to different ways to set and implement daily step goals and then were entered into the same gamification intervention.

Figure 2.  Mean Daily Step Counts
Mean Daily Step Counts

Depicted are mean daily steps by week during the introductory intervention (weeks 1-8), maintenance intervention (weeks 9-16), and follow-up (weeks 17-24) using imputed data.

Figure 3.  Mean Daily Minutes of Moderate to Vigorous Physical Activity
Mean Daily Minutes of Moderate to Vigorous Physical Activity

Depicted are daily minutes of moderate to vigorous physical activity by week during the introductory intervention (weeks 1-8), maintenance intervention (weeks 9-16), and follow-up (weeks 17-24) using imputed data.

Table 1.  Participant Characteristics
Participant Characteristics
Table 2.  Physical Activity Outcomesa
Physical Activity Outcomesa
1.
Thompson  PD, Eijsvogels  TMH.  New physical activity guidelines: a call to activity for clinicians and patients.   JAMA. 2018;320(19):1983-1984. doi:10.1001/jama.2018.16070 PubMedGoogle ScholarCrossref
2.
Piercy  KL, Troiano  RP, Ballard  RM,  et al.  The physical activity guidelines for Americans.   JAMA. 2018;320(19):2020-2028. doi:10.1001/jama.2018.14854 PubMedGoogle ScholarCrossref
3.
Gal  R, May  AM, van Overmeeren  EJ, Simons  M, Monninkhof  EM.  The effect of physical activity interventions comprising wearables and smartphone applications on physical activity: a systematic review and meta-analysis.   Sports Med Open. 2018;4(1):42. doi:10.1186/s40798-018-0157-9 PubMedGoogle ScholarCrossref
4.
Patel  MS, Asch  DA, Volpp  KG.  Wearable devices as facilitators, not drivers, of health behavior change.   JAMA. 2015;313(5):459-460. doi:10.1001/jama.2014.14781 PubMedGoogle ScholarCrossref
5.
Ding  D, Lawson  KD, Kolbe-Alexander  TL,  et al; Lancet Physical Activity Series 2 Executive Committee.  The economic burden of physical inactivity: a global analysis of major non-communicable diseases.   Lancet. 2016;388(10051):1311-1324. doi:10.1016/S0140-6736(16)30383-X PubMedGoogle ScholarCrossref
6.
Hallal  PC, Bauman  AE, Heath  GW, Kohl  HW  III, Lee  IM, Pratt  M.  Physical activity: more of the same is not enough.   Lancet. 2012;380(9838):190-191. doi:10.1016/S0140-6736(12)61027-7 PubMedGoogle ScholarCrossref
7.
Andersen  LB, Mota  J, Di Pietro  L.  Update on the global pandemic of physical inactivity.   Lancet. 2016;388(10051):1255-1256. doi:10.1016/S0140-6736(16)30960-6 PubMedGoogle ScholarCrossref
8.
Locke  EA, Latham  GP.  The development of goal setting theory: a half century retrospective.   Motivation Sci. 2019;5(2):93-105. doi:10.1037/mot0000127 Google ScholarCrossref
9.
Locke  EA, Latham  G, Smith  KJ. A Theory of Goal Setting & Task Performance. Prentice Hall; 1990.
10.
Elliot  AJ. Advances in Motivation Science. Academic Press; 2014.
11.
Swann  C, Rosenbaum  S, Lawrence  A, Vella  SA, McEwan  D, Ekkekakis  P.  Updating goal-setting theory in physical activity promotion: a critical conceptual review.   Health Psychol Rev. 2021;15(1):34-50. doi:10.1080/17437199.2019.1706616 PubMedGoogle ScholarCrossref
12.
Gerber  Y, Myers  V, Goldbourt  U,  et al; Israel Study Group on First Acute Myocardial Infarction.  Neighborhood socioeconomic status and leisure-time physical activity after myocardial infarction: a longitudinal study.   Am J Prev Med. 2011;41(3):266-273. doi:10.1016/j.amepre.2011.05.016 PubMedGoogle ScholarCrossref
13.
Boone-Heinonen  J, Diez Roux  AV, Kiefe  CI, Lewis  CE, Guilkey  DK, Gordon-Larsen  P.  Neighborhood socioeconomic status predictors of physical activity through young to middle adulthood: the CARDIA study.   Soc Sci Med. 2011;72(5):641-649. doi:10.1016/j.socscimed.2010.12.013 PubMedGoogle ScholarCrossref
14.
Tucker-Seeley  RD, Subramanian  SV, Li  Y, Sorensen  G.  Neighborhood safety, socioeconomic status, and physical activity in older adults.   Am J Prev Med. 2009;37(3):207-213. doi:10.1016/j.amepre.2009.06.005 PubMedGoogle ScholarCrossref
15.
Asch  DA, Volpp  KG.  On the way to health.   LDI Issue Brief. 2012;17(9):1-4.PubMedGoogle Scholar
16.
Patel  MS, Benjamin  EJ, Volpp  KG,  et al.  Effect of a game-based intervention designed to enhance social incentives to increase physical activity among families: the BE FIT randomized clinical trial.   JAMA Intern Med. 2017;177(11):1586-1593. doi:10.1001/jamainternmed.2017.3458 PubMedGoogle ScholarCrossref
17.
Patel  MS, Small  DS, Harrison  JD,  et al.  Effectiveness of behaviorally designed gamification interventions with social incentives for increasing physical activity among overweight and obese adults across the United States: the STEP UP randomized clinical trial.   JAMA Intern Med. 2019;179(12):1624-1632. doi:10.1001/jamainternmed.2019.3505 PubMedGoogle ScholarCrossref
18.
Patel  MS, Asch  DA, Rosin  R,  et al.  Framing financial incentives to increase physical activity among overweight and obese adults: a randomized, controlled trial.   Ann Intern Med. 2016;164(6):385-394. doi:10.7326/M15-1635 PubMedGoogle ScholarCrossref
19.
Patel  MS, Asch  DA, Rosin  R,  et al.  Individual versus team-based financial incentives to increase physical activity: a randomized, controlled trial.   J Gen Intern Med. 2016;31(7):746-754. doi:10.1007/s11606-016-3627-0 PubMedGoogle ScholarCrossref
20.
Chokshi  NP, Adusumalli  S, Small  DS,  et al.  Loss-framed financial incentives and personalized goal-setting to increase physical activity among ischemic heart disease patients using wearable devices: the ACTIVE REWARD randomized trial.   J Am Heart Assoc. 2018;7(12):e009173. doi:10.1161/JAHA.118.009173 PubMedGoogle Scholar
21.
