European Journal of Obstetrics & Gynecology and Reproductive Biology
Measuring physical activity in pregnancy: a comparison of accelerometry and self-completion questionnaires in overweight and obese women
Introduction
There is growing interest in the potential influence of physical activity (PA) on pregnancy outcome. In non-pregnant adults, PA has beneficial effects on glucose metabolism [1]. Evidence relating to pregnancy outcomes is mixed: some studies report that higher levels of PA may reduce the risks of gestational diabetes and pre-eclampsia [2], [3], [4], but others have found no beneficial effect [5].
Most studies assessing the effect of PA on pregnancy outcome have assessed PA by self report, often assessing only recreational or leisure time activity [6], [7]. Such measures rarely allow accurate assessment of light intensity activity and sedentary time, or of PA-related energy expenditure (PAEE), and may therefore be prone to misclassification. Some questionnaires have been specifically designed to overcome these limitations and to identify the duration of activity at different intensity, including sedentary time [8], [9].
The development of activity monitors such as accelerometers has facilitated objective measurement of the duration and intensity of body movement. Accelerometers demonstrate a high degree of reproducibility and validity for quantifying duration and intensity of PA, and correlate with energy expenditure in a variety of populations and settings [10].
There are specific challenges to measuring PA in pregnancy. It is a time of significant physiological change, and PA tends to decline as pregnancy progresses [11], [12]. The shape of the pregnant abdomen can alter the placement and tilt of any measurement devices, so affecting recording. Discomfort caused by the elastic belt and forgetting to re-attach the belt will also lead to inaccurate, incomplete data [13]. A significant proportion of PA in women with young children is derived from domestic chores and childcare, which are rarely explicitly measured by questionnaire [14]. Few PA measurement tools have been developed or evaluated for use in pregnancy. Chasan-Taber et al. developed a questionnaire specifically for use in pregnancy, but found only low to moderate correlation with accelerometry-derived estimates [15]. A small number of studies have used accelerometers in pregnancy [13], [16], [17]. In a study of 57 healthy primiparous women, Rousham et al. found that the correlation between self-reported (via seven day recall interview), and accelerometer-derived estimates of PA declined as pregnancy progressed [16], as did compliance with wearing the device. Harrison et al. compared accelerometry, pedometry and the International Physical Activity Questionnaire (IPAQ), in the second trimester of pregnancy [13]. They found that accelerometer and IPAQ estimates of PA did not correlate and there was poor absolute agreement.
Obese and overweight pregnant women are at higher risk of developing gestational diabetes and pre-eclampsia, and may particularly benefit from remaining physically active during pregnancy. The aim of this study was to assess the reproducibility, absolute and relative agreement of two self-completion questionnaires with objective measurement by accelerometry for the first time in overweight and obese pregnant women.
Section snippets
Study population
Pregnant women aged 16 or more, with a first trimester body mass index (BMI) ≥ 25 kg/m2 (based on measured weight and self-reported height), with normal first trimester ultrasound scan and singleton pregnancy were eligible for inclusion. Recruitment took place between October 2007 and January 2008 at the Royal Victoria Infirmary, a tertiary centre in Newcastle upon Tyne, UK, which had approximately 5000 deliveries per year during the recruitment period. Women were excluded if they were unable to
Study population (Table 1)
A total of 59 women had either valid accelerometry or questionnaire data, or both. Fifty-five women recorded at least three days of accelerometry at the first data collection point, 57 completed RPAQ questionnaires before and after accelerometry measurement and 56 completed AWAS questionnaires. Four women completed both questionnaires, but had no accelerometry data (one due to technical failure) and three completed at least three days of accelerometry but did not complete both questionnaires.
Comments
This study compared the measurement of PA using self-completion questionnaires with objective measurement of accelerometry in overweight and obese pregnant women. We found that both questionnaires, but particularly AWAS, over-estimated time spent in MVPA compared with accelerometry. There was no correlation between accelerometry or either questionnaire in measuring MVPA, and substantial disagreement in classification of those achieving at least 30 min of MVPA. There was low to moderate
Conclusions
The AWAS and RPAQ questionnaires show limited validity compared with accelerometry in pregnant women. Both questionnaires over-estimated MVPA activity and showed poor ability to discriminate duration of MVPA between women at individual or group level. Accelerometry measurement was feasible and acceptable. Objective methods should be used where possible in studies investigating PA and pregnancy outcome, unless and until questionnaires with demonstrably better validity have been developed.
Competing interests
None.
Authors’ contributions
All authors contributed to the design of the study, interpretation of the results, and critical revision of the manuscript. RB conceived the study and drafted the manuscript. CM recruited participants and collected the data. PWGT performed the statistical analysis.
Acknowledgements
We thank the study participants, and Brianna Fjeldsoe, Alison Marshall and Yvette Miller for use of the AWAS questionnaire. We thank Herve Besson for processing the RPAQ data to produce summary variables. Newcastle upon Tyne Hospitals NHS Foundation Trust acted as sponsors of the study and provided the salary for the research midwife but took no part in the collection, analysis and interpretation of data and in the writing of the manuscript.
References (33)
- et al.
Estimating physical activity energy expenditure, sedentary time, and physical activity intensity by self-report in adults
Am J Clin Nutr
(2010) - et al.
Reproducibility and relative validity of the short questionnaire to assess health-enhancing physical activity
J Clin Epidemiol
(2003) - et al.
Physical activity and diabetes prevention
J Appl Physiol
(2005) - et al.
Prospective study of gestational diabetes mellitus risk in relation to maternal recreational physical activity before and during pregnancy
Am J Epidemiol
(2004) - et al.
Associations of physical activity and inactivity before and during pregnancy with glucose tolerance
Obstet Gynecol
(2006) - et al.
A prospective analysis of recreational physical activity and preeclampsia risk
Med Sci Sports Exerc
(2008) - et al.
Does leisure time physical activity in early pregnancy protect against pre-eclampsia? Prospective cohort in Danish women
BJOG
(2009) - et al.
Physical activity during pregnancy and maternal-child health outcomes: a systematic literature review
Cadernos de Saude Publica
(2008) - et al.
Assessment of recreational physical activity during pregnancy in epidemiologic studies of birthweight and length of gestation: methodologic aspects
Women Health
(2007) - et al.
International physical activity questionnaire: 12-country reliability and validity
Med Sci Sports Exerc
(2003)
Validity and repeatability of the EPIC-Norfolk Physical Activity Questionnaire
Int J Epidemiol
Accelerometers and pedometers: methodology and clinical application
Curr Opin Clin Nut Metab Care
Activity patterns and time allocation during pregnancy: a longitudinal study of British women
Ann Hum Biol
Physical activity patterns during pregnancy in a diverse population of women
J Women's Health
Measuring physical activity during pregnancy
Int J Behav Nutr Phys Act
Physical activity in women with young children: how can we assess ‘anything that's not sitting’?
Women Health
Cited by (35)
Physical activity and pregnancy outcomes: an expert review
2023, American Journal of Obstetrics and Gynecology MFMCitation Excerpt :For example, most devices do not accurately capture swimming, cycling, and other low vertical impact activity, many of which are common activities undertaken during pregnancy, although technology is evolving.49 Bell et al49 compared accelerometer data with activity survey data in a cohort of pregnant individuals who were overweight and pregnant individuals with obesity at a median of 12 weeks of gestation. These researchers found a low correlation between the accelerometer (ActiGraph GT1M; ActiGraph LLC, Pensacola, FL) and survey data.
Clinical Practice Guideline No. 391 - Pregnancy and maternal obesity. Part 1: Pre-conception and prenatal care
2019, Journal of Obstetrics and Gynaecology CanadaGuideline No. 391-Pregnancy and Maternal Obesity Part 1: Pre-conception and Prenatal Care
2019, Journal of Obstetrics and Gynaecology CanadaCitation Excerpt :Special support may be needed for women who have back or joint pain or persistent nausea and vomiting of pregnancy.123 Devices such as pedometers or Fitbits have been shown to increase physical activity levels.125,126 In addition to the preceding recommendation, women should be advised to take 11 000 steps per day.127
Walking for health during pregnancy: A literature review and considerations for future research
2019, Journal of Sport and Health ScienceCitation Excerpt :Given similar correlations of the PPAQ and PIN3 with accelerometer and findings by Shephard,68 who described the difficulty with individuals self-reporting PA intensity, it does not seem that assessing PA intensity as a construct independent from activity type results in improved PA assessment. Other questionnaires developed for pregnant and nonpregnant individuals have shown similar agreement with device-based measures.14,77–82 Collectively, these findings indicate the agreement between questionnaires and activity monitors in pregnant women is modest at best.
Maternal physical activity prior to and during pregnancy does not moderate the relationship between maternal body mass index and infant macrosomia
2019, Journal of Science and Medicine in SportCitation Excerpt :First, maternal PA and BMI were measured via self-reporting methods. This type of assessment is notorious for producing less accurate estimates of behavior compared to objective assessments.25 Consequently, the use of these imprecise and error-prone tools likely led to biased, misestimated associations.26
Trajectories of objectively-measured physical activity and sedentary time over the course of pregnancy in women self-identified as inactive
2016, Preventive Medicine ReportsCitation Excerpt :A few studies have used both self-reported and objective measures to quantify physical activity levels during pregnancy (Ruifrok et al., 2014). Bell et al. measured physical activity using self-report and accelerometer measures in 59 pregnant women at one time point (12-week gestation) (Bell et al., 2013). Self-reported MVPA was significantly higher (81–127 min/day) than that recorded using accelerometers (35 min/day).