Translation and cross-cultural adaptation of the Pregnancy Physical Activity Questionnaire (PPAQ) to the Brazilian population

Overview

Objective:
To evaluate the total activity performed by women with low-risk pregnancy as well as translate and pursue a cross-cultural adaptation of the Pregnancy Physical Activity Questionnaire (PPAQ) to the Brazilian reality.

Design:
We conducted a cross-sectional quantitative study including 305 women between 16 and 40 years of age with low-risk pregnancies.

Setting:
The Department of Public Health, State University of Ceará, Fortaleza, Brazil.

Methods:
We applied an adapted version of PPAQ to assess the levels of physical activity and the intensity in the metabolic equivalent task (MET), which could be distinguished as follows: sedentary (< 1.5 METs), light (1.5–3.0 METs), moderate activity (3.0–6.0 METs), and vigorous activity (>6.0 METs). The Kruskal–Wallis test was used to compare energy expenditure (MET) with socio-demographic variables.

Results:
More than half of the participants performed activities that were classified as light (51.4%). If we group the activities that were categorized as sedentary and light, this value increases to 74.7%, showing a high prevalence of insufficiently active pregnant women. Lower energy expenditure was observed in the third gestational trimester among pregnant women with lower educational level, single women, and mixed-race women (p < 0.05).

Conclusion:
There is a prevalence of physical inactivity during the three trimesters of pregnancy. The results validate PPAQ for the Brazilian population to serve as a basis for future public policies focused on combating the health problems of mother–infant pairs.

Keywords:
physical activity, pregnancy, energy expenditure, adaptation, questionnaire; Brazilian population

INTRODUCTION

Currently, there seems to be a growing global concern regarding the quality of life and harmful effects of physical inactivity. Approximately two million deaths worldwide occur due to a sedentary lifestyle, which represents a strong risk factor for developing coronary heart disease along with high blood pressure and increased cholesterol [16]. Regular exercises are associated with reduced adverse health conditions including premature death, cardiovascular disease, stroke, type 2 diabetes, and colon and breast cancer. Exercise also has a positive impact on the prevention and management of arterial hypertension and obesity, overall reduction of violence, and promotion of social integration, strengthened by strategies such as supporting healthy diets and discouraging the use of tobacco, alcohol, and drugs. Moreover, exercising promotes wellness by reducing stress, anxiety, and depression [8].

Aspects related to the benefits of physical activity during labor are associated with endocrine system adjustments during pregnancy, affecting the joints and pelvic ligaments and improving flexibility. High estrogen levels contribute to muscle relaxation, soften the cartilage, and allow for a wider range of movements by increasing synovial fluid in the joints. These mechanisms facilitate the passage of the fetus and childbirth [8].

Aerobic exercises are strongly recommended during pregnancy, including water workouts, walking, and cycling. In contrast, competitive sports, martial arts, weightlifting, and sudden movements or jumps can lead to joint injury and should therefore, be avoided [2]. Gunnar Borg’s rating of perceived exertion scale is a useful tool to determine the intensity of exercise. The suggested exercise for sedentary women is somewhat hard, whereas for those who exercise regularly, the difficulty varies from being hard to very hard. Elite athletes should not exceed 70% to 80% of their maximum heart rate [17].

In 2004, Chasan-Taber et al. [6] designed and validated a specific and reliable self-administered questionnaire, in English, to assess the level of physical activity during pregnancy termed the Pregnancy Physical Activity Questionnaire (PPAQ). The purpose of this questionnaire was to establish a global instrument to detect the level of physical activity among pregnant women.

The main goal of this study was to evaluate the total activity performed by women with low-risk pregnancy during the first, second, and third trimesters, as well as translate and pursue a cross-cultural adaptation of PPAQ to the Brazilian population, thus comparing the levels of physical activity with socio-demographic variables.

MATERIALS AND METHODS

We conducted a cross-sectional quantitative study to best describe the duration, frequency, and intensity of the total activity throughout pregnancy, along with socio-demographic characteristics. The quantitative design was primarily based on the use of a questionnaire containing closed-ended questions, which was adapted to suit the Brazilian population. The selected socio-demographic variables were also obtained to incorporate different women’s perception.

The study included 305 women who were between 16 and 40 years of age with low-risk pregnancies and who visited the basic health units in Fortaleza, Brazil, from November 2006 to December 2007. The pregnant women were divided in the following gestational trimester: 158, 103 and 44 at first, second and third trimester, respectively. The exclusion criteria were as follows: twin pregnancies, chronic diseases interfering with physical activity, chronic kidney disease, and a history of previous premature birth. The research ethics committee of the State University of Ceará approved the study under the protocol number 06193205-1, and all volunteering patients signed an informed consent. The criteria to classify the pregnancies as low-risk were as follows: absence of previous chronic diseases including arterial hypertension or diabetes, previous miscarriage, previous still birth pre-eclampsia, and/or pregnancy-induced hypertension, gestational diabetes, smoking during pregnancy, and alcohol consumption during pregnancy.

Besides gathering socio-demographic and clinical data, we used the Physical Activity Questionnaire for Pregnant Women (PAQPW), a translation and adaptation of PPAQ for the Brazilian population.

According to PAQPW, the type of exercise, intensity, duration, and frequency should be recorded in minutes or hours per day during the current gestational trimester. According to the intensity, the activity pattern according to the metabolic equivalent task (MET) can be distinguished as follows: sedentary (<1.5 METs), light (1.5–3.0 METs), moderate activity (3.0–6.0 METs), and vigorous activity (>6.0 METs) [6]. MET is equivalent to the relative oxygen consumption at rest; thus, 1 MET equals to 3.5 mL/kg/min. To determine MET scores, the rate of relative oxygen consumption at rest should be divided by 3.5, which will reflect how many times higher physical activity is compared to the resting state [6]. PPAQ aims to report the time period that the pregnant women spent developing their daily routine activities, including the following 32 modalities: household/caregiving (13 activities), occupational (5 activities), sports/exercise (8 activities), transportation (3 activities), and inactivity (3 activities). Furthermore, the respondents select the category that best indicates the amount of time spent in each activity: none; <0.5 h/day; 0.5 to almost 1 h/day; 1 to almost 2 h/day; 2 to almost 3 h/day, and ≥3 h/day. This brief particular questionnaire was selected because it could be self-administered, was easily understandable, and a substantial proportion of the total physical activity of individuals in developing countries are limited to those related to their occupation and transportation.

The time spent on each activity within the last 3 months, as well as the type and frequency of physical activities, were evaluated. The original version of PPAQ can be self-administered; however, we conducted an assisted interview if there was a high incidence of individuals with low health literacy in Brazil [9].

Cross-cultural adaptation of PPAQ

PPAQ was originally written in English and comprises 36 questions. Our final adapted version, i.e., PAQPW, comprised 33 questions [18]. The first and the second questions contain information about the woman’s last menstrual period and the expected date of delivery. The remaining 31 questions were aimed to compute the average energy expenditure attributable to each daily routine activity, which is usually performed by Brazilian pregnant women.

Translation into Portuguese

We adopted the World Health Organization methodology [15] for translating and adapting the instrument. During the cross-cultural adaptation process, two Brazilian translators, aware of the goals of this scientific research and fluent in English, prepared two adapted versions of PPAQ (forward translation). An independent translator, whose mother tongue is English and who had no knowledge of the questionnaire, translated the instrument back to English (back-translation). The conceptual and cultural equivalence were emphasized instead of linguistics, using the assistance of a bilingual expert panel. The researchers and the editor-in-chief compared both versions, and changes were made when there were disagreements. The final version was then compared to the original text in English and showed no divergence.

Adequacy of PPAQ

With the purpose of adjusting the original questionnaire to the Brazilian reality, the adapted and translated version (PAQPW) was submitted to six health professionals with recognized scientific expertise, who judged the statements and made valuable suggestions and contributions concerning the following aspects: context (appropriateness of the assertions), clarity (easy comprehension), and relevance (whether it really measures what it claims to be measuring).

Therefore, these professionals were requested to evaluate each assertion using a Likert scale, with scores ranging from 1 to 5 (1 – totally irrelevant, 2 – irrelevant, 3 – neutral, 4 – relevant, 5 – extremely relevant). The responses were tabulated in Excel 2003 (Microsoft Corp., Redmond, WA, USA) and then analyzed by the main author (FTS).

The questionnaire was applied by only one examiner (FTS) who had a bachelor’s degree in physical education and 3 years’eexperience in women’s health.

Computing energy expenditure

The data were collected and organized using PAQPW. Thus, we obtained answers with respect to the time spent on each activity throughout the last gestational trimester. We multiplied the time spent in each activity per day (h) by its intensity (MET) to estimate the average weekly energy expenditure (MET-h · wk^§-1). Furthermore, each activity was classified in terms of intensity based on the following criteria: sedentary (<1.5 METs), light (1.5–3.0 METs), moderate (3.0–6.0 METs), or vigorous (>6.0 METs). The pregnant women indicated the category that best estimated the amount of time spent in each activity.

