Background Many patients exhibit multiple chronic disease risk behaviors. Research provides little information about advice that can maximize simultaneous health behavior changes.
Methods To test which combination of diet and activity advice maximizes healthy change, we randomized 204 adults with elevated saturated fat and low fruit and vegetable intake, high sedentary leisure time, and low physical activity to 1 of 4 treatments: increase fruit/vegetable intake and physical activity, decrease fat and sedentary leisure, decrease fat and increase physical activity, and increase fruit/vegetable intake and decrease sedentary leisure. Treatments provided 3 weeks of remote coaching supported by mobile decision support technology and financial incentives. During treatment, incentives were contingent on using the mobile device to self-monitor and attain behavioral targets; during follow-up, incentives were contingent only on recording. The outcome was standardized, composite improvement on the 4 diet and activity behaviors at the end of treatment and at 5-month follow-up.
Results Of the 204 individuals randomized, 200 (98.0%) completed follow-up. The increase fruits/vegetables and decrease sedentary leisure treatments improved more than the other 3 treatments (P[[nbsp]][[lt]][[nbsp]].001). Specifically, daily fruit/vegetable intake increased from 1.2 servings to 5.5 servings, sedentary leisure decreased from 219.2 minutes to 89.3 minutes, and saturated fat decreased from 12.0% to 9.5% of calories consumed. Differences between treatment groups were maintained through follow-up. Traditional dieting (decrease fat and increase physical activity) improved less than the other 3 treatments (P[[nbsp]][[lt]][[nbsp]].001).
Conclusions Remote coaching supported by mobile technology and financial incentives holds promise to improve diet and activity. Targeting fruits/vegetables and sedentary leisure together maximizes overall adoption and maintenance of multiple healthy behavior changes.
Nonadherence with lifestyle change advice is cited as a major barrier to effective preventive care.1,2 Many physicians express skepticism that patients will change unhealthy behaviors; they also report lack of time and training to counsel patients effectively.3,4
Poor-quality diet and inactivity are well-established behavioral risk factors for cardiovascular disease, cancer, and diabetes mellitus.5-8 Although healthy lifestyle changes can reduce morbidity and premature mortality,9-12 fewer than 25% of US adults meet the dietary guidelines,13 and 25% report no leisure-time physical activity.14 Suboptimal diet and a sedentary behavior pattern tend to cluster as risk behaviors,15-18 heightening disease risk19 and creating an opportunity to intervene comprehensively, efficiently, and, perhaps, even synergistically on more than 1 risk behavior simultaneously. However, data are sparse regarding how to change multiple lifestyle behaviors simultaneously, especially when in-person contact time is limited, as in the medical encounter.
Increasingly, patients use mobile devices to manage activities across life domains, including health.20,21 This study's interventions leveraged handheld technology to create efficient interventions that make self-monitoring more convenient, extend decision support into life contexts where lifestyle choices are made,21,22 and convey time-stamped behavioral data to paraprofessionals who provide coaching remotely. Participants received a handheld device and financial incentives initially to adopt recommended changes and subsequently to report behavior intermittently.
We designed the Make Better Choices trial to determine which combination of advice to change 1 dietary behavior (high saturated fat or low fruit and vegetable intake) and 1 activity behavior (high sedentary leisure or low physical activity) would maximize healthy diet and activity change during treatment and follow-up. Per behavioral choice theory,23,24 we predicted that increasing intake of fruits/vegetables and decreasing sedentary leisure would maximize healthy change by fostering healthy substitution (of fruits/vegetables for fat and physical activity for sedentary leisure) and complementary behavior change (decreased fat accompanying decreased sedentary leisure). We also tested 2 alternative predictions about which treatment would maximize healthy change: traditional dieting (decrease fat and increase physical activity) because of its familiarity or increase healthy behaviors (increase fruits/vegetables and physical activity) because it requires the least inhibition of rewarding behaviors.24
The study design and methods are discussed in detail elsewhere24 and are described briefly herein.
