- Click here for explanation of classification scheme.

To explore the relationship between recreational physical activity performed during the year before and during the first 20 weeks of pregnancy and the risk of GDM.  Also, to examine the specific roles of walking and stair climbing in relation to risk of GDM.

Subjects were recruited as part of a case-control study designed to examine the epidemiology of preeclampsia.  Patients were identified who met the diagnostic criteria for GDM by the National Diabetes Data Group.  Women were classified as having GDM if two or more of the four plasma glucose values resulting from a diagnostic 100 g, 3h OGTT were abnormal according to the following criteria:  fasting > 105 mg/dl; 1-h > 190 mg/dl; 2-h > 165 mg/dl; 3-h > 145 mg/dl.  There were 191 eligible cases and 155 participants provided consent for participation. 

Controls were selected by identifying women who did not develop pregnancy induced HTN and who did not meet the NDDG criteria for GDM during the study pregnancy.

Not applicable

Recruitment Subjects were recruited as part of a case-control study designed to examine the epidemiology of preeclampsia.  Patients were identified who met the diagnostic criteria for GDM by the National Diabetes Data Group.  

Design Case Control

Blinding used (if applicable)  Neither the interviewers nor the participants were aware of any of the specific study hypotheses 

Intervention (if applicable) An in-person interview questionnaire was administered during a participant's postpartum hospital stay.  Information collected included maternal sociodemographic, medical, reproductive and lifestyle characteristics.  Women were also asked to report their height and weight during the 3 months prior to the study pregnancy. 

Women were asked to specify what recreational activities they engaged in during  the year before and during the first 20 weeks of pregnancy.  For each activity, information related to frequency and average time spent was assessed.  Women were asked about their usual walking pace (<2, 2, 3-4, >4 miles per h) and distance (miles) walked each day.  Women also reported  the number of flights of stairs climbed each day.

Statistical Analysis

1.  Logistic regression procedures were used to estimate the odds ratios of GDM in relation to varying levels of the various dimensions of physical activity examined (eg time, intensity, energy expenditure). 

 

2.  In univariate and multivariate analyses, tests for linear trend assessing increasing categories of physical activity among those physically active were conducted by treating the median values of each category as continuous variables. 

 

3.  To assess confounding, covariates were entered into the logistic regression model one at a time and the adjusted and unadjusted odd ratios were compared one at a time.

 

4.  Final logistic regression models included covariates that altered unadjusted odds ratios for physical activity variables by at least 10%, as well as those covariates of a priori interest (maternal age and parity) for which we attempted to frequency-match and those which we previously reported were associated with maternal physical activity (maternal education, socioeconomic status).

 

Timing of Measurements  Data was collected during participants postpartum hospital stay.

Dependent Variables

• Risk of Gestational Diabetes

Independent Variables

1.  Recreational activity performed the year before pregnancy, including:

• any recreational activity -- yes/no.
• time spent in activity (h/week) -- none, .1-1.9, 2.0-3.9, 4.0-5.9, >6.0
• energy expended performing activity (MET-h/week) -- none, .1-9.9, 10-19.9, 20-29.9, >30
• Maximum intensity (METs) -- not active, moderate, vigorous
• Distance walked (miles/d) -- <1.0, 1.1-2.0, 2.1-3.0, >3.0, missing
• Typical distance and pace of walking (daily) -- <2 miles/casual, <2 miles/brisk >2 miles/casual, > 2 miles/brisk, missing
• Flights of stairs climbed daily -- none, 1-4, 5-9, >10, missing

2.  Recreational activity performed during the first 20 weeks of pregnancy with the same variables as those listed for the year before pregnancy

3.  Walking

4.  Stair use

Control Variables pre-pregnancy BMI, maternal age, race, nulliparity, maternal education, socioeconomic status, smoking

 

Initial N: There were 191 eligible cases and 155 participants provided consent for participation. 155 GDM and 386 control participants

Attrition (final N):  N/A

Age:

	GDM N	GDM %	Control N	Control %
<19	1	.6	23	6.0
20-34	93	60.0	249	64.5
>35	61	39.4	114	29.5

Ethnicity:

	GDM N	GDM %	Control N	Control %
Non-Hispanic White	98	63.2	271	70.2
African-American	14	9.0	30	7.8
Other	41	26.5	83	21.5
Missing Data	2	1.3	2	.5

Other relevant demographics:

Household Income	GDM N	GDM %	Control N	Control %
<29999	37	23.9	68	17.6
30000-69999	64	41.3	126	32.6
>70000	47	30.3	172	44.6
Missing Data	7	4.5	20	5.2

 

High school education - 31 (20%) of GDM mothers had less than a high school education, compared to 76 (19.7%) of controls

Unmarried - 31 (20%) of GDM mothers were unmarried, compared to 87 (22.5%) of controls

Smoking - 29 (18.7%) of GDM mothers smoked, compared to 50 (13%) of controls

Nulliparous - 62 (40%) of GDM mothers had no previous children as compared to 203 (52.6%) of controls

Anthropometrics:

Prepregnancy BMI	GDM N	GDM %	Control N	Control %
<20	10	6.5	84	21.8
20 - 24.9	43	27.7	217	56.2
25 - 29.9	42	27.1	55	14.2
>30	60	38.7	30	7.8

Location: Labor and Delivery Wards at two Hospitals in Washington State, one in Seattle and one in Tacoma

 

Physical activity during the year before pregnancy

Women who participated in any recreational physical activity during the year before the index pregnancy expereinced a 51% reduction in risk of GDM when compared to women who were inactive.  This reduction in risk remained evident after controlling for confounding (OR = .45; 95% CI .28-.74)

Engaging in recreational physical activty for .1-1.9h per week was associated with a 51% reduction in risk (adjusted OR = .49; 95%CI .26-.92), while exercising 2.0-3.9h per week conferred a 47% reduction in risk (OR = .53; 95% CI .28-1.01).  With increased time spent in activity, the risk reduction appeared to flatten.

When compared with inactive women, women participating in vigorous physical activitiy (> 6 METs) experienced a 71% reduced risk of GDM (OR = .29; 95% CI .16-.51) and those participating in moderate physical activity (3-5.9 METs) experienced a 35% reduction in risk (OR =.65; 95% CI .38-1.10), although the latter did not approach statistical significance. 

Aprroximately 55% of both cases and controls reported that they walked over 1 mile per day.  Overall, there was no clear evidence of a linear trend in rsik of GDM with the number of of miles waked per day (P=.42).  There was also no clear pattern related to distance walked and pace of walking. 

Daily star climbing was associated with a decreased risk fo developing GDM.  For women who climbed 1-4, 5-9, and >10 flights of stairs daily, compared with those who did not, the adjusted odds ratios were .71 (95% CI .41-1.25), .48 (95% CI .23-1.02) and .47 (95% CI .26-.93), