Controlled outcome evaluation of the First Step Program: a daily physical activity intervention for individuals with type II diabetes

Transcription

1 (2004) 28, & 2004 Nature Publishing Group All rights reserved /04 $ PAPER Controlled outcome evaluation of the First Step Program: a daily physical activity intervention for individuals with type II diabetes C Tudor-Locke 1,2 *, RC Bell 3, AM Myers 2,4, SB Harris 5, NA Ecclestone 2, N Lauzon 2,4 and NW Rodger 2,6,7 1 Department Exercise and Wellness, Arizona State University East, Mesa, AZ, USA; 2 Canadian Centre for Activity and Aging, The University of Western Ontario and St Joseph s Health Centre, London, Ontario, Canada; 3 Agriculture, Food and Nutrition Sciences, The University of Alberta, Edmonton, Alberta, Canada; 4 Department of Health Studies & Gerontology, University of Waterloo, Waterloo, Ontario, Canada; 5 Centre for Studies in Family Medicine, The University of Western Ontario, London, Canada; 6 The Lawson Diabetes Centre, The University of Western Ontario and St Joseph s Health Care, London, Ontario, Canada; and 7 Division of Endocrinology, St Joseph s Health Care and The University of Western Ontario, London, Canada OBJECTIVE: To conduct a randomised trial of a physical activity (PA) intervention, The First Step Program (FSP) for adults with type II diabetes. DESIGN: A 16-week intervention study and 24-week follow-up assessment. PARTICIPANTS: A total of 47 overweight/obese, sedentary individuals (age ¼ y; BMI ¼ kg/m 2 ) recruited through a diabetes education centre. MEASUREMENTS: Primary outcome: daily PA assessed by pedometer (steps/day). Secondary outcomes: anthropometric measures (weight, BMI, waist girth, hip girth); indicators of cardiovascular health (resting heart rate and blood pressure); glycemic control (fasting glucose, insulin, HbA1c, glucose concentration 120 min postglucose load); plasma lipid status (total cholesterol, HDL cholesterol, LDL cholesterol, and triglycerides). RESULTS: Relative to the CONTROL group, FSP participants increased their PA steps/day (approximately 30 min/day) during the intervention (Po0.0001). Waist and hip girth decreased (approximately 2 3 cm), but did not differ significantly between groups. Significant changes did not emerge for any of the other variables. CONCLUSIONS: The FSP is a practical intervention that elicits an immediate and profound change in walking behaviour. Such change is an important first step towards increasing the volume and/or intensity of PA necessary to improve long-term health outcomes in this largely sedentary and overweight or obese population. Relapse by 24 weeks indicates that other strategies such as booster sessions are needed to maintain lifestyle change. Further research must determine realistic and responsive health outcomes for this population that are achievable through practical, real-world programming. (2004) 28, doi: /sj.ijo Published online 21 October 2003 Keywords: walking exercise; behaviour modification; pedometer; goal-setting Introduction While exercise is considered a cornerstone in the treatment and management of type II diabetes, a substantial proportion of this population is not physically active. A national survey found that 66% of individuals with diabetes reported no regular leisure-time physical activity (PA). 1 Another study 2 *Correspondence: C Tudor-Locke, Department Exercise and Wellness, Arizona State University East, Mesa, AZ 85212, USA,. Tudor-Locke@asu.edu Received 31 January 2003; revised 27 June 2003; accepted 3 August 2003 found almost 55% of those with type II diabetes reported no weekly PA via walking. The importance of regular PA for this population cannot be overstated. Low cardiovascular fitness and physical inactivity are independent predictors of all-cause mortality in men with type II diabetes. 3 Regular PA may contribute to improvements in body composition, blood lipid profiles, hypertension, and glycemic control 4 and reduce the risk of coronary heart disease by as much as 55%. 5 Physical inactivity likely contributes to many of the functional limitations or disabilities observed in this group. For