Patel  MS, Volpp  KG, Rosin  R,  et al.  A randomized trial of social comparison feedback and financial incentives to increase physical activity.   Am J Health Promot. 2016;30(6):416-424. doi:10.1177/0890117116658195 PubMedGoogle ScholarCrossref
22.
Patel  MS, Volpp  KG, Rosin  R,  et al.  A randomized, controlled trial of lottery-based financial incentives to increase physical activity among overweight and obese adults.   Am J Health Promot. 2018;32(7):1568-1575. doi:10.1177/0890117118758932 PubMedGoogle ScholarCrossref
23.
Case  MA, Burwick  HA, Volpp  KG, Patel  MS.  Accuracy of smartphone applications and wearable devices for tracking physical activity data.   JAMA. 2015;313(6):625-626. doi:10.1001/jama.2014.17841 PubMedGoogle ScholarCrossref
24.
Schultz  WM, Kelli  HM, Lisko  JC,  et al.  Socioeconomic status and cardiovascular outcomes: challenges and interventions.   Circulation. 2018;137(20):2166-2178. doi:10.1161/CIRCULATIONAHA.117.029652 PubMedGoogle ScholarCrossref
25.
Goff  DC  Jr, Lloyd-Jones  DM, Bennett  G,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines.  2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines.   Circulation. 2014;129(25)(suppl 2):S49-S73. doi:10.1161/01.cir.0000437741.48606.98 PubMedGoogle Scholar
26.
Bassett  DR  Jr, Wyatt  HR, Thompson  H, Peters  JC, Hill  JO.  Pedometer-measured physical activity and health behaviors in US adults.   Med Sci Sports Exerc. 2010;42(10):1819-1825. doi:10.1249/MSS.0b013e3181dc2e54 PubMedGoogle ScholarCrossref
27.
Kang  M, Rowe  DA, Barreira  TV, Robinson  TS, Mahar  MT.  Individual information-centered approach for handling physical activity missing data.   Res Q Exerc Sport. 2009;80(2):131-137. doi:10.1080/02701367.2009.10599546 PubMedGoogle ScholarCrossref
28.
Kahneman  D, Tversky  A.  Prospect theory: an analysis of decision under risk.   Econometrica. 1979;47(2):263. doi:10.2307/1914185 Google ScholarCrossref
29.
Dai  H, Milkman  KL, Riis  J.  The fresh start effect: temporal landmarks motivate aspirational behavior.   Management Science. 2014;60(10):2563-2582. doi:10.1287/mnsc.2014.1901 Google ScholarCrossref
30.
Marshall  SJ, Levy  SS, Tudor-Locke  CE,  et al.  Translating physical activity recommendations into a pedometer-based step goal: 3000 steps in 30 minutes.   Am J Prev Med. 2009;36(5):410-415. doi:10.1016/j.amepre.2009.01.021 PubMedGoogle ScholarCrossref
31.
Tudor-Locke  C, Sisson  SB, Collova  T, Lee  SM, Swan  PD.  Pedometer-determined step count guidelines for classifying walking intensity in a young ostensibly healthy population.   Can J Appl Physiol. 2005;30(6):666-676. doi:10.1139/h05-147 PubMedGoogle ScholarCrossref
32.
Burman  CF, Sonesson  C, Guilbaud  O.  A recycling framework for the construction of Bonferroni-based multiple tests.   Stat Med. 2009;28(5):739-761. doi:10.1002/sim.3513 PubMedGoogle ScholarCrossref
33.
Young  R, Johnson  DR.  Handling missing values in longitudinal panel data with multiple imputation.   J Marriage Fam. 2015;77(1):277-294. doi:10.1111/jomf.12144 PubMedGoogle ScholarCrossref
34.
Rubin  DB.  Multiple Imputation for Nonresponse in Surveys. Wiley; 1987. doi:10.1002/9780470316696
35.
Bull  FC, Al-Ansari  SS, Biddle  S,  et al.  World Health Organization 2020 guidelines on physical activity and sedentary behaviour.   Br J Sports Med. 2020;54(24):1451-1462. doi:10.1136/bjsports-2020-102955 PubMedGoogle ScholarCrossref
36.
Gebel  K, Ding  D, Chey  T, Stamatakis  E, Brown  WJ, Bauman  AE.  Effect of moderate to vigorous physical activity on all-cause mortality in middle-aged and older Australians.   JAMA Intern Med. 2015;175(6):970-977. doi:10.1001/jamainternmed.2015.0541 PubMedGoogle ScholarCrossref
37.
Patall  EA, Cooper  H, Robinson  JC.  The effects of choice on intrinsic motivation and related outcomes: a meta-analysis of research findings.   Psychol Bull. 2008;134(2):270-300. doi:10.1037/0033-2909.134.2.270 PubMedGoogle ScholarCrossref
38.
Deci  EL, Ryan  RM. Intrinsic Motivation and Self-Determination in Human Behavior: Perspectives in Social Psychology. Plenum; 1985.
39.
Liberman  N, Förster  J.  Goal Gradients, Expectancy, and Value: Goal-Directed Behavior. Psychology Press; 2012.
40.
Lee  IM, Shiroma  EJ, Lobelo  F, Puska  P, Blair  SN, Katzmarzyk  PT; Lancet Physical Activity Series Working Group.  Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy.   Lancet. 2012;380(9838):219-229. doi:10.1016/S0140-6736(12)61031-9 PubMedGoogle ScholarCrossref
41.
Yang  Q, Cogswell  ME, Flanders  WD,  et al.  Trends in cardiovascular health metrics and associations with all-cause and CVD mortality among US adults.   JAMA. 2012;307(12):1273-1283. doi:10.1001/jama.2012.339 PubMedGoogle ScholarCrossref
42.
Saint-Maurice  PF, Troiano  RP, Bassett  DR  Jr,  et al.  Association of daily step count and step intensity with mortality among US adults.   JAMA. 2020;323(12):1151-1160. doi:10.1001/jama.2020.1382 PubMedGoogle ScholarCrossref
43.
Spartano  NL, Davis-Plourde  KL, Himali  JJ,  et al.  Association of accelerometer-measured light-intensity physical activity with brain volume: the Framingham Heart Study.   JAMA Netw Open. 2019;2(4):e192745. doi:10.1001/jamanetworkopen.2019.2745 PubMedGoogle Scholar
44.
Lee  I-M, Shiroma  EJ, Kamada  M, Bassett  DR, Matthews  CE, Buring  JE.  Association of step volume and intensity with all-cause mortality in older women.   JAMA Intern Med. 2019;179(8):1105-1112. doi:10.1001/jamainternmed.2019.0899 PubMedGoogle ScholarCrossref
Limit 200 characters
Limit 25 characters
Conflicts of Interest Disclosure

Identify all potential conflicts of interest that might be relevant to your comment.