Adults aged 21 to 60 years were recruited through community advertisements. Eligible individuals were required to report all of the following: (1) intake of fewer than 5 fruits/vegetables daily,25,26 (2) greater than 8% caloric intake from saturated fat, (3) less than 60 min/d of moderate or vigorous physical activity, and (4) greater than 90 min/d of sedentary leisure (television, movies, recreational Internet use, and video games).
Figure 1 diagrams participant flow through the trial. All the procedures were approved by the institutional review boards of the University of Illinois at Chicago and Northwestern University.
2-week baseline phase (and final eligibility screening)
Candidates who self-reported all 4 risk behaviors were screened by a bachelor[[rsquo]]s-level research assistant (coach: J.V., M.S., or A.D.). The coach trained participants to estimate accurately and to use a handheld device to record and upload dietary intake, moderate- to vigorous-intensity physical activity, and targeted sedentary leisure. During the 2-week baseline phase, participants wore an accelerometer, recorded diet and activity on the handheld device, and uploaded data daily.
Candidates who displayed all 4 risk behaviors, as evidenced by handheld device and accelerometer data, were randomized (stratified by sex) using a computer-generated sequence of randomly permuted blocks.
Intervention (behavioral treatment) phase
Coaches tailored behavioral strategies based on participants' baseline data. For example, those asked to decrease fat were shown 10 foods that supplied their greatest saturated fat grams and were coached to reduce portion size or number for those foods. For the first week of treatment, daily goals were set midway between the baseline behavior and the ultimate daily goal. Beginning the second treatment week, full goals were set for the 2 targeted behaviors to which the participant was randomized: 5 fruit/vegetable servings, saturated fat intake less than 8% of calories, physical activity of at least 60 min/d, or sedentary leisure less than 90 min/d. Participants were expected to reach their behavioral targets during treatment week 2 and to maintain them during week 3. During the 3 treatment weeks, participants uploaded data daily and communicated as needed with their coaches via telephone or e-mail, per preference, to overcome challenges. Participants could earn a $175 incentive for meeting the goals for both targeted behaviors during the treatment phase.
To explore the potential for maintenance of healthy behavior changes, the study included a 20-week follow-up. Immediately after the treatment period, participants were informed that attainment of diet and activity targets was no longer required and that payment was now contingent solely on recording and transmitting handheld device data on a predetermined schedule. Recording was required daily for the first week after treatment, for 3 consecutive days in posttreatment weeks 2 and 3, biweekly for the next 6 weeks, and then monthly until final follow-up. Participants could earn incrementally larger financial incentives (from $30 to $80) for uploading data during consecutive follow-ups. All the incentives were received at the end of follow-up.
Participants used a personal digital assistant to record and self-regulate their targeted behaviors. They were advised to carry the device and record immediately after executing a behavior. During treatment and follow-up, the handheld device displayed 2 decision support feedback [[ldquo]]thermometers[[rdquo]]: 1 for diet (fruits/vegetables or fat) and 1 for activity (physical activity or sedentary leisure)
(eFigure and
eAppendix). Once activated, goal thermometers were continually updated in response to data entry. They also enabled participants to look up the potential impact of a food or activity choice.
Outcomes were assessed by daily self-report recordings on the handheld device. Fat and fruit/vegetable consumption were measured from dietary intake recordings. To prevent superfluous calories (eg, in sweetened beverages) from inflating the fat gram allowance, the saturated fat goal for those randomized to decrease fat was determined using the Harris-Benedict equation27 to estimate calories needed to maintain weight. Minutes of physical and sedentary activity were measured cumulatively by an end-of-day 24-hour activity log in which participants accounted for every 15-minute block of each day. Previous studies have established the validity of self-report diet and activity assessments administered via a mobile device.28-32 To further encourage honest recording, we implemented a validated bogus pipeline protocol33 whereby participants submitted grocery receipts, accelerometer data, and urine samples that they believed would be used to evaluate their self-reports.
