Physical activity patterns six weeks after delivery in postpartum women with prior gestational diabetes mellitus

Physical activity patterns of Sri Lankan mothers with a history of gestational diabetes mellitus (GDM) at 6 weeks postpartum

GDM is considered a threat to maternal and child health. It is estimated that 16% of live births to women have some form of hyperglycemia in pregnancy and 84% of those cases are due to GDM.^[1] GDM mothers are reported to have a 7.43-fold relative risk of developing type 2 diabetes mellitus (T2DM) in future life, compared to non-GDM mothers.^[2] International Diabetes federation further reports that vast majority of cases of antenatal hyperglycemia were from middle- and low-income countries.^[1] Since GDM is common among Asian populations,^[3] it is vital to have an effective preventive program/strategy to prevent progression of GDM to overt diabetes.

Physical activity (PA) during postpartum period is known to enhance maternal insulin sensitivity,^[4] reduce the risk of T2DM^[5] without negatively affecting milk production or composition.^[6]

Furthermore, it is reported that regular exercise training significantly reduces the HbA1c^[7] which is considered the gold stand for monitoring glucose control.^[8] Walking, pelvic floor exercises, and stretching are recommended for women immediately after the delivery.^[5,6] Although the PA recommendation for postpartum mothers is for more than 150 min/week,^[5,9] maximum benefits are achieved by gradually increasing the intensity from moderate to vigorous activity.^[5,10] However, many obstacles such as lack of family and social support^[11] their exercise beliefs and disparities in access to health-care^[10,12] hinder them from engaging in PA. In addition, lack of time, tiredness, excessive child care demands, and lack of awareness of the type of PA that is beneficial to prevent T2DM are also reported as barriers.^[11]

The physiological changes due to pregnancy persist for 4 to 6 weeks postpartum and exercise routines should ideally be introduced gradually during this period.^[10] Since PA guidelines for women in Asia pacific region are sparse, a meticulous assessment of the baseline PA patterns is justifiable as an initial step of developing a feasible and an easily implementable PA program for these women. Hence, the objective of this study was to evaluate the PA patterns of Sri Lankan GDM mothers at 6 weeks postpartum.

Study design and setting

This descriptive cross-sectional community based component was a part of a quasi-experimental study conducted in selected Medical Officer of Health (MOH) divisions of three districts in Sri Lanka. Four MOH divisions were selected from each district by cluster randomization method.

Study participants

Postpartum mothers with a history of GDM in their index pregnancy, who delivered a single healthy baby, were invited to the study at 6 weeks postpartum. GDM had been diagnosed by a consultant obstetrician and gynecologist based on ADA guidelines^[13] at 24–28 weeks of gestation. The mothers with a history of GDM in their previous pregnancies and/or a history of any other chronic illnesses and employees in health sector were excluded from the study. Thirty to thirty-five mothers were recruited purposefully from each district to achieve a total sample size of hundred.

Data collection

Data collection was conducted in the office of the Public Health Midwife (PHM) of selected MOH divisions. Informed written consent was obtained from all mothers before they were recruited. Ethical clearance was obtained from the relevant Ethics Committee. Data on socio-demographic variables, anthropometry, glycemic control, and PA were obtained.

Assessment of socio-demographic variables

A pre-tested, self-administered questionnaire was used to collect socio-demographic data.

Anthropometric measurements

Weight, height, waist, and hip circumferences were recorded by the standard techniques.^[14] BMI (kg/m²) was calculated from weight and height measurements and a BMI of less than 25, between 25 and 30 and more than 30 kg/m² were considered as normal weight, overweight, and obese, respectively.^[14] Waist to hip (WHR) and waist to height ratios were calculated. The WHR cut off for abdominal obesity was considered as 0.85.^[14] A waist to height ratio of <5, 5, or >5 but <6 and 6 or >6 were considered as no risk, increased risk, and very high risk, respectively.^[15]

Assessment of HbA1c

Glycemic control was assessed by HbA1c at and a value of A1c below 5.7% was considered normal while 5.7–6.4% and ≥6.5% were considered as pre-diabetes and diabetes, respectively.^[16]