2 114 instance, 50% of women and 32% of men over age 60 y have difficulty or are unable to walk a quarter mile. 6 The American Diabetes Association continues to endorse the US Surgeon General s public health recommendation 7 that all adults should accumulate 30 min or more of moderate intensity PA (equivalent to brisk walking) on most, if not all days of the week. Although persons with type II diabetes believe they should get more exercise, most are not interested in joining formal exercise programs. 8 Rather, they desire activity options tailored to individual abilities, lifestyle and preferences. 9 The First Step Program (FSP), developed and evaluated systematically and in collaboration with diabetes educators and adults with type II diabetes, is described in detail elsewhere Briefly, the FSP is a facilitated, behaviour modification program based on the theoretical principles of self-efficacy and social support, the common clinical practices of goal-setting, self-monitoring and feedback, as well as the simple premise that walking can be gradually increased throughout the day at work and home and during leisure. The FSP comprises four weekly group meetings (adoption phase), followed by a 12-week adherence phase. Inspired by the initial work of Yamanouchi et al, 13 the FSP uses a pedometer as a motivational and monitoring tool. In the FSP, pedometers are used to establish baseline level of PA, show participants how many steps they normally take in a given amount of time (eg, 10 min of walking), and facilitate personal goal-setting, self-monitoring and feedback. Pedometers correlate (r ¼ ) with more expensive accelerometers in controlled and field conditions. 14,15 Formative evaluation showed that the FSP is a feasible and acceptable approach for adults with type II diabetes. 12 A preliminary outcome evaluation with a small sample suggested that the FSP may be effective in increasing level of PA and improving blood pressure and waist girth. 11 The current study examined the effectiveness of the FSP with a larger sample and a control group. In addition to the primary outcome of PA (measured as steps/day), we also examined whether increased PA was related to improvements in cardiovascular health, glycemic control and lipid profiles. of steps/day was established during pilot testing and is corroborated by research that indicates that individuals who take more than E9000 steps/day (75th percentile of distribution) are more likely to be classified as normal weight, and therefore sufficiently active. 17 Medical history, including diagnosis and treatment for diabetes, were obtained from chart review. Participant flow through the study is depicted in Figure 1. A pool of 146 individuals was identified as eligible to participate; 137 (94%) of whom could be reached by telephone. A total of 10 people (7%) contacted were found to be in exercise classes, thus ineligible. Potential participants were told about the study requirements, including randomization to either the FSP or a wait-list control group. In all, 60 individuals (47%) accepted our invitation to participate while 67 (53%) did not. Lack of time was the primary reason provided by those who chose not to participate (64%); additional reasons are detailed in Figure 1. Those individuals who agreed to participate were randomly assigned to either the FSP group or the wait-list control group (CONTROL). With the person s permission, their family physician was notified about the programme and study participation. Ethical approval was obtained from the university review board for research involving human subjects. Assessment procedures Study participants were assessed at baseline and 16 weeks later (completion of the adherence phase for the FSP group). At this point, study participants were asked to return their pedometers and there was no further contact until they were Contacted about study (n=137) Interested in participating (n=60); Randomized at baseline CONTROL group (n=30) Eligible participants (n=146) Unable to contact (n=9) Not participating (n=77); No time (n=49), Participating in exercise program (n=10), Not interested (n=7), Transportation issue (n=4), Family issues (n=1), Not specified (n=6). FSP group (n=30); First Step Program Methods Study design and recruitment Participants were recruited from a diabetes education centre (the Lawson Diabetes Centre in London, Ontario) based on these criteria: (1) aged y; (2) minimum 3 months postdiagnosis of type II diabetes; (3) treated by diet alone or by oral hypoglycaemic medications (not insulin); (4) no PA limitations or documented heart conditions; (5) not currently in an exercise program; and (6) o8800 steps/day (defined as insufficiently active determined via a 3-day, blinded pedometer protocol 16 as described below). This level Motivational postcard at Weeks 6 & 10: "Thanks for being in our study". Post-intervention assessment at Week 16 (n=23). Figure 1 Follow-up assessment at Week 24 (n=22). Week 1-4; group meetings once per week; individually record pedometer values daily. Weeks 5-16: no group meetings, individually record pedometer values daily. Motivational postcard at Weeks 6 & 10: "Thanks for trying the First Step Program". Post-intervention assessment at Week 16 (n=24). Follow-up assessment at Week 24 (n=16). Progress of participants through outcome evaluation of the FSP.