Composite diet-activity improvement score
Rather than measure only single behaviors, we assessed simultaneous overall change on all 4 behaviors. To place the 4 behaviors (fruits/vegetables, fat, physical activity, and sedentary leisure) on a common scale to quantify overall change, we developed a [[ldquo]]Composite Diet-Activity Improvement Score[[rdquo]] that weighted each behavior equally. We transformed all the variables to better approximate normality by using square root transformation for the count outcomes (fruits/vegetables, physical activity, and sedentary leisure) and arc sine transformation for the percentage outcome (fat).34 Then we standardized each individual health behavior using a modified z score (where 1 U represents a 1-SD change), with higher values representing greater healthy lifestyle improvement. To reflect improvement relative to baseline, we standardized z scores for time points after baseline relative to the overall baseline distribution. We calculated the mean of all 4 individual z scores at each time point, as recommended,35 to derive a composite index that expressed each participant's overall healthy change across multiple diet and activity behaviors.
The primary analytic aim was to determine which behavioral treatment maximizes initiation of healthy diet and activity changes, measured as improvement from baseline through the treatment phase. The secondary aim was to examine maintenance of change on the Composite Diet-Activity Improvement Score. A sample size of 200 (50 per treatment) was projected to yield power of 0.85 to detect a 0.5-SD difference in diet and activity improvement between treatments.
Statistical analyses were performed using a commercially available software program (SAS, version 9.2; SAS Institute, Inc). Study hypotheses were tested using 3 a priori planned contrasts comparing the predicted best treatment with all others combined: (1) decrease fat and increase physical activity vs others combined, (2) increase fruit/vegetable intake and physical activity vs others combined, and (3) increase fruit/vegetable intake and decrease sedentary vs others combined. The analyses used a linear mixed model for longitudinal data,36 with the Composite Diet-Activity Improvement Score as the dependent variable. For the within-participants factor of time (baseline, intervention week 1 and weeks 2 and 3, and follow-ups 1-8), we tested 2 comparisons for the intervention phase (intervention week 1 vs baseline and intervention weeks 2 and 3 vs week 1) and 3 for the follow-up phase (an average of all follow-ups vs the final treatment phase time point and linear and quadratic trends during follow-up). These 5 time comparisons are included in all the analyses. We also created interaction terms for each hypothesis by each time contrast. Inferences focused on treatment [[times]] time interactions, which compare the difference in behavior change over time between treatments, specifically testing for the components of the treatment [[times]] time interaction in a hierarchical manner (ie, treatment [[times]] follow-up first, then treatment [[times]] intervention phase). For the variance-covariance structure of the longitudinal data, we used a heterogeneous Toeplitz structure36 that allowed the variances to vary across time and the correlations to vary across time lags. All the analyses were performed on an intention-to-treat basis and included sex as a covariate. Across the 11 time points, the proportion of missing data ranged from 4% to 11%.
We used separate mixed-effects regression models to predict each participant's change from baseline through the treatment phase for each of the 4 behaviors. We then correlated these 4 individual change estimates to examine whether changes in targeted behaviors were correlated with changes in untargeted behaviors.
The final sample of 204 adults included 48 men, 46.6% with nonwhite race/ethnicity, 26.0% with no more than a high school education, and a mean (SD) age of 33 (11) years. Demographic information is summarized in Table 1. Except for 1 individual, all the participants attained the behavioral targets; most did so promptly. The median time taken to achieve consumption of 5 fruits/vegetables was 9 days (ie, 2 days after the 5 fruits/vegetables goal was set). The median time taken to attain the sedentary, fat, and physical activity targets was 8 days (ie, 1 day after the targeted amount was set).
Treatment effects on composite diet-activity improvement scores
The treatment [[times]] intervention phase interaction was highly significant ([[khgr]]26[[nbsp]]=[[nbsp]]38.1, P[[nbsp]][[lt]][[nbsp]].001), indicating that the treatment groups differed significantly on the Composite Diet-Activity Improvement Score. Furthermore, this treatment effect remained significant throughout the subsequent follow-up period (observed means are graphed in Figure 2A).
The increase fruits/vegetables and decrease sedentary leisure treatment increased Composite Diet-Activity Improvement Scores more than the alternative treatments in the first week of intervention (t383[[nbsp]]=[[nbsp]]3.16, P[[nbsp]]=[[nbsp]].002). The superiority of the increase fruits/vegetables and decrease sedentary leisure treatment, evident after 1 week, was maintained through the end of the 3-week treatment period and through follow-up week 20 (Figure 2B).
The decrease fat and increase physical activity treatment decreased Composite Diet-Activity Improvement Scores compared with other treatments in the first week of intervention (t382[[nbsp]]=[[nbsp]][[minus]]2.17, P[[nbsp]]=[[nbsp]].03), a disadvantage that persisted through the end of treatment and follow-up.
Treatment effects on individual diet and activity behaviors
Figure 3 shows the effect of the increase fruits/vegetables and decrease sedentary leisure treatment on standardized improvement over time in the 4 behaviors. Raw (unstandardized) change is included, indicating that mean (SD) fruit/vegetable intake changed from 1.2 (0.9) servings per day at baseline to 5.5 (1.0) servings per day at the end of treatment and 2.9 (2.3) servings per day at the end of follow-up. Mean (SD) minutes per day of sedentary leisure changed from 219.2 (93.8) at baseline to 89.3 (65.5) at the end of treatment and 125.7 (108.7) at the end of follow-up. Mean (SD) daily calories from saturated fat changed from 12.0% (2.2%) at baseline to 9.4% (1.9%) at the end of treatment and 9.9% (3.4%) at the end of follow-up.
Group means for each behavior in natural units (ie, minutes per day, daily servings, or percentage of daily caloric intake) are presented in Table 2 for all treatments and are graphed in Figure 4.
In the increase fruits/vegetables and decrease sedentary leisure treatment group, we examined correlations among the individual behavior change estimates to determine whether changes in targeted behaviors were correlated with changes in untargeted behaviors. The degree to which participants decreased their sedentary leisure time (targeted) correlated positively with the degree to which they also reduced their fat intake (untargeted) (r52[[nbsp]]=[[nbsp]]0.29, P[[nbsp]]=[[nbsp]].04). Correlations among other behavior change pairs were nonsignificant and ranged from r52[[nbsp]]=[[nbsp]][[ndash]]0.14 to 0.15.
This study demonstrates the feasibility of changing multiple unhealthy diet and activity behaviors simultaneously, efficiently, and with minimal face-to-face contact by using mobile technology, remote coaching, and incentives. The increase fruits/vegetables and decrease sedentary leisure treatment maximized healthy lifestyle change compared with the other interventions. In addition to producing targeted improvements in intake of fruits/vegetables and sedentary leisure time, the treatment produced untargeted improvement in saturated fat intake. The superiority of the increase fruits/vegetables and decrease sedentary leisure treatment present after 1 week of intervention persisted through the end of the 3-week treatment phase and was maintained. As expected, since participants were no longer asked to maintain healthy changes, lifestyle gains diminished once treatment ended. Nevertheless, substantial improvements (1 SD compared with baseline) in fruits/vegetables, sedentary leisure, and fat persisted through the 5-month follow-up. From baseline to the end of treatment to the end of follow-up, respectively, mean servings per day of fruits/vegetables changed from 1.2 to 5.5 to 2.9, mean minutes per day of sedentary leisure from 219.2 to 89.3 to 125.7, and daily calories from saturated fat from 12.0% to 9.4% to 9.9%. Although they were neither asked nor reinforced to maintain eating or activity improvements, 86.5% of the 185 participants from whom exit interviews were obtained said they [[ldquo]]definitely[[rdquo]] or [[ldquo]]somewhat[[rdquo]] tried to maintain gains.
Consistent with behavioral choice theory,23,24 decreasing recreational screen time was complemented by a reduction in saturated fat intake. Previous research has demonstrated that manipulating screen time changes energy and fat intake in children.37 This is the first study to show that reducing sedentary leisure time decreases fat intake in adults. Increased interest in sedentary behavior has been driven by epidemiologic data associating sedentarism with cancer,38 obesity,39,40 type 2 diabetes mellitus,19,39-41 depression,42 cardiovascular events,19,43 and increased mortality,19,43,44 independent of physical activity. Reduced screen time may be an important behavioral target not only to reverse direct adverse effects of prolonged sitting45 but also to disrupt pairing of screen time with high-fat snacking.