Assessment of PA

The International PA Questionnaire (IPAQ) short version was used to assess the levels and patterns of PA. This is a standardized instrument commonly used to measure PA of different populations and varying sociocultural settings. The reliability and validity of the IPAQ short version have been widely assessed. Some questions in the standard questionnaire were replaced with others to suit the Sri Lankan setting. It was translated to Sinhala and back translated to English to reduce interpretation errors. IPAQ was administered by a trained investigator as the mothers were in varying literacy levels. In addition, the mothers were instructed to list their activities in a separate log book for a period of 1 week before the interview and IPAQ data were matched with log book entries to minimize subjective errors. Both the frequency (the number of days per week) and the duration (the average time in minutes) of moderate and vigorous intensity activities and walking were assessed separately. Data were cleaned by removing the activities which lasted for <10 min. The cutoff limits for the PA categories were based on the current PA guidelines.^[17]

Data analysis

Statistical analysis

Data were analyzed using SPSS version 21 statistical software package. Descriptive statistics were used to describe socio-demographic, anthropometric measurements, and HbA1c values. Details of specific type of physical activities in the cohort, namely, the vigorous-intensity, moderate-intensity, and walking were presented as means, medians, and the interquartile ranges. Siting time was presented as the median and interquartile range. The levels of PA in the cohort determined by categorical scores were presented as frequencies. All data were truncated to normalize the distribution. The differences between means were compared by independent sample t-test and ANOVA. The significance was set at P < 0.05. Relationship between the HbA1c and the specific types of physical activities was determined by zero-order correlation. Partial correlation was run to determine the relationship between HbA1c and the types of physical activities while controlling the factors (the number of children and the number of family members) that showed a significant association with physical activities.

Description of the study sample

The mean age (±SD) of study participants was 33.6 ± 5.8 years. About half the sample of mothers were housewives (n = 53) and 76% have studied only up to the ordinary level. Monthly income of the majority (n = 75) was below 50,000 SLR (313 USD). Exactly half of the sample (n = 50) had five or more dependents in their families. The majority of the mothers (n = 75) had 1or 2 children. Fifty-five mothers had a positive family history of diabetes mellitus out of which in 25, both parents were diabetics.

Table 1 shows the anthropometric characteristics and HbA1c values of the subjects in the study sample categorized according to recommended cutoff values.

Description of PA

The median (IQR) energy expenditure of the study group (non-truncated) was 3659.50 MET-minutes/week (IQR = 123–12,000) and the truncated PA was 3250 MET-minutes/ week (IQR = 123–5106). Seventy-three percent of mothers participated in at least one moderate intensity activity continuously for more than 10 min in the 1 week period immediately before data collection and the common activities included were hand washing of clothes (67%) and sweeping the house and the garden (65%). However, when vigorous intensity activities are considered, only 10% participated in at least one activity continuously for more than 10 min during a period of 1 week and the common activities were drawing water from a well, chopping firewood, and digging the ground for gardening. Although walking was the most common activity in the group (76%), only 58% of mothers had walked for more than 150 min/week. In the categorical analysis, it was found that 52% of mothers had a high level of PA while 19% and 29% had moderate and low levels of PA, respectively. The median sitting time of the study group was 1260 min (IQR = 1260–3360)/week. Almost all mothers (98%) were engaged in breastfeeding or lulling the baby to sleep during their sitting time.

Table 2 shows the comparison of physical activity levels (MET-minutes per week) of subjects against socio-demographic characteristics, anthropometric measurements, and HbA1c values of the study population. Moderate intensity activity was significantly higher among mothers who have 5 or more than 5 family members and walking was significantly higher among the mothers who have 5 or more family members and mothers who have more than 2 children.

Table 3 shows the correlations between HbA1c values and physical activity levels before controlling and after controlling for the number of family members and the number of children. None of the physical activity levels correlated significantly with HbA1c at 6 weeks postpartum.