3 invited to a follow-up assessment at 24 weeks. Each assessment was identical and included: (1) PA using a pedometer (2) anthropometric measurements; (3) resting heart rate and blood pressure; (4) an oral glucose tolerance test; and (5) blood samples for determination of insulin, haemoglobin A1c (HbA1c) and plasma lipid profiles. Baseline levels of PA were determined during the initial recruitment process, followed by the rest of the assessment. This order was reversed at 16 and 24 weeks; at the face-toface assessment, study participants were given a package containing a new pedometer and a padded, prestamped, addressed envelope (see below). To assess PA (defined as steps/day), each person wore a pedometer (Yamax SW-200, Yamax Corporation, Tokyo, Japan) over three consecutive days (including one weekend day) during their waking hours while engaging in usual activities. 11,16 These pedometers record within 1% of all steps taken under controlled conditions 18 and are considered valid for assessing PA in obese individuals. 19 Study participants were trained regarding proper placement and use of the pedometer. Each person was instructed to set the pedometer to zero the first morning, then seal the cover with a sticker. Pedometers were returned to the study centre after the third day of wear in the envelope provided. Upon receipt at the study centre, pedometer seals were checked. Steps were adjusted according to a correction factor (803 steps; based on pilot findings) to account for steps added during mailing. 11,16,20 Mean steps/day was calculated as the corrected total steps divided by 3 days of wear. No participants reported engaging in activities that could not be detected by a pedometer (eg swimming, cycling). Aside from the pedometer measures, all other assessments were conducted at the diabetes education centre. Body mass and height were measured wearing street clothing and without shoes. Body mass index (BMI) was calculated (kg/ m 2 ). Waist and hip girths were measured using standard protocols over light clothing. 21 Resting heart rate and blood pressure were measured by a registered nurse. For the determination of oral glucose tolerance, participants were asked to take medications as usual, fast overnight (minimum 10 h) and abstain from caffeine, alcohol and smoking for 12 h prior. A 2-h, 75 g oral glucose tolerance test was performed. Blood glucose, total cholesterol, HDL cholesterol, LDL cholesterol and triglycerides were determined using an autoanalyzer (Beckman-Coulter s LX-20; Brea, CA, USA). Fasting plasma insulin was determined using a microparticle-based enzyme immunoassay and an Abbott IMX analyzer (Abbott Laboratories, Dainabot, Japan). Glycosylated haemoglobin (HbA1c) was determined by an automated glycohaemoglobin analyzer A1C 2.2 plus (Tosoh Medics Inc.; Foster City, CA, YSA). Separation of haemoglobin fractions was achieved by injecting 3 ml of diluted haemolyzed sample on an HPLC column (instrument calibrated to DCCT values). FSP Intervention The 16-week FSP was delivered in a standard fashion 11,12 by PA experts (CT-L, NL). During the adoption phase (initial 4 weeks), participants were asked to attend four weekly group meetings (individual attendance was recorded). They were given pedometers and the program manual containing goalsetting and problem-solving exercises, as well as calendars for self-monitoring steps/day. No specific advice was given concerning diet or glycemic control. Other forms of PA were not systematically assessed as they tend to be relatively uncommon in this population 1 and quantification tends to be problematic. 22 During the adherence phase (subsequent 12 weeks), participants were asked to use their pedometers and calendars for goal-setting and self-monitoring. Postcards were mailed at 6 and at 10 weeks thanking people for participating in the FSP. Simultaneously, the CONTROL group received postcards thanking them for their participation in the study. Pedometers and calendars were returned at the 16-week assessment. Calendars were examined to verify pedometer use, the number of days/week personal goals were achieved (for each of the first 4 weeks), and to calculate the mean steps/day (up to 16 weeks in those with complete data, n ¼ 14). Data treatment and statistical analysis All data are presented as mean7s.d. unless otherwise stated; a P-value of 0.05 was considered statistically significant. Baseline differences between the FSP and CONTROL groups were examined using unpaired t-tests or w 2 analysis. A repeated measures ANOVA (with group as the between factor and time as the within factor) was used to examine steps/day, anthropometric variables, indicators of cardiovascular fitness, glycemic control and blood lipid parameters. Nonparametric statistics were used when variables were not normally distributed. A delta variable (D ¼ 16-weekFbaseline) was constructed for each parameter. Spearman correlation coefficients were used to examine whether change in PA (D steps/day) was related to various health improvements. For post hoc analyses of glycemic and plasma lipid profiles, subjects were grouped according to type of diabetes treatment (ie, diet alone vs oral hypoglycemic medications). Results A total of 60 individuals (55% male, 45% females, age ¼ y; BMI ¼ kg/m 2 ), qualified and agreed to participate in the study. Seven participants in the CONTROL group (three males, four females) and six in the FSP (four males, two females) did not return for the 16-week assessment. Baseline characteristics did not differ significantly between participants who did and did not return for the 16-week assessment. The findings reported below are based on the 47 participants (24 FSP and 23 CONTROL) assessed at both baseline and 16 weeks. 115

4 116 Baseline characteristics Demographic and baseline clinical characteristics were similar in FSP and CONTROL group participants (Tables 1 and 2). Generally, participants in this study were in their mid-50s and had BMIs that would be considered overweight to obese 23 (range of BMI ¼ kg/m 2 ). Average time since diagnosis of type II diabetes was months. Aside from diabetes, the most commonly reported health problems were high blood pressure (55% of total sample), high serum cholesterol (43%) and allergies (25%). In total, 17% of the participants smoked cigarettes regularly, 13% smoked occasionally, 38% were former smokers and 32% had never smoked. Of the sample, 23% had college or university graduate education, 30% had some postsecondary, 21% had a high school diploma and 25% had some secondary school (25%). One person was unemployed (0.02%), 13% were retired and the remaining individuals categorised themselves as nonoffice worker (43%), office worker (36%) or homemaker (6%). Program compliance Of the 24 participants in the FSP, 16 (67%) attended all four group sessions; the remainder attended three of the four sessions. Calendars were returned by 21 participants (88%) at the 16-week follow-up (the others were reportedly lost). During the adoption phase, FSP participants consistently achieved or exceeded their personal goals 3 4 days each week. Daily step goals generally increased each week, although the greatest increment in steps/day was achieved the first week (Figure 2). Compliance recording daily steps in calendars dropped from 100% at the beginning of the adoption phase to 88% at the beginning of the adherence phase (when group meetings were discontinued) and stabilised at 58% during the last 4 weeks of the adherence phase (while there was no on-going contact). No FSP participants reported having noteworthy musculoskeletal pain or problems during the intervention. Assessment at 16 weeks FSP participants increased their PA by approximately 3000 steps/day from baseline (Po0.01) (Table 3), a significant improvement relative to CONTROL group (Po0.0001). This change by FSP participants was not related to continued recording in calendar. Body weight and BMI remained constant in the two groups over the 16 weeks studied. Waist girth decreased by 1.8 cm in the FSP group compared to 0.4 cm in the CONTROL group Table 1 Baseline personal, physical activity and anthropometric characteristics Characteristics All participants (n ¼ 47) FSP (n ¼ 24) Control (n ¼ 23) Sex 26M/11F 12M/12F 14M/9F Age (y) Treatment n (%) Diet only 26 (55.3) 14 (58.3) 12 (52.2) Oral hypoglycaemics 21 (47.4) 10 (41.7) 11 (47.8) Physical activity (steps/day) Height (cm) Weight (kg) BMI (kg/m 2 ) Waist girth (cm) Hip girth (cm) % Subjects with waist hip ratio 41 19% 25% 13% Note: Continuous variables are mean7s.d. Table 2 Baseline indicators of cardiovascular fitness, glycemia and lipid status Indicator All participants (n ¼ 47) FSP (n ¼ 24) Control (n ¼ 23) Resting HR (bpm) Resting systolic (mmhg) Resting diastolic (mmhg) Fasting glucose (mm) h glucose (mm) Insulin (pmol/l) HbA1c (%) Total cholesterol (mmol/l) HDL (mmol/l) LDL (mmol/l) Triglycerides (mmol/l) Note: Continuous variables are mean7s.d.

5 Average steps/day Figure 2 (n ¼ 14) Baseline Weeks (Table 3). Although change in waist girth was significant over time (P ¼ 0.025), there was no significant group difference (P ¼ 0.128). Hip girth also decreased by 3.0 cm in the FSP participants, compared to 0.4 cm in the CONTROL group. This difference, however, was not statistically significant over time (P ¼ 0.09) nor between groups (P ¼ 0.190). None of the remaining indicators of cardiovascular fitness, glycemia or lipid status differed significantly over time or between groups (Tables 3 and 4). 7 8 Pattern of physical activity from complete FSP participant records For the total sample, baseline steps/day were inversely correlated with fasting blood glucose (r ¼ 0.45, P ¼ 0.001), glucose at 120 min postglucose load (r ¼ 0.35, P ¼ 0.02) and HbA1c (r ¼ 0.42, P ¼ 0.003); however, no relationship emerged with fasting insulin (r ¼ 0.12, P ¼ 0.42). For the FSP group, there was a significant relationship between D steps/day and D fasting blood glucose (r ¼ 0.43, P ¼ 0.04; CONTROL: r ¼ 0.11, P ¼ 0.59). This relationship was particularly strong among FSP participants treated with oral hypoglycemic medications (r ¼ 0.76, P ¼ 0.01, n ¼ 10) compared to diet alone (r ¼ 0.53, P ¼ 0.1, n ¼ 11). In the FSP participants treated with oral hypoglycemic medications, a significant inverse relationship was observed between D steps/day and D HbA1c (r ¼ 0.71, P ¼ 0.02), as well as with D circulating triglyceride concentrations (r ¼ 0.64, P ¼ 0.04). Again, these relationships were not observed in the CON- TROL group, nor among FSP participants treated with diet alone. Correlations between D steps/day and D indicators of cardiovascular fitness were not statistically significant. Of potential clinical significance, however, was an inverse relationship between D steps/day and D resting heart rate for FSP participants treated with diet alone (r ¼ 0.53, P ¼ 0.06); the subgroup with the largest D steps/day ( steps/day). Assessment at 24 weeks Of those assessed at 16 weeks, 16 (67%) FSP and 22 (96%) CONTROL group participants returned for the follow-up at 117 Table 3 Physical activity, anthropometric and cardiovascular status at week 16 and change from baseline (D) FSP group (n ¼ 24) Control group (n ¼ 23) Variable Week 16 D Week 16 D Physical activity (steps/day) a,b b Weight (kg) Waist girth (cm) Hip girth (cm) Resting HR (bpm) Resting systolic (mmhg) Resting diastolic (mmhg) Note: D ¼ (16-weekFbaseline).Values are means7s.d. a Difference between groups, *Po b Difference over time, Po Table 4 Glycemia and lipid status at week 16 and change from baseline (D) FSP group (n ¼ 24) Control group (n ¼ 23) Variable Week 16 D Week 16 D Fasting glucose (mm) h glucose (mm) Insulin (pmol/l) HbA1c (%) Total cholesterol (mmol/l) HDL (mmol/l) LDL (mmol/l) Triglycerides (mmol/l) Note: D ¼ (16-weekFbaseline). Values are means7s.d. No significant differences between groups or over time.

6 weeks. While steps/day for the FSP group were still higher on average compared to the CONTROL group ( vs ), the group difference was no longer significant (P ¼ 0.17). No significant longer-term changes emerged for any of the other health indicators. Discussion This study confirms our preliminary findings 11 that the FSP is effective in eliciting an immediate and profound increase in daily walking behaviour (approximately 3000 steps/day above baseline) during the intervention. Since individuals with type II diabetes take E2200 steps in 20 min of selfpaced walking (1), this average incremental change of 3369 steps/day for the FSP group,represents approximately 31 min of extra walking per day. These findings also confirm previous studies 24 and further demonstrates the utility of the pedometer approach with a population known to have little interest in, and high dropout rates from, formal exercise programs. 4,8 The group meetings were well attended, participants achieved or exceeded their step goals 3 4 days/week, and steps/day (both blinded protocol and calendars) support a true behaviour change, sustained above baseline levels for at least 12 weeks beyond formal interactions. Concerning the glycemia indicators, the most interesting finding was the inverse relationship between D steps/day with D blood glucose, HbA1c and triglyceride concentrations in FSP participants treated with oral hypoglycemic medications. Although we did not collect these data, there may be an interaction between improved insulin secretion (from the medication) and slightly improved insulin sensitivity (from increased activity). Moderate improvements in both factors may be more effective in people with type II diabetes than a dramatic change in either one