The traditional dieting regimen (decrease fat and increase physical activity) produced less healthy change than did the other treatments: it improved only 2 behaviors, and increased physical activity did not persist. The requirement to inhibit rewarding behaviors had no systematic effect on diet-activity change: the increase fruits/vegetables and physical activity treatment was no more advantageous than were the other treatments, and the decrease fat and sedentary leisure treatment was no more disadvantageous.24
These results are germane to physicians trying to help patients improve multiple health risk behaviors. Physicians play an important role by advising and assisting patients to accomplish healthy behavioral changes,9,46 especially since a trusting relationship with a provider is associated with greater adherence to advice.3 However, limits on physicians' time combined with movement toward new systems of patient-centered, team-based care47 creates an opportunity to reconsider the optimal locus and configuration of health behavior change counseling. Results suggest feasibility and potential benefit of a systems reconfiguration that reinforces health behavior change by connecting patients with mobile technology, incentives, and remote, nonphysician coaches.
Several study limitations warrant consideration. Generalizability of the findings is limited by the constraints that the study was conducted in a research setting, and only a quarter of the sample was male. Use of a screening phase to confirm the presence of the risk behaviors may additionally limit generalizability to those with entrenched unhealthy diet and activity behaviors. Also, the amount of the financial incentive was larger than would be feasible for some settings. It remains to be determined whether such rapid and full acquisition of behavior change targets would occur with smaller incentives. Furthermore, the fact that primary outcome measures were self-reported raises the possibility that participants might have overstated their behavioral improvements to earn incentives. We find that unlikely for several reasons. First, treatment differences remained after controlling for the effects of financial motivation and social desirability. Second, the sample ranked financial motives lowest among their reasons to join the trial. Third, maintaining diet and activity improvements yielded no financial reward during follow-up, an altered contingency made apparent to participants by staff reminders and by discontinuation of study procedures (urine samples, accelerometry, and grocery receipts) that could have verified self-reports. Yet, participants maintained substantial improvements, and most said they did so intentionally.
Finally, although physical activity was increased by treatments that targeted it, it was the one behavior not improved by the increase fruits/vegetables and decrease sedentary leisure treatment. We currently are testing whether all 4 risk behaviors can be improved by targeting physical activity simultaneously or sequentially with fruits/vegetables and sedentary leisure.
The strengths of the study include the ethnic diversity of the sample and minimal loss to follow-up. Also, the sample was deliberately chosen to present challenges for behavior change. The requirement that participants unremittingly display all 4 risk behaviors throughout baseline screening, even while self-monitoring, yielded a sample with risk behaviors that were refractory to lower-intensity behavioral intervention. A key innovation was the use of mobile technologies that connect and provide decision support to patients and coaches, reducing the need for professionals to perform counseling. Another strength was the conservative, comparative research design that contrasts active treatments.
Interventions that target multiple, prevalent, covariant risk behaviors simultaneously have the potential to be powerfully efficient and cost-effective. Yet, many multiple behavior change interventions have achieved limited success,48-50 presumably because their interventions were insufficiently intensive.51 As mobile technologies become increasingly ubiquitous, they afford a scalable platform to extend continuing support for healthy behavior change pervasively into the environment, with the potential to improve population health.
Correspondence: Bonnie Spring, PhD, Department of Preventive Medicine, Northwestern University, 680 N Lake Shore Dr, Ste 1220, Chicago, IL 60611 (bspring@northwestern.edu).
Accepted for Publication: February 20, 2012.
Author Contributions: Dr Spring had full access to all the study data and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Spring[[cedil]] Schneider, McFadden, and Hedeker. Acquisition of data: Spring, Schneider, McFadden, Vaughn, Kozak, Smith, and DeMott. Analysis and interpretation of data: Spring, Schneider, McFadden, Vaughn, Moller, Epstein, Hedeker, Siddique, and Lloyd-Jones. Drafting of the manuscript: Spring, Schneider, McFadden, Vaughn, Kozak, Smith, Moller, DeMott, Hedeker, Siddique, and Lloyd-Jones. Critical revision of the manuscript for intellectual content: Spring, Schneider, McFadden, Vaughn, Kozak, Moller, Epstein, Hedeker, Siddique, and Lloyd-Jones. Statistical analysis: Spring, Schneider, McFadden, Moller, Epstein, Hedeker, and Siddique. Obtained funding: Spring. Administrative, technical, or material support: Spring, McFadden, Vaughn, Smith, and DeMott. Study supervision: Spring, Kozak, and Lloyd-Jones.