The specific MET values revealed the need to perform the activities described in each of the 31 questions that followed the pattern described [question number: demanded MET value]: [3: 2.5], [4: 2.0], [5: 3.0], [6: 2.7], [7: 4.0], [8: 3,0], [9: 4.0], [10: 1.8], [11: 1.0], [12: 3.2], [13: 2.5], [14: 2.3], [15: 3.0], [16: 4.5], [17: 2.5], [18: 4.0], [19: 1.5], [20: 3.5], [21: 5.0], [22: 6.0], [23: 7.0], [24: 3.5] [25: 6.0], [26: 4.5], [27: 3.0], [28: 3.5], [29: 1.8], [30: 3.0], [31: 2.0], [32: 4.0], and [33: 3.3]. Table 1 contains detailed information of each question of PAQPW. The MET intensity of each activity was based on the compendium of physical activities [10].

To compute the average daily energy expenditure, we multiplied the demanded MET value in each question by the pre-determined duration category published by Chasan–Taber et al. [6],as per the instructions for using PPAQ (Table 2). The final value was multiplied by 7 (days/week), except for questions 15, 16, and 20–28 (values already in weekly form). After calculating the final score for each question, the values were added up, and subsequently divided by the total number of questions and then by 7 days to determine the mean value of daily energy expenditure in METs. Therefore, we could classify the intensity level of daily physical activity performed by the women interviewed.

Statistical Analysis

The data were collected using the software Epi Info version 6.04b (Centers for Disease Control and Prevention, Atlanta, GA, USA). Subsequently, the data was transferred to Excel 2003 (Microsoft Corp., Redmond, WA, USA) and processed using SPSS version 16.0 (SPSS Inc., Chicago, IL, USA). A descriptive assessment of the main variables and a comparative analysis was performed to look for any associations between the independent and dependent variables. Moreover, the frequency of the gestational periods, percentage, mean, and median values in the participants was calculated. We used the Kruskal–Wallis test once for nominal and measurement variables. It was classified into two competing hypotheses as follows: the null hypothesis (H0), in which the means are equal and the alternative hypothesis (Ha), in which there is at least a different pair. The significance level (p value) was set at 5% and we selected a chi-square variable with φ = n − 1 degrees of freedom.

RESULTS

Cross-cultural adaptation

To understand the issues that need to be addressed in relation to the context, clarity, and relevance, we developed a graphic score of all survey questions (PPAQ), scored on a scale from 0 to 5 by six recognized scientific professionals all with postdoctoral degrees in the health sciences. We adopted a cutoff point above the average scores of 2.5 (neutral, relevant, and extremely relevant). Figure 1 shows the median values of the questions of PAQPW after the modifications in the original issues from PPAQ, which were adapted by the Brazilian population.

Level of physical activity during pregnancy

The pre-pregnancy body mass index (BMI) was obtained in 221/305 pregnant women and were classified as follows: low weight 26.7% (59/221), eutrophic 55.7% (123/221), being overweight 10.9% (24/221), and obesity 6.9% (15/221). A high level of physical activity was obtained in 241/305 of the pregnant women and was classified as follows: 1.40–1.52 meters (29% = 70/241), 1.53–1.56 meters (21.2% = 51/241), 1.57–1.60 meters (25.7% = 62/241), and 1.61–1.82 meters (24.1% = 52/241).

The intensity level of the exercises performed by the 305 respondent pregnant women was as follows: 23.3% sedentary, 51.4% light, 24.3% moderate, and 1.0% vigorous.

Table 3 shows the physical activity intensity level based on the average energy expenditure according to the trimester of pregnancy. We observed a higher proportion of sedentary women and women performing light activity during the third trimester. Moderate activity was predominant during the first trimester of pregnancy.

Table 4 summarizes the intensity of physical activity according to the type of activity performed. Unlike other types of activities, most pregnant women practice moderate exercise while executing household tasks (54.4%). In contrast, most of them were classified as being sedentary for caregiving activities (42.4%), occupational activities (93.4%), and transportation (50.8%). Concerning leisure activities, in 80.7% of the interviewed women, the intensity level was classified as light. None of the respondents practiced sports/exercise activities.

Table 5 shows the energy expenditure (MET) according to socio-demographic variables. Lower energy expenditure was observed in the third gestational trimester (p = 0.007) among pregnant women with lower educational level (p = 0.025), for single women (p = 0.001), and among mixed-race women (p = 0.001).

DISCUSSION

Energy expenditure through physical activity is a well-known protective factor that improves overall health. Therefore, assessing the levels of physical activity in a specific population allows health care services to propose effective public health policies. PAQPW is a questionnaire designed to detect the level of different physical activities in the daily lives of pregnant women.

In our study, the prevalence of exercise classified as light (insufficiently active) was 51.4%, which is consistent with the study of Matsudo et al. [10] that identified 46.5% of insufficiently active subjects among 2001 Brazilian individuals in the state of São Paulo. In a validation study conducted by Chasan-Taber et al. [6], the preferred type of exercise among pregnant women was watching television/videos and walking slowly, thus classified as light to moderate.

