Measuring physical activity during pregnancy

Harrison et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:19 http://www.ijbnpa.org/content/8/1/19 RESEARCH Open Access Measuring physical activity during pregnancy Cheryce L Harrison1, Russell G Thompson3, Helena J Teede1,2, Catherine B Lombard1* Abstract Background: Currently, little is known about physical activity patterns in pregnancy with prior estimates predominantly based on subjective assessment measures that are prone to error. Given the increasing obesity rates and the importance of physical activity in pregnancy, we evaluated the relationship and agreement between subjective and objective physical activity assessment tools to inform researchers and clinicians on optimal assessment of physical activity in pregnancy. Methods: 48 pregnant women between 26-28 weeks gestation were recruited. The Yamax pedometer and Actigraph accelerometer were worn for 5-7 days under free living conditions and thereafter the International Physical Activity Questionnaire (IPAQ) was completed. IPAQ and pedometer estimates of activity were compared to the more robust and accurate accelerometer data. Results: Of 48 women recruited, 30 women completed the study (mean age: 33.6 ± 4.7 years; mean BMI: 31.2 ± 5.1 kg/m2) and 18 were excluded (failure to wear [n = 8] and incomplete data [n = 10]). The accelerometer and pedometer correlated significantly on estimation of daily steps (r = 0.69, p < 0.01) and had good absolute agreement with low systematic error (mean difference: 505 ± 1498 steps/day). Accelerometer and IPAQ estimates of total, light and moderate Metabolic Equivalent minutes/day (MET min-1 day-1) were not significantly correlated and there was poor absolute agreement. Relative to the accelerometer, the IPAQ under predicted daily total METs (105.76 ± 259.13 min-1 day-1) and light METs (255.55 ± 128.41 min-1 day-1) and over predicted moderate METs (-112.25 ± 166.41 min-1 day-1). Conclusion: Compared with the accelerometer, the pedometer appears to provide a reliable estimate of physical activity in pregnancy, whereas the subjective IPAQ measure performed less accurately in this setting. Future research measuring activity in pregnancy should optimally encompass objective measures of physical activity. Trial Registration: Australian New Zealand Clinical Trial Registry Number: ACTRN12608000233325. Registered 7/5/ 2008. Background Regular physical activity plays a fundamental role in health and is positively associated with a reduced risk of obesity, hypertension, diabetes and cardiovascular disease (CVD) and a variety of other health conditions. Additional advantages of regular physical activity apply to specific settings, such as pregnancy, with benefits including improved emotional well being and body image and reduced risk of gestational diabetes mellitus [1-3], excess maternal weight gain [4] and complications during labour [5]. Exercise during pregnancy also has * Correspondence: 1 Jean Hailes Foundation for Women’s Health Research Unit, School of Public Health and Preventive Medicine, Monash University, Clayton, Victoria, Australia Full list of author information is available at the end of the article positive effects on the foetus including improved stress response in utero and reduced risk of childhood obesity [6]. In the presence of limited pregnancy specific research, current activity recommendations are informed by general advice for healthy adults [7]. In the absence of any contraindications, pregnant women are advised to engage in physical activity of moderate intensity and duration (~30 minutes) on most days of the week [7]. Despite the potential health benefits, there is limited accurate information about physical activity patterns during pregnancy [8]. Previous epidemiological research suggests that most women (~50-60%) do not participate in regular physical activity during pregnancy [9,10]. However, these estimates are based around use of crude measures that are not validated [10] and may be prone to error [11]. A recent review highlighted the © 2011 Harrison et al; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/2.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Harrison et al. International Journal of Behavioral Nutrition and Physical Activity 2011, 8:19 http://www.ijbnpa.org/content/8/1/19 [22,23]. Although pedometers have been applied in pregnancy, there are theoretical concerns about the body shape changes in pregnancy. Therefore the role of pedometers in pregnancy remains unclear [8]. With accuracy of most popular subjective and objective tools unknown in pregnancy, there is a need to evaluate these against comprehensive techniques, including accelerometers, to inform on optimal evaluation of physical activity during pregnancy. In this setting, we aimed to evaluate the relationship and agreement between the accelerometer and subjective (IPAQ) and objective (pedometer) physical activity measurement tools in pregnancy. As the second trimester is considered the most comfortable period during pregnancy [12], we studied women between 26-28 weeks gestation consistent with previous research [8,12,15]. Methods Subjects Enrollment This study involved a subset of women recruited for a larger randomised controlled trial (RCT) promoting healthy living (intervention to improve diet and physical activity) in pregnancy versus standard information only (controls) from early pregnancy (13-16 weeks gestation) to six weeks postpartum. A flow diagram of participants, according to the CONSORT statement, are available in Figure 1[24]. Women were recruited for the larger RCT study from May 2008 onwards through invitation flyers given at the first maternal appointment in early pregnancy at a large tertiary hospital in Australia. Inclusion criteria for the larger RCT study included women who were overweight (Body Mass Index; BMI ≥ 25.0 kg/m2) Recruited for larger RCT study (n 185) Participants progressively invited for sub-study (n 59) Declined (n 11) Accepted (n 48) Did not complete (n 18) Analysis inadequacy of previous research in measuring physical activity during pregnancy, with studies predominantly using subjective long-term, self-recall of activity with no published evidence of their reliability or validity [11]. Further, there are limited pregnancy specific studies that have objectively assessed physical activity with pedometers [8,12-14] or accelerometers [15,16]. Given the limited research using objective, comprehensive and validated methods, there is currently no commonly accepted measurement tool used to assess physical activity during pregnancy. In non-pregnant populations, the gold standard measurement of free-living energy expenditure is the use of doubly labelled water (DLW) [17] which can also be used to estima (...truncated)