Conflicts of interest comprise financial interests, activities, and relationships within the past 3 years including but not limited to employment, affiliation, grants or funding, consultancies, honoraria or payment, speaker's bureaus, stock ownership or options, expert testimony, royalties, donation of medical equipment, or patents planned, pending, or issued.

Err on the side of full disclosure.

If you have no conflicts of interest, check "No potential conflicts of interest" in the box below. The information will be posted with your response.

Not all submitted comments are published. Please see our commenting policy for details.

Limit 140 characters
Limit 3600 characters or approximately 600 words
    Views 1,446
    Citations 0
    Original Investigation
    September 1, 2021

    Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events: The ENGAGE Randomized Clinical Trial

    Author Affiliations
    • 1Department of Medicine, Perelman School of Medicine, University of Pennsylvania, Philadelphia
    • 2The Wharton School, University of Pennsylvania, Philadelphia
    • 3Penn Medicine Nudge Unit, University of Pennsylvania, Philadelphia
    • 4Penn Center for Health Incentives and Behavioral Economics, University of Pennsylvania, Philadelphia
    • 5Crescenz Veterans Affairs Medical Center, Philadelphia
    • 6Now with Ascension Health, St Louis, Missouri
    • 7Virginia Department of Medical Assistance Services, Richmond
    • 8Department of Internal Medicine, Virginia Commonwealth University School of Medicine, Richmond
    JAMA Cardiol. Published online September 1, 2021. doi:10.1001/jamacardio.2021.3176
    Visual Abstract. Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events
    Effect of Goal-Setting Approaches Within a Gamification Intervention to Increase Physical Activity Among Economically Disadvantaged Adults at Elevated Risk for Major Adverse Cardiovascular Events
    Key Points

    Question  What is the best way to set and implement goals within a gamification intervention to increase physical activity among economically disadvantaged adults at elevated risk for major adverse cardiovascular events?

    Findings  In this randomized clinical trial of 500 patients from lower-income neighborhoods with elevated risk for major adverse cardiovascular events, a 4-month behaviorally designed gamification intervention significantly increased physical activity when daily step goals were self-chosen (rather than assigned) and implemented immediately (rather than gradually) relative to an attention control group. Increases in daily steps and minutes of moderate to vigorous physical activity were sustained during the 2-month follow-up period.

    Meaning  Among economically disadvantaged adults at elevated risk for major adverse cardiovascular events, programs designed to increase physical activity may be more effective if goals are self-chosen and implemented immediately.

    Abstract

    Importance  Health promotion efforts commonly communicate goals for healthy behavior, but the best way to design goal setting among high-risk patients has not been well examined.

    Objective  To test the effectiveness of different ways to set and implement goals within a behaviorally designed gamification intervention to increase physical activity.

    Design, Setting, and Participants  Evaluation of the Novel Use of Gamification With Alternative Goal-setting Experiences was conducted from January 15, 2019, to June 1, 2020. The 24-week randomized clinical trial included a remotely monitored 8-week introductory intervention period, 8-week maintenance intervention period, and 8-week follow-up period. A total of 500 adults from lower-income neighborhoods in and around Philadelphia, Pennsylvania, who had either an atherosclerotic cardiovascular disease (ASCVD) condition or a 10-year ASCVD risk score greater than or equal to 7.5% were enrolled. Participants were paid for enrolling in and completing the trial.

    Interventions  All participants used a wearable device to track daily steps, established a baseline level, and were then randomly assigned to an attention control or 1 of 4 gamification interventions that varied only on how daily step goals were set (self-chosen or assigned) and implemented (immediately or gradually).

    Main Outcome Measures  The primary outcome was change in mean daily steps from baseline to the 8-week maintenance intervention period. Other outcomes included changes in minutes of moderate to vigorous physical activity. All randomly assigned participants were included in the intention-to-treat analysis.

    Results  Of the 500 participants, 331 individuals (66.2%) were Black, 114 were White (22.8%), and 348 were women (69.6%). Mean (SD) age was 58.5 (10.8) years and body mass index was 33.2 (7.8). A total of 215 participants (43.0%) had an ASCVD condition. Compared with the control arm, participants with self-chosen and immediate goals had significant increases in the number of daily steps during the maintenance intervention period (1384; 95% CI, 805-1963; P < .001) that were sustained during the 8-week follow-up (1391; 95% CI, 785-1998; P < .001). This group also had significant increases in daily minutes of moderate to vigorous physical activity during the maintenance intervention (4.1; 95% CI, 1.8-6.4; P < .001) that were sustained during follow-up (3.5; 95% CI, 1.1-5.8; P = .004). No other gamification arms had consistent increases in physical activity compared with the control arm. No major adverse events were reported.

    Conclusions and Relevance  In this trial among economically disadvantaged adults at elevated risk for major adverse cardiovascular events, a gamification intervention led to increases in physical activity that were sustained during 8 weeks of follow-up when goals were self-chosen and implemented immediately.

    Trial Registration  ClinicalTrials.gov Identifier: NCT03749473

    Introduction

    Health promotion efforts commonly communicate goals for healthy behavior. For example, the Centers for Disease Control and Prevention recommends that adults achieve at least 150 minutes of moderate to vigorous physical activity (MVPA) per week.1,2 Activity trackers, such as wearable devices and smartphone applications, also often include a preset goal of 10 000 steps per day.3 However, most people are unable to reach this target and those who do often cannot sustain it.4-7

    One potential explanation for the lack of achievement is that inadequate attention is given to the process of goal setting, which is the development of an action plan designed to motivate an individual toward a goal.8 Previous evidence has led to the development of goal-setting theory that postulates that specific and more challenging goals lead to better performance than nonspecific, easy goals.9 However, these findings are conditional on 4 moderating factors: ability to attain the goal, commitment to the goal, receiving feedback on performance, and having situational resources to achieve the goal.8,10 These factors are more often associated with individuals who are already active and have adequate resources.11 Adults from neighborhoods with lower socioeconomic status have had more challenges with increasing physical activity than the general population.12-14 Therefore, more insights are needed on how to set and implement goals for individuals who may be less active and economically disadvantaged.