Financial Disclosure: None reported.
Funding/Support: The Make Better Choices trial was supported by grant HL075451 from the National Institutes of Health (NIH) (Dr Spring), the Robert H. Lurie Comprehensive Cancer Center grant (NIH P30 CA060553), and grant F31 MH070107 from the NIH (Dr Schneider).
Additional Contributions: Michael J. Coons, PhD, Sherry Pagoto, PhD, Christine Dutton Pellegrini, PhD, and Alex Pictor, BA, provided technical assistance.
1.Colditz GA, Taylor PR. Prevention trials: their place in how we understand the value of prevention strategies.
Annu Rev Public Health. 2010;31:105-12020070190
PubMedGoogle ScholarCrossref 2.Smith TW, Orleans CT, Jenkins CD. Prevention and health promotion: decades of progress, new challenges, and an emerging agenda.
Health Psychol. 2004;23(2):126-13115008655
PubMedGoogle ScholarCrossref 3.Bazata DD, Robinson JG, Fox KM, Grandy S.SHIELD Study Group. Affecting behavior change in individuals with diabetes: findings from the Study to Help Improve Early Evaluation and Management of Risk Factors Leading to Diabetes (SHIELD).
Diabetes Educ. 2008;34(6):1025-103619075084
PubMedGoogle ScholarCrossref 4.Vogt F, Hall S, Marteau TM. General practitioners' and family physicians' negative beliefs and attitudes towards discussing smoking cessation with patients: a systematic review.
Addiction. 2005;100(10):1423-143116185204
PubMedGoogle ScholarCrossref 5.Mokdad AH, Marks JS, Stroup DF, Gerberding JL. Actual causes of death in the United States, 2000.
JAMA. 2004;291(10):1238-124515010446
PubMedGoogle ScholarCrossref 6.US Department of Agriculture, US Department of Health and Human Services. Dietary Guidelines for Americans. 7th ed. Washington, DC: US Government Printing Office; 2010
7.Lloyd-Jones DM, Hong Y, Labarthe D,
et al; American Heart Association Strategic Planning Task Force and Statistics Committee. Defining and setting national goals for cardiovascular health promotion and disease reduction: the American Heart Association's strategic Impact Goal through 2020 and beyond.
Circulation. 2010;121(4):586-61320089546
PubMedGoogle ScholarCrossref 9.Lin JS, O[[rsquo]]Connor E, Whitlock EP, Beil TL. Behavioral counseling to promote physical activity and a healthful diet to prevent cardiovascular disease in adults: a systematic review for the U.S. Preventive Services Task Force.
Ann Intern Med. 2010;153(11):736-75021135297
PubMedGoogle Scholar 10.Knowler WC, Barrett-Connor E, Fowler SE,
et al; Diabetes Prevention Program Research Group. Reduction in the incidence of type 2 diabetes with lifestyle intervention or metformin.
N Engl J Med. 2002;346(6):393-40311832527
PubMedGoogle ScholarCrossref 11.Svetkey LP, Harsha DW, Vollmer WM,
et al. PREMIER: a clinical trial of comprehensive lifestyle modification for blood pressure control: rationale, design and baseline characteristics.
Ann Epidemiol. 2003;13(6):462-47112875806
PubMedGoogle ScholarCrossref 12.Lindstr[[ouml]]m J, Ilanne-Parikka P, Peltonen M,
et al; Finnish Diabetes Prevention Study Group. Sustained reduction in the incidence of type 2 diabetes by lifestyle intervention: follow-up of the Finnish Diabetes Prevention Study.
Lancet. 2006;368(9548):1673-167917098085
PubMedGoogle ScholarCrossref 13.Centers for Disease Control and Prevention. Prevalence and trends data 2009: Behavioral Risk Factor Surveillance System.