alone, and warrants further examination. In contrast, the relationship between D steps/day and D blood glucose was not observed in FSP participants treated with diet alone. For this group with characteristically limited insulin secretion, a larger improvement in insulin sensitivity (than induced by this self-directed programme) may be necessary to affect blood glucose levels. The dose response relationship between improved insulin sensitivity and steps/ day has yet to be established, and would need to account for the role of any change in body weight. Yamanouchi et al 13 found that a substantial amount of walking (average steps/day) combined with diet over 6 8 weeks resulted in improved insulin sensitivity and a 7.8 kg weight loss. The current, emphasis on daily accumulation of 30 min or more of moderate intensity activity 7 may be insufficient to elicit improvements in glycemia in people who are typically insulin resistant and have reduced insulin secretion. This is an important consideration when counselling people with type II diabetes. Further research is needed to determine levels of PA that bring about important health benefits (including improved glycemic control) under realworld living conditions. Results from such studies will help patients and practitioners establish realistic goals and timeframes for achieving health benefits. Generally, the physiological effects of increased walking in this study were subtle. Participation in the FSP was associated with slight reductions in waist and hip girths (without weight change) and was not associated with changes in any of the indicators of cardiovascular fitness. Longer-term or more intense activity may be required to observe significant changes in these parameters. Given the small magnitude of changes observed in these outcomes, future studies will require at least 84 participants per group to detect a statistically significant 0.5 1% difference in HbA1c between groups. This was beyond the scope of the present study focused on the primary outcome of PA behaviour, but should be considered for future research. The present findings must be interpreted cautiously considering response rates and attrition documented at 16 weeks, and particularly at 24 weeks follow-up. Only 41% of the initial pool identified from medical charts could be reached by telephone and were eligible and interested in participating (60 of the 146). While lack of time was the primary reason cited by the 67 individuals who chose not to take part, it is unclear whether the issue was time to attend the four meetings, time to complete the assessments, or both. Those who participated were likely to be highly motivated and interested. We had no formal contact with participants after 16 weeks and this may have contributed to reduced response to the 24 week follow-up invitation. The present results, therefore, should be interpreted in light of the fact that programme recruitment and adherence reported reflects the balance between participant motivation and the extra demands (eg, assessments and monitoring beyond typical care) of a research study. In summary, notwithstanding the study limitations, this study showed that the FSP is an effective real-world programme for sedentary or insufficiently active, overweight and obese adults with type II diabetes. By the end of the programme, participants walked an additional half hour/day, constituting an important first step towards adopting a more active lifestyle. The physiological impact of this behavioural change was subtle and appears to be most obvious in persons taking oral hypoglycaemic agents. By 24 weeks, steps/day decreased without continued contact, suggesting that additional strategies (such as booster sessions ) are needed to re-engage FSP graduates in order to sustain PA behaviour change and elicit long-term health benefits. For those individuals who are successful in sustaining an elevated number of steps/day, the next step would be to encourage additional increases and/or attention to intensity and duration of exercise bouts. This study will help guide expectations and goals for initial adoption of a more active lifestyle in people with type II diabetes. Future research is important to: (1) determine realistic and responsive health outcomes for this population; and (2) continue to