    In this study, our objective was to use a randomized clinical trial to evaluate the effectiveness of different approaches to set and implement goals within a behaviorally designed gamification intervention to increase physical activity. We recruited patients from lower-income neighborhoods in and around Philadelphia, Pennsylvania, who were at elevated risk for major adverse cardiovascular events and conducted the trial remotely using wearable devices and an automated technology platform.15

    Methods
    Study Design

    Evaluation of the Novel Use of Gamification With Alternative Goal-setting Experiences (ENGAGE) was a randomized clinical trial conducted between January 15, 2019, and June 1, 2020, consisting of an 8-week introductory intervention period, 8-week maintenance intervention period, and 8-week follow-up period. Analyses were conducted between September 15, 2020, and January 14, 2021. The trial protocol was approved by the University of Pennsylvania Institutional Review Board (Supplement 1). This study followed Consolidated Standards of Reporting Trials (CONSORT) reporting guideline for randomized clinical trials.

    The study was conducted using Way to Health,15 a research technology platform at the University of Pennsylvania used previously for remote monitoring and physical activity interventions.16-22 Participants used the study website to create an account, provide informed consent online, and complete baseline and validated survey assessments. Each participant received regular study communications by email and text message. Eligible participants were mailed a wrist-worn wearable device (Fitbit Alta or Fitbit Inspire; Fitbit Inc) and connected their devices to the Way to Health platform for remote data collection to track daily step counts. Earlier work has demonstrated these types of devices are accurate for tracking step counts.23 All participants received $30 for enrolling in the trial, $30 for completing the 8-week maintenance intervention period and surveys, and $40 for completing the 8-week follow-up and surveys.

    Participants

    Potential participants were identified from the electronic health records at Penn Medicine and sent email invitations to learn more about the study online. The study team conducted outreach to approximately 19 000 patients. Recruitment occurred from January 15 to December 9, 2019.

    Participants were recruited from lower-income neighborhoods in and around Philadelphia that were identified using the 2018 US Census Data. Recruitment began with zip codes in West Philadelphia and then progressed through the remaining zip codes from lowest to higher income until the trial population was filled. Lower neighborhood socioeconomic status, specifically, lower income level, has been found to be associated with a higher risk for cardiovascular disease and lower levels of physical activity.12-14,24 Individuals were eligible if they were aged 18 years or older, able to read and provide informed consent to participate, had either an atherosclerotic cardiovascular disease (ASCVD) condition or a 10-year ASCVD risk score greater than or equal to 7.5% calculated according to the 2013 American College of Cardiology/American Heart Association guidelines,25 and owned a smartphone or tablet compatible with the wearable device. Potential participants were excluded if they had a baseline daily step count greater than 10 000, a condition that made their participation infeasible (eg, inability to provide informed consent or speak/read/write English), a condition that made participation unsafe (eg, pregnancy or being told by a physician not to exercise), were already enrolled in another study targeting physical activity, or had any medical conditions or other reasons that would prohibit them from participating in the 24-week trial.

    Baseline Step Count

    After the wearable device was set up and connected to the study program, the participant was asked to adjust to wearing the device for several weeks. During this 2-week run-in period, a baseline step count was estimated using the second week of data—a method used in previous work.16,17,20 The first week of data was ignored to diminish the potential upward bias of the estimate from higher activity during initial device use. To prevent potential mismeasurement, we ignored any daily values less than 1000 steps because evidence indicates these values are unlikely to represent capture of actual activity during the whole day.26,27 If less than 4 days of data were available during the second week (n = 40), the participant was contacted to inquire about any device issues and the run-in period was extended until at least 4 days of data were captured.

    Randomization

    After establishing a baseline step count, participants were randomized electronically by stratifying on their baseline step count (≤5000, 5001-7500, or >7500 steps per day) and using block sizes of 5 groups with 3 participants per group. All investigators, statisticians, and data analysts were blinded to arm assignments until the study and analysis were completed. Participants were randomized to either the control group or 1 of 4 gamification intervention arms with different approaches to goal setting.

    Interventions

    Participants in the control arm were asked to use the wearable device but did not establish daily step goals. Each day during the 24-week study, participants in the control group received daily feedback via text message on the previous day’s step count. This group received no other interventions.

    Participants in the gamification arms were randomly assigned to have step goals either assigned (2000-step increase from baseline) or self-chosen (select a goal between 1000 and 3000 steps above baseline). Participants in the self-chosen arms could change their goal at any time during the study as long as it was within the provided range. When the study began, goals were implemented either immediately (strive for the goal beginning on day 1) or gradually over the 8-week introductory period (step targets increase by 12.5% each week for 8 weeks to full goal by week 9). After 16 weeks, the gamification interventions were turned off and all participants received the same daily feedback via text message as the control group for the final 8 weeks.

    Aside from differences in goal setting, participants in the gamification arms received the same intervention implemented as follows. The game included points and levels that were run automatically (participants did not have to actively play the game—just strive for step goals) and provided daily feedback via text message on their progress. The design was adapted from prior work that incorporated principles from behavioral economics.16,17 At the beginning of each week the participant received 70 points (10 for each day of that week). If the participant did not achieve their step goal on a given day, they lost 10 points from their balance. This design leverages loss aversion, which has been demonstrated to motivate behavior change more effectively with losses compared with gains.18,20,28 Third, at the end of each week, if the participant had at least 40 points, they moved up a level (levels from lowest to highest: blue, bronze, silver, gold, and platinum). If not, participants dropped a level. All participants began at the middle of the 5 levels (silver level). Each week, participants received a fresh set of 70 points on Monday, a design that emphasizes the fresh-start effect, which is the tendency for aspirational behavior around temporal landmarks, such as the beginning of the year, month, or week.29

    Outcome Measures

    The primary outcome was the change in daily steps from baseline to the maintenance intervention period (weeks 9 to 16). The secondary outcome was the change in daily steps from baseline to the 8-week follow-up period. Exploratory outcomes included the change in minutes of MVPA from baseline to the maintenance intervention and follow-up periods. Based on established methods, MVPA was estimated by totaling the number of minutes per day that had a step cadence of 100 steps or more per minute.30,31

    Statistical Analysis

    A priori power calculations used a 2-phased testing procedure that has an overall familywise error rate that is controlled at 0.05.32 In the first phase, the 4 intervention arms were compared with the control arm. We estimated that a sample of 500 participants (100 per arm) would provide at least 80% power to detect a 1000-step difference between each intervention arm and the control arm, with an SD of 2000 steps, a 10% dropout rate, and a conservative Bonferroni adjustment of the type I error rate with a 2-sided α level of 0.0125. In the second phase, only intervention arms that were significantly different from the control arm were compared with each other using a conservative Bonferroni adjustment of the type I error rate with a 2-sided α level of 0.0125 to adjust for up to 4 comparisons.