2011. http://apps.nccd.cdc.gov/BRFSS. Accessed May 30, 2011 14.Centers for Disease Control and Prevention. State Indicator Report on Physical Activity. Atlanta, GA: US Dept of Health and Human Services; 2010
15.Emmons KM, Marcus BH, Linnan L, Rossi JS, Abrams DB.Working Well Research Group. Mechanisms in multiple risk factor interventions: smoking, physical activity, and dietary fat intake among manufacturing workers.
Prev Med. 1994;23(4):481-4897971876
PubMedGoogle ScholarCrossref 16.Fine LJ, Philogene GS, Gramling R, Coups EJ, Sinha S. Prevalence of multiple chronic disease risk factors: 2001 National Health Interview Survey.
Am J Prev Med. 2004;27(2):(suppl)
18-2415275670
PubMedGoogle ScholarCrossref 17.Gillman MW, Pinto BM, Tennstedt S, Glanz K, Marcus B, Friedman RH. Relationships of physical activity with dietary behaviors among adults.
Prev Med. 2001;32(3):295-30111277687
PubMedGoogle ScholarCrossref 18.O'Brien K. Living Dangerously: Australians With Multiple Risk Factors for Cardiovascular Disease: Bulletin No. 24. Canberra: Australian Institute of Health and Welfare; 2005
19.Gr[[oslash]]ntved A, Hu FB. Television viewing and risk of type 2 diabetes, cardiovascular disease, and all-cause mortality: a meta-analysis.
JAMA. 2011;305(23):2448-245521673296
PubMedGoogle ScholarCrossref 20.Pew Research Center. Internet and American Life Project: Tracking Survey. Washington, DC: Pew Research Center; 2011
21.Baumgart DC. Smartphones in clinical practice, medical education, and research.
Arch Intern Med. 2011;171(14):1294-129621788549
PubMedGoogle ScholarCrossref 23.Bickel WK, ed, Vuchinich RE, ed. Reframing Health Behavior Change With Behavioral Economics. Mahwah, NJ: Lawrence Erlbaum Associates, Inc; 2000
24.Spring B, Schneider K, McFadden HG,
et al. Make Better Choices (MBC): study design of a randomized controlled trial testing optimal technology-supported change in multiple diet and physical activity risk behaviors.
BMC Public Health. 2010;10:58620920275
PubMedGoogle ScholarCrossref 25.National Cancer Institute. 5 a Day for Better Health Program. Bethesda, MD: National Cancer Institute, National Institutes of Health; 2001
26.US Department of Agriculture, US Department of Health and Human Services. Nutrition and Your Health: Dietary Guidelines for Americans. Washington, DC: US Government Printing Office; 1995
27.Harris J, Benedict F. A Biometric Study of Basal Metabolism in Man. Washington, DC: Carnegie Institute of Washington; 1919
29.Fukuoka Y, Kamitani E, Dracup K, Jong SS. New insights into compliance with a mobile phone diary and pedometer use in sedentary women.
J Phys Act Health. 2011;8(3):398-40321487139
PubMedGoogle Scholar 30.Sternfeld B, Jiang SF, Picchi T, Chasan-Taber L, Ainsworth B, Quesenberry CP Jr. Evaluation of a cell phone-based physical activity diary.
Med Sci Sports Exerc. 2012;44(3):487-49521857369
PubMedGoogle ScholarCrossref 31.Clark BK, Sugiyama T, Healy GN, Salmon J, Dunstan DW, Owen N. Validity and reliability of measures of television viewing time and other non-occupational sedentary behaviour of adults: a review.
Obes Rev. 2009;10(1):7-1618631161
PubMedGoogle ScholarCrossref 32.Long JD, Littlefield LA, Estep G,
et al. Evidence review of technology and dietary assessment.
Worldviews Evid Based Nurs. 2010;7(4):191-20419804586
PubMedGoogle ScholarCrossref 33.Roese NJ, Jamieson DW. Twenty years of bogus pipeline research: a critical review and meta-analysis.
Psychol Bull. 1993;114(2):363-375
Google ScholarCrossref 34.Mosteller F, Tukey JW. Data Analysis and Regression: A Second Course in Statistics. Reading, MA: Addison-Wesley; 1977
35.Prochaska JJ, Velicer WF, Nigg CR, Prochaska JO. Methods of quantifying change in multiple risk factor interventions.
Prev Med. 2008;46(3):260-26518319099
PubMedGoogle ScholarCrossref 36.Hedeker D, Gibbons RD. Longitudinal Data Analysis. Hoboken, NJ: John Wiley [[amp]] Sons Inc; 2006
37.Epstein LH, Roemmich JN, Paluch RA, Raynor HA. Influence of changes in sedentary behavior on energy and macronutrient intake in youth.
Am J Clin Nutr. 2005;81(2):361-36615699222
PubMedGoogle Scholar 38.Gierach GL, Chang SC, Brinton LA,
et al. Physical activity, sedentary behavior, and endometrial cancer risk in the NIH-AARP Diet and Health Study.
Int J Cancer. 2009;124(9):2139-214719123463
PubMedGoogle ScholarCrossref 39.Hu FB, Li TY, Colditz GA, Willett WC, Manson JE. Television watching and other sedentary behaviors in relation to risk of obesity and type 2 diabetes mellitus in women.
JAMA. 2003;289(14):1785-179112684356
PubMedGoogle ScholarCrossref 40.Healy GN, Wijndaele K, Dunstan DW,
et al. Objectively measured sedentary time, physical activity, and metabolic risk: the Australian Diabetes, Obesity and Lifestyle Study (AusDiab).
Diabetes Care. 2008;31(2):369-37118000181
PubMedGoogle ScholarCrossref 41.Hu FB, Manson JE, Stampfer MJ,
et al. Diet, lifestyle, and the risk of type 2 diabetes mellitus in women.
N Engl J Med. 2001;345(11):790-79711556298
PubMedGoogle ScholarCrossref 42.Hamer M, Stamatakis E, Mishra GD. Television- and screen-based activity and mental well-being in adults.
Am J Prev Med. 2010;38(4):375-38020307805
PubMedGoogle ScholarCrossref 43.Stamatakis E, Hamer M, Dunstan DW. Screen-based entertainment time, all-cause mortality, and cardiovascular events: population-based study with ongoing mortality and hospital events follow-up.
J Am Coll Cardiol. 2011;57(3):292-29921232666
PubMedGoogle ScholarCrossref 44.Patel AV, Bernstein L, Deka A,
et al. Leisure time spent sitting in relation to total mortality in a prospective cohort of US adults.
Am J Epidemiol. 2010;172(4):419-42920650954
PubMedGoogle ScholarCrossref 45.Hamilton MT, Healy GN, Dunstan DR, Zderic TW, Owen N. Too little exercise and too much sitting: inactivity physiology and the need for new recommendations on sedentary behavior.
Curr Cardiovasc Risk Rep. 2009;2(8):292-298
Google Scholar 46.Goldstein MG, Whitlock EP, DePue J.Planning Committee of the Addressing Multiple Behavioral Risk Factors in Primary Care Project. Multiple behavioral risk factor interventions in primary care: summary of research evidence.
Am J Prev Med. 2004;27(2):(suppl)
61-7915275675
PubMedGoogle ScholarCrossref 47.Schuetz B, Mann E, Everett W. Educating health professionals collaboratively for team-based primary care.
Health Aff (Millwood). 2010;29(8):1476-148020679650
PubMedGoogle ScholarCrossref 48.Multiple Risk Factor Intervention Trial Research Group. Multiple Risk Factor Intervention Trial: risk factor changes and mortality results.
JAMA. 1982;248(12):1465-14777050440
PubMedGoogle ScholarCrossref 49.Fortmann SP, Varady AN. Effects of a community-wide health education program on cardiovascular disease morbidity and mortality: the Stanford Five-City Project.
Am J Epidemiol. 2000;152(4):316-32310968376
PubMedGoogle ScholarCrossref 50.Luepker RV, R[[aring]]stam L, Hannan PJ,
et al. Community education for cardiovascular disease prevention: morbidity and mortality results from the Minnesota Heart Health Program.
Am J Epidemiol. 1996;144(4):351-3628712192
PubMedGoogle ScholarCrossref 51.Spring B. Translational behavioral medicine: a pathway to better health.
Translational Behavioral Med. 2011;1:1-3
Google ScholarCrossref