More than half of the whole sample (51.4%) performed activities that were classified as light. If we group activities categorized as sedentary and light, this value increases to 74.7%, showing a high prevalence of insufficiently active pregnant women. Comparing the energy expenditure during the three trimesters of pregnancy, a gradual reduction over the course of gestation becomes evident. Considering the first, second, and third trimesters, 24.7%, 16.5%, and 34.1% of the women were classified as being sedentary, respectively. Regarding light activities, these numbers were 44.9%, 57.3%, and 61.4%, respectively. Observing the ratings related to moderate and vigorous activities, a striking reduction was found, which appears to be associated with a cultural component.

Considering the different types of physical activities included in this questionnaire, most of them were predominantly classified as sedentary, with the exception of leisure activities and household tasks, primarily classified as light and moderate, respectively.

PPAQ has already been transcribed and validated in other countries such as Vietnam and Japan [11, 14], confirming the importance of its adaptation to the reality of each country. In the Vietnamese study, the validity and reproducibility of PPAQ was tested by Bland–Altman plots and intraclass correlation coefficient (ICC). In the reproducibility of the analysis, the ICC values for “sedentary,” “light,” “moderate,” and “vigorous” were 0.94, 0.88, 0.90, and 0.87, respectively [14]. PPAQ that was transcribed and adapted to the Japanese population comprised 36 items, unlike our study that comprised 33 items. The Japanese study included the item “cycling for reasons other than for recreation or exercise”, because this activity is very common among pregnant women in Japan [11]. Moreover, the age of Japanese mothers in the study ranged from 25 to 38 years of age [11], unlike our study, where 16 to 40 year-old women were included, showing a higher incidence of teenage pregnancy among Brazilian women.

PPAQ was recently transcribed to assess its validity and accuracy for French obese pregnant women. Forty-nine obese women were evaluated and PPAQ was filled twice within a seven-day interval to recall physical activity in the last 3 months. ICCs between the two assessments were 0.90 for “total activity,” 0.86 for “light” and “moderate,” and 0.81 for “vigorous” [5]. In another recent study, Cohen et al. [7] evaluated the use of PPAQ to identify behaviors associated with appropriate weight gain during pregnancy. Sixty-one Canadian pregnant women in the second and third trimesters were evaluated. PPAQ comprised 32 questions, and although 15 of them were associated with increased physical activity in pregnant women, only the walking and cycling were associated with a healthy weight gain.

The results of our study have shown that a great number of our patients (23.3%) were classified as being sedentary. Bungum et al. [4], assessing nulliparous women, observed that sedentary pregnant women had 4.5 times the risk of births by cesarean section than physically active pregnant women, after controlling for factors such as age, type of anesthesia, changes in BMI prior to pregnancy, induced labor, and the type of hospital birth. Melzer et al. [12] suggest that the adoption or continuation of a sedentary lifestyle during pregnancy contributes to the development of certain disorders such as hypertension, maternal and childhood obesity, gestational diabetes, dyspnea, and pre-eclampsia. In a recent systematic review, Nascimento et al. [13] observed that exercises during pregnancy are associated with higher cardiorespiratory fitness, prevention of urinary incontinence and low back pain, reduced symptoms of depression, gestational weight gain control, and for cases of gestational diabetes, and a reduced number of women who required insulin. Furthermore, there was no association with reduction in birth weight or preterm birth rate.

The limitations of this study are that the questionnaire was applied to a low socio-economic population of Northwest Brazil with low levels of education, which could compromise the correct understanding of the issues and consequently the validity of the study. The assisted interview method is not objective because women may respond purposefully misrepresented; however, we believe that the assisted interview is a unique form to surmount the low cultural and education level of our population. Furthermore, PAPQW was established in agreement with the social-cultural aspects of the Brazilian population to minimize this weakness.

In summary, there is a prevalence of physical inactivity during the three trimesters of pregnancy. A majority of the women had a physical activity level that was classified as sedentary and light, and the moderate level was mainly reflected in the household tasks category. Regarding general social-demographic aspects, the sample indicated a majority of women who were mixed-race, married, with low levels of education, and low-wage jobs. These results validate PPAQ for the Brazilian population as a useful tool for future public policies focused on combating the health problems of mother–infant pairs.

Corresponding author

Prof. Edward Araujo Júnior, PhD

Department of Obstetrics, Paulista School of Medicine

Federal University of São Paulo (EPM-UNIFESP)

Rua Belchior de Azevedo, 156, apto. 111 Torre Vitoria

São Paulo – SP, Brazil

CEP 05039-030

e-mail: araujojred@terra.com.br