    All randomly assigned participants were included in the intention-to-treat analysis. For each participant on each day of the study (participant-day level), the number of steps achieved was obtained as a continuous variable. Data can be missing for any day if the participant did not use the wearable device, did not upload data, or reported they had an event that prevented them from normal activity (eg, hospital admission). Missing data rates are available in eTable 1 in Supplement 2) and were similar to previous physical activity interventions.16,17,20 For the prespecified main analysis, we used multiple imputation for step values that were either missing or for values less than 1000 steps per day. This method has been used in prior work16,17,20 and in this study because evidence indicates that daily step values less than 1000 may not represent full data capture.26,27 Five imputations were conducted using the mice package in R, version 3.4.0 (R Foundation for Statistical Computing) that allows for participant random effects with this data structure.33 The following predictors of missing data were included: study arm, calendar month, week of study, baseline daily steps, age, sex, race/ethnicity, educational level, marital status, household income level, self-reported health, ASCVD condition (yes/no), and participant random effect. Results were combined using Rubin standard rules.34 Sensitivity analyses were conducted using collected data without multiple imputation, both with and without step values less than 1000 per day. For minutes of MVPA, data were missing on the same participant days for which step values were missing. The same imputation method was used except with baseline minutes of MVPA as a predictor for imputation rather than baseline step count.

    Similar to earlier work,16,17,20 adjusted analyses used PROC GLIMMIX, version 9.4 (SAS Institute Inc) to fit generalized mixed-effects models with a random intercept and participant random effects, and to account for the repeated measures of daily step counts. In the main adjusted models, we included fixed effects for calendar month and treatment arm and adjusted for either baseline step count or baseline minutes of MVPA. To test the robustness of our findings, we fit a fully adjusted model that also included indicators for age, sex, race/ethnicity, educational level, marital status, annual household income, self-reported health, body mass index, and presence of an ASCVD condition. We assumed a normal distribution and obtained difference in number of steps or minutes of MVPA between arms for the intervention and follow-up periods using the least square means command.

    In secondary analyses, we examined the main effects of the goal-setting approaches. The same models were fit using treatment arms to evaluate how goals were set (control, self-chosen, and assigned) or how goals were implemented (control, immediate, and gradual). We conducted 2 sensitivity analyses to evaluate the impact of the COVID-19 pandemic (participation between March 1 and June 1, 2020, and associated stay-at-home orders), which affected 11.8% (59 of 500) of participants in their final months of the trial. First, we evaluated the main imputed models for the 441 participants who completed the trial before the pandemic began. Second, we evaluated the main imputed models for all participants and added a variable to adjust for the pandemic at the participant-day level.

    Results
    Participant Sample

    In this trial, 500 participants were randomized (Figure 1). Participants in the sample had a mean (SD) age of 58.5 (10.8) years and body mass index of 33.2 (7.8) (calculated as weight in kilograms divided by height in meters squared); 348 were women (69.6%), 152 were men (30.4%), 331 (66.2%) were Black, 114 (22.8%) were White, and 215 (43.0%) had an ASCVD condition (Table 1). The distribution of the sample by zip code level median annual household income was 28.6% (143 of 500) from less than $30 000, 67.6% (338 of 500) from less than $40 000, and 91.6% (458 of 500) from less than $50 000 (eTable 2 in Supplement 2).

    For the overall sample, the mean (SD) baseline daily step count was 5777 (2266) and the baseline daily minutes of MVPA was 6.1 (7.4). There were no significant differences in baseline physical activity measures across the treatment arms. The proportion of days with missing data was similar to earlier studies16-22and ranged among the arms from 12.2% (678 of 5544) to 19.1% (1059 of 5544) during the introductory intervention, 20.7% (1113 of 5376) to 32.6% (1806 of 5544) during the maintenance intervention, and 24.6% (1322 of 5376) to 37.2% (2063 of 5544) during follow-up.

    Physical Activity Outcomes

    Figure 2 and Figure 3 depict the unadjusted daily step counts and minutes of MVPA by week and study arm. During the 24-week trial, the control arm had a mostly steady increase in physical activity from baseline, ranging from approximately 300 to 500 steps per day and 0 to 2 minutes of MVPA per day. Three of the gamification arms (both arms with assigned goals and the arm with self-chosen and gradual goals) also had mostly steady increases in physical activity, ranging from approximately 800 to 1200 steps per day and 2 to 5 minutes of MVPA per day. The gamification arm with self-chosen and immediate goals had the largest increase in physical activity, ranging from 1600 to 1900 steps per day and 4 to 6 minutes of MVPA per day.

    In the main adjusted models, compared with the control group (Table 2), participants with self-chosen and immediate goals had significant increases in daily steps during the maintenance intervention period (1384; 95% CI, 805-1963; P < .001) that were sustained during follow-up (1391; 95% CI, 785-1998; P < .001). This group also had significant increases in daily minutes of MVPA during the maintenance intervention (4.1; 95% CI, 1.8-6.4; P < .001) that were sustained during follow-up (3.5; 95% CI, 1.1-5.8; P = .004). No other gamification arms had consistent increases in physical activity compared with the control group. These findings were consistent in fully adjusted models and sensitivity analyses (eTables 3-7 in Supplement 2). These findings were also consistent in sensitivity models excluding and adjusting for participants enrolled during the COVID-19 pandemic (eTable 8 and eTable 9 in Supplement 2).

    In main effects models, self-chosen goals significantly increased daily steps (967; 95% CI, 464-1470; P < .001) and daily minutes of MVPA (3.3; 95% CI, 1.3-5.3; P = .001) compared with the control group during the maintenance intervention period (eTable 10 in Supplement 2). The effects were sustained during follow-up. Immediate goals also significantly increased daily steps (956; 95% CI, 446-1465; P < .001) and daily minutes of MVPA (3.6; 95% CI, 1.6-5.6; P < .001) compared with the control group during the maintenance intervention period (eTable 10 in Supplement 2).

    Participant Satisfaction and Safety

    During the 6-month trial, there was only 1 reported adverse event related to the study, which was arthritic knee pain related to increased walking (eTable 11 in Supplement 2). Participants reported high satisfaction with their experience in the study, and 93.3% of participants who responded at the end of the trial stated they planned to continue using the wearable device (eTable 12 in Supplement 2).

    Discussion

    In this randomized clinical trial of adults with elevated risk for major adverse cardiovascular events who were from lower-income neighborhoods, we found that differences in the approach to setting and implementing daily step goals resulted in significant differences in physical activity levels. Participants who were asked to select a daily step goal and then asked to strive for it immediately increased their activity by nearly 1400 steps per day more than control participants, representing a 23% relative increase from baseline. These changes were sustained during the 2-month follow-up period. These participants also had sustained increases in minutes of MVPA, which has been demonstrated to be associated with improvements in health outcomes.2,35,36 To our knowledge, this is one of the first randomized trials to demonstrate the effect that differences in the design of goal setting can have on physical activity, particularly among adults from lower-income neighborhoods who were at elevated risk for a major adverse cardiovascular event.

    Our findings reveal several insights for future health promotion efforts. First, providing participants the ability to choose a step goal led to greater increases in physical activity than did assigning a step goal. It may have been that the act of choosing the goal led to greater intrinsic motivation to achieve it because the individual felt they had more of a role in determining their experience. A meta-analysis of 41 studies evaluated the outcome of choice on intrinsic motivation and found that not only did providing a choice increase intrinsic motivation, it also increased task effort and performance.37 The finding that choosing one's goal led to higher activity levels is also supported by self-determination theory, which postulates that higher intrinsic motivation leads an individual to be more self-motivated and self-determined.38 Higher activity levels may also have been due to the ability for one to better choose the right goal for themselves. A challenge to any program assigning the same goal to everyone is that it will likely be too high for some people and too low for others, which could have a negative influence on performance.

    Second, asking participants to strive for the step goal immediately upon starting the program led to greater increases in physical activity than gradually increasing the step goal to the target. Participants in the gradual arms may have been discouraged by a step goal that continued to increase. This premise is supported by evidence on goal gradients in which individuals are more motivated to try harder as they get closer to the goal.39 Goals in the gradual arms in this study may have instead been experienced by participants as if the goal were moving away from the individual and led to lower motivation. Another possibility is that gradual goals did not lead to higher step counts because forming a habit early in the gradual arm was more difficult because in each of the first 8 weeks participants were asked to strive for something different.

    Third, these findings are particularly important in this sample of individuals who were from lower-income neighborhoods and had higher rates of cardiovascular disease. Previous work evaluating the effectiveness of gamification to increase physical activity was conducted among mostly healthy family members and employees from a consulting firm who were younger, had higher incomes, and were otherwise mostly healthy.16,17 Adults of lower socioeconomic status have had more challenges with increasing physical activity than the general population.24 Higher levels of physical activity have been demonstrated to reduce the risk of ASCVD and related events (eg, myocardial infarction or stroke),40,41 but more than half of adults in the US and a quarter of people worldwide do not engage in sufficient physical activity.2,35 The interventions in this study are highly scalable in that they used an automated technology platform and monitored activity remotely.

    Three of the 4 gamification arms (both arms with assigned goals and the arm with self-chosen and gradual goals) had smaller increases in physical activity compared with the control arm of approximately 500 to 600 steps per day and 2 to 3 minutes of MVPA per day. Our study was not powered to detect these more-modest increases in activity levels. Nonetheless, these findings may indicate that behaviorally designed gamification led to some increases in physical activity and that these changes were higher and statistically different from control when combined with self-chosen and immediate goals.

    Limitations

    This study has limitations. First, this was a small sample of participants enrolled from lower-income neighborhoods in and around Philadelphia, which limits generalizability. However, to our knowledge, this is one of the first studies to test the effect of alternative goal-setting strategies in this high-risk population. Second, this study was 6 months in duration. Although most previous goal-setting work has been of much shorter duration,37 the limited follow-up in our trial indicates need for further work to evaluate longer-term outcomes. Third, although increases in step counts and minutes of MVPA have been found to lead to improvements in clinical outcomes,36,42-44 further work is needed to evaluate clinical outcomes directly. Fourth, the findings from this trial were among individuals who were willing to participate, and all arms, including the control, received incentives to enroll and complete the study. Future research testing these approaches in a pragmatic study and in a more general sample would be useful. Fifth, the wearable devices used in this study did not track heart rate and therefore we were unable to evaluate device wear-time. Sixth, participants were recruited based on zip code–level median annual household income and not individual-level income.

    Conclusions

    Health promotion efforts commonly set goals for healthy behavior. Differences in the design of goal setting may affect improving these outcomes. The findings from this randomized clinical trial indicate that individuals increase physical activity more when they have a role in selecting their goals and when the target goals are implemented immediately. These insights may inform future research and the design of goal setting for other behaviors.

    Back to top
    Article Information

    Accepted for Publication: June 21, 2021.

    Published Online: September 1, 2021. doi:10.1001/jamacardio.2021.3176

    Corresponding Author: Mitesh S. Patel, MD, MBA, Ascension Health, 4600 Edmundson Rd, St Louis, MO 63134 (mitesh.patel3@ascension.org).

    Author Contributions: Dr Patel had full access to all of the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

    Concept and design: Patel, Bachireddy, Volpp.

    Acquisition, analysis, or interpretation of data: All authors.

    Drafting of the manuscript: Patel, Harrison.

    Critical revision of the manuscript for important intellectual content: Bachireddy, Small, Harrison, Harrington, Oon, Rareshide, Snider, Volpp.

    Statistical analysis: Bachireddy, Small, Rareshide.

    Obtained funding: Patel, Bachireddy, Volpp.

    Administrative, technical, or material support: Bachireddy, Harrison, Harrington, Volpp.

    Supervision: Patel, Volpp.

    Conflict of Interest Disclosures: Dr Patel reported receiving personal fees as owner of Catalyst Health LLC, fees from Life.io as an advisory board member, fees from HealthMine Services as an advisory board member and stockholder, and fees from Holistic Industries as an advisory board member outside the submitted work. Dr Volpp reported receiving personal fees for consulting from CVS Caremark; grants from Humana, Hawaii Medical Services Association; and WW; personal fees for speaking from WW, Center for Corporate Innovation, Greater Philadelphia Business Coalition on Health, Lehigh Valley Medical Center, Vizient, American Gastroenterological Association Tech Conference, and Bridges to Population Health Meeting; fees as part owner of VAL Health consulting firm; and personal fees for consulting from Irish MedTech Summit outside the submitted work. No other disclosures were reported.

    Funding/Support: This study was supported by an internal grant from the Perelman School of Medicine and the University of Pennsylvania Health System through the Penn Medicine Nudge Unit.

    Role of the Funder/Sponsor: The funding organizations had no role in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the manuscript; and decision to submit the manuscript for publication.

    Data Sharing Statement: See Supplement 3.

    References
    1.
    Thompson  PD, Eijsvogels  TMH.  New physical activity guidelines: a call to activity for clinicians and patients.   JAMA. 2018;320(19):1983-1984. doi:10.1001/jama.2018.16070 PubMedGoogle ScholarCrossref
    2.
    Piercy  KL, Troiano  RP, Ballard  RM,  et al.  The physical activity guidelines for Americans.   JAMA. 2018;320(19):2020-2028. doi:10.1001/jama.2018.14854 PubMedGoogle ScholarCrossref
    3.
    Gal  R, May  AM, van Overmeeren  EJ, Simons  M, Monninkhof  EM.  The effect of physical activity interventions comprising wearables and smartphone applications on physical activity: a systematic review and meta-analysis.   Sports Med Open. 2018;4(1):42. doi:10.1186/s40798-018-0157-9 PubMedGoogle ScholarCrossref
    4.
    Patel  MS, Asch  DA, Volpp  KG.  Wearable devices as facilitators, not drivers, of health behavior change.   JAMA. 2015;313(5):459-460. doi:10.1001/jama.2014.14781 PubMedGoogle ScholarCrossref
    5.
    Ding  D, Lawson  KD, Kolbe-Alexander  TL,  et al; Lancet Physical Activity Series 2 Executive Committee.  The economic burden of physical inactivity: a global analysis of major non-communicable diseases.   Lancet. 2016;388(10051):1311-1324. doi:10.1016/S0140-6736(16)30383-X PubMedGoogle ScholarCrossref
    6.
    Hallal  PC, Bauman  AE, Heath  GW, Kohl  HW  III, Lee  IM, Pratt  M.  Physical activity: more of the same is not enough.   Lancet. 2012;380(9838):190-191. doi:10.1016/S0140-6736(12)61027-7 PubMedGoogle ScholarCrossref
    7.
    Andersen  LB, Mota  J, Di Pietro  L.  Update on the global pandemic of physical inactivity.   Lancet. 2016;388(10051):1255-1256. doi:10.1016/S0140-6736(16)30960-6 PubMedGoogle ScholarCrossref
    8.
    Locke  EA, Latham  GP.  The development of goal setting theory: a half century retrospective.   Motivation Sci. 2019;5(2):93-105. doi:10.1037/mot0000127 Google ScholarCrossref
    9.
    Locke  EA, Latham  G, Smith  KJ. A Theory of Goal Setting & Task Performance. Prentice Hall; 1990.
    10.
    Elliot  AJ. Advances in Motivation Science. Academic Press; 2014.
    11.
    Swann  C, Rosenbaum  S, Lawrence  A, Vella  SA, McEwan  D, Ekkekakis  P.  Updating goal-setting theory in physical activity promotion: a critical conceptual review.   Health Psychol Rev. 2021;15(1):34-50. doi:10.1080/17437199.2019.1706616 PubMedGoogle ScholarCrossref
    12.
    Gerber  Y, Myers  V, Goldbourt  U,  et al; Israel Study Group on First Acute Myocardial Infarction.  Neighborhood socioeconomic status and leisure-time physical activity after myocardial infarction: a longitudinal study.   Am J Prev Med. 2011;41(3):266-273. doi:10.1016/j.amepre.2011.05.016 PubMedGoogle ScholarCrossref
    13.
    Boone-Heinonen  J, Diez Roux  AV, Kiefe  CI, Lewis  CE, Guilkey  DK, Gordon-Larsen  P.  Neighborhood socioeconomic status predictors of physical activity through young to middle adulthood: the CARDIA study.   Soc Sci Med. 2011;72(5):641-649. doi:10.1016/j.socscimed.2010.12.013 PubMedGoogle ScholarCrossref
    14.
    Tucker-Seeley  RD, Subramanian  SV, Li  Y, Sorensen  G.  Neighborhood safety, socioeconomic status, and physical activity in older adults.   Am J Prev Med. 2009;37(3):207-213. doi:10.1016/j.amepre.2009.06.005 PubMedGoogle ScholarCrossref
    15.
    Asch  DA, Volpp  KG.  On the way to health.   LDI Issue Brief. 2012;17(9):1-4.PubMedGoogle Scholar
    16.
    Patel  MS, Benjamin  EJ, Volpp  KG,  et al.  Effect of a game-based intervention designed to enhance social incentives to increase physical activity among families: the BE FIT randomized clinical trial.   JAMA Intern Med. 2017;177(11):1586-1593. doi:10.1001/jamainternmed.2017.3458 PubMedGoogle ScholarCrossref
    17.
    Patel  MS, Small  DS, Harrison  JD,  et al.  Effectiveness of behaviorally designed gamification interventions with social incentives for increasing physical activity among overweight and obese adults across the United States: the STEP UP randomized clinical trial.   JAMA Intern Med. 2019;179(12):1624-1632. doi:10.1001/jamainternmed.2019.3505 PubMedGoogle ScholarCrossref
    18.
    Patel  MS, Asch  DA, Rosin  R,  et al.  Framing financial incentives to increase physical activity among overweight and obese adults: a randomized, controlled trial.   Ann Intern Med. 2016;164(6):385-394. doi:10.7326/M15-1635 PubMedGoogle ScholarCrossref
    19.
    Patel  MS, Asch  DA, Rosin  R,  et al.  Individual versus team-based financial incentives to increase physical activity: a randomized, controlled trial.   J Gen Intern Med. 2016;31(7):746-754. doi:10.1007/s11606-016-3627-0 PubMedGoogle ScholarCrossref
    20.
    Chokshi  NP, Adusumalli  S, Small  DS,  et al.  Loss-framed financial incentives and personalized goal-setting to increase physical activity among ischemic heart disease patients using wearable devices: the ACTIVE REWARD randomized trial.   J Am Heart Assoc. 2018;7(12):e009173. doi:10.1161/JAHA.118.009173 PubMedGoogle Scholar
    21.
    Patel  MS, Volpp  KG, Rosin  R,  et al.  A randomized trial of social comparison feedback and financial incentives to increase physical activity.   Am J Health Promot. 2016;30(6):416-424. doi:10.1177/0890117116658195 PubMedGoogle ScholarCrossref
    22.
    Patel  MS, Volpp  KG, Rosin  R,  et al.  A randomized, controlled trial of lottery-based financial incentives to increase physical activity among overweight and obese adults.   Am J Health Promot. 2018;32(7):1568-1575. doi:10.1177/0890117118758932 PubMedGoogle ScholarCrossref
    23.
    Case  MA, Burwick  HA, Volpp  KG, Patel  MS.  Accuracy of smartphone applications and wearable devices for tracking physical activity data.   JAMA. 2015;313(6):625-626. doi:10.1001/jama.2014.17841 PubMedGoogle ScholarCrossref
    24.
    Schultz  WM, Kelli  HM, Lisko  JC,  et al.  Socioeconomic status and cardiovascular outcomes: challenges and interventions.   Circulation. 2018;137(20):2166-2178. doi:10.1161/CIRCULATIONAHA.117.029652 PubMedGoogle ScholarCrossref
    25.
    Goff  DC  Jr, Lloyd-Jones  DM, Bennett  G,  et al; American College of Cardiology/American Heart Association Task Force on Practice Guidelines.  2013 ACC/AHA guideline on the assessment of cardiovascular risk: a report of the American College of Cardiology/American Heart Association Task Force on Practice Guidelines.   Circulation. 2014;129(25)(suppl 2):S49-S73. doi:10.1161/01.cir.0000437741.48606.98 PubMedGoogle Scholar
    26.
    Bassett  DR  Jr, Wyatt  HR, Thompson  H, Peters  JC, Hill  JO.  Pedometer-measured physical activity and health behaviors in US adults.   Med Sci Sports Exerc. 2010;42(10):1819-1825. doi:10.1249/MSS.0b013e3181dc2e54 PubMedGoogle ScholarCrossref
    27.
    Kang  M, Rowe  DA, Barreira  TV, Robinson  TS, Mahar  MT.  Individual information-centered approach for handling physical activity missing data.   Res Q Exerc Sport. 2009;80(2):131-137. doi:10.1080/02701367.2009.10599546 PubMedGoogle ScholarCrossref
    28.
    Kahneman  D, Tversky  A.  Prospect theory: an analysis of decision under risk.   Econometrica. 1979;47(2):263. doi:10.2307/1914185 Google ScholarCrossref
    29.
    Dai  H, Milkman  KL, Riis  J.  The fresh start effect: temporal landmarks motivate aspirational behavior.   Management Science. 2014;60(10):2563-2582. doi:10.1287/mnsc.2014.1901 Google ScholarCrossref
    30.
    Marshall  SJ, Levy  SS, Tudor-Locke  CE,  et al.  Translating physical activity recommendations into a pedometer-based step goal: 3000 steps in 30 minutes.   Am J Prev Med. 2009;36(5):410-415. doi:10.1016/j.amepre.2009.01.021 PubMedGoogle ScholarCrossref
    31.
    Tudor-Locke  C, Sisson  SB, Collova  T, Lee  SM, Swan  PD.  Pedometer-determined step count guidelines for classifying walking intensity in a young ostensibly healthy population.   Can J Appl Physiol. 2005;30(6):666-676. doi:10.1139/h05-147 PubMedGoogle ScholarCrossref
    32.
    Burman  CF, Sonesson  C, Guilbaud  O.  A recycling framework for the construction of Bonferroni-based multiple tests.   Stat Med. 2009;28(5):739-761. doi:10.1002/sim.3513 PubMedGoogle ScholarCrossref
    33.
    Young  R, Johnson  DR.  Handling missing values in longitudinal panel data with multiple imputation.   J Marriage Fam. 2015;77(1):277-294. doi:10.1111/jomf.12144 PubMedGoogle ScholarCrossref
    34.
    Rubin  DB.  Multiple Imputation for Nonresponse in Surveys. Wiley; 1987. doi:10.1002/9780470316696
    35.
    Bull  FC, Al-Ansari  SS, Biddle  S,  et al.  World Health Organization 2020 guidelines on physical activity and sedentary behaviour.   Br J Sports Med. 2020;54(24):1451-1462. doi:10.1136/bjsports-2020-102955 PubMedGoogle ScholarCrossref
    36.
    Gebel  K, Ding  D, Chey  T, Stamatakis  E, Brown  WJ, Bauman  AE.  Effect of moderate to vigorous physical activity on all-cause mortality in middle-aged and older Australians.   JAMA Intern Med. 2015;175(6):970-977. doi:10.1001/jamainternmed.2015.0541 PubMedGoogle ScholarCrossref
    37.
    Patall  EA, Cooper  H, Robinson  JC.  The effects of choice on intrinsic motivation and related outcomes: a meta-analysis of research findings.   Psychol Bull. 2008;134(2):270-300. doi:10.1037/0033-2909.134.2.270 PubMedGoogle ScholarCrossref
    38.
    Deci  EL, Ryan  RM. Intrinsic Motivation and Self-Determination in Human Behavior: Perspectives in Social Psychology. Plenum; 1985.
    39.
    Liberman  N, Förster  J.  Goal Gradients, Expectancy, and Value: Goal-Directed Behavior. Psychology Press; 2012.
    40.
    Lee  IM, Shiroma  EJ, Lobelo  F, Puska  P, Blair  SN, Katzmarzyk  PT; Lancet Physical Activity Series Working Group.  Effect of physical inactivity on major non-communicable diseases worldwide: an analysis of burden of disease and life expectancy.   Lancet. 2012;380(9838):219-229. doi:10.1016/S0140-6736(12)61031-9 PubMedGoogle ScholarCrossref
    41.
    Yang  Q, Cogswell  ME, Flanders  WD,  et al.  Trends in cardiovascular health metrics and associations with all-cause and CVD mortality among US adults.   JAMA. 2012;307(12):1273-1283. doi:10.1001/jama.2012.339 PubMedGoogle ScholarCrossref
    42.
    Saint-Maurice  PF, Troiano  RP, Bassett  DR  Jr,  et al.  Association of daily step count and step intensity with mortality among US adults.   JAMA. 2020;323(12):1151-1160. doi:10.1001/jama.2020.1382 PubMedGoogle ScholarCrossref
    43.
    Spartano  NL, Davis-Plourde  KL, Himali  JJ,  et al.  Association of accelerometer-measured light-intensity physical activity with brain volume: the Framingham Heart Study.   JAMA Netw Open. 2019;2(4):e192745. doi:10.1001/jamanetworkopen.2019.2745 PubMedGoogle Scholar
    44.
    Lee  I-M, Shiroma  EJ, Kamada  M, Bassett  DR, Matthews  CE, Buring  JE.  Association of step volume and intensity with all-cause mortality in older women.   JAMA Intern Med. 2019;179(8):1105-1112. doi:10.1001/jamainternmed.2019.0899 PubMedGoogle ScholarCrossref
    ×