7 evaluate and refine practical programs that can be delivered in real-world contexts. Based on these findings, we have incorporated booster sessions into the FSP and are currently evaluating strategies for national dissemination within diabetes education centres across Canada comparing delivery via diabetes educators vs peer leaders. Acknowledgements This project was supported by a grant from the Canadian Diabetes Association. Additional support for the development of First Step Program educational resources was provided by a Canadian Diabetes Association Award supported by Bayer Corporation. The results herein were previously presented as a poster at the Canadian Diabetes Association Annual Meeting in References 1 Ford ES, Herman WH. Leisure-time physical activity patterns in the US diabetic population. Findings from the 1990 National Health Interview SurveyFHealth Promotion and Disease Prevention Supplement. Diabetes Care 1995; 18: Hays LM, Clark DO. Correlates of physical activity in a sample of older adults with type 2 diabetes. Diabetes Care 1999; 22: Wei M, Gibbons LW, Kampert JB, Nichaman MZ, Blair SN. Low cardiorespiratory fitness and physical inactivity as predictors of mortality in men with type 2 diabetes. Ann Intern Med 2000; 132: Tudor-Locke C, Bell RC, Myers AM. Re-visiting the role of physical activity and exercise in the treatment of type 2 diabetes. Can J Appl Physiol 2000; 25: Manson JE, Hu FB, Rich-Edwards JW, Colditz GA, Stampfer MJ, Willett WC, Speizer FE, Hennekens CH. A prospective study of walking as compared with vigorous exercise in the prevention of coronary heart disease in women. N Engl J Med 1999; 341: Gregg EW, Beckles GL, Williamson DF, Leveille SG, Langlois JA, Engelgau MM, Narayan KM. Diabetes and physical disability among older US adults. Diabetes Care 2000; 23: US Department of Health Human Services. Physical activity and health: a report of the surgeon general. US Department of Health and Human Services, Centers for Disease Control and Prevention, National Center for Chronic Disease Prevention and Promotion: Atlanta, GA; Searle MS, Ready AE. Survey of exercise and dietary knowledge and behaviour in persons with type II diabetes. Can J Public Health 1991; 82: Tudor-Locke C, Myers AM, Rodger NW, Ecclestone NA. Towards acceptable exercise guidelines in type 2 diabetes: an examination of current standards and practices. Can J Diabetes Care 1998; 22: Tudor-Locke C, Myers AM, Rodger NW. Development of a theorybased daily activity intervention for individuals with type 2 diabetes. Diabetes Educ 2001; 27: Tudor-Locke C, Myers AM, Bell RC, Harris S, Rodger NW. Preliminary outcome evaluation of The First Step Program: a daily physical activity intervention for individuals with type 2 diabetes. Patient Educ Couns 2002; 47: Tudor-Locke C, Myers AM, Rodger NW. Formative evaluation of The First Step Program: a practical intervention to increase daily physical activity. Can J Diabetes Care 2000; 24: Yamanouchi K, Takashi T, Chikada K, Nishikawa T, Ito K, Shimizu S, Ozawa N, Suzuki Y, Maeno H, Kato K, Oshida Y, Sato Y. Daily walking combined with diet therapy is a useful means for obese NIDDM patients not only to reduce body weight but also to improve insulin sensitivity. Diabetes Care 1995; 18: Leenders NYJM, Sherman WM, Nagaraja HN, Kien CL. Evaluation of methods to assess physical activity in free-living conditions. Med Sci Sports Exerc 2001; 33: Bassett Jr DR, Ainsworth BE, Swartz AM, Stratch SJ, O Brien WL, King GA. Validity of four motion sensors in measuring moderate intensity physical activity. Med Sci Sports Exerc 2000; 32: S471 S Tudor-Locke C, Bell RC, Myers AM, Harris SB, Lauzon N, Rodger NW. Pedometer-determined ambulatory activity in individuals with type 2 diabetes. Diabetes Res Clin Pract 2002; 55: Tudor-Locke C, Ainsworth BE, Whitt MC, Thompson R, Addy CL, Jones DA. The relationship between pedometer-determined ambulatory activity and body composition variables. Int J Obes 2001; 25: Bassett DR, Ainsworth BE, Leggett SR, Mathien CA, Main JA, Hunter DC, Duncan GE. Accuracy of five electronic pedometers for measuring distance walked. Med Sci Sports Exerc 1996; 28: Swartz AM, Bassett Jr DR, Moore JB, Thompson DL, Strath SJ. Accuracy of an electronic pedometer in adults with varying body mass index levels. Med Sci Sports Exerc 2003; 35: S Tudor-Locke C, Myers AM. Methodological considerations for researchers and practitioners using pedometers to measure physical (ambulatory) activity. Res Q Exerc Sport 2001; 72: Canadian Society for Exercise Physiology. The Canadian physical activity, fitness and lifestyle appraisal. The Canadian Society for Exercise Physiology: Ottawa; Tudor-Locke C, Myers AM. Challenges and opportunities for measuring physical activity in sedentary adults. Sports Med 2001; 31: World Health Organization. Obesity: preventing and managing the global epidemicfreport of a WHO consultation on obesity. World Health Organization: Geneva; Tudor-Locke C. Taking steps toward increase physical activity: using pedometers to measure and motivate. Res Dig 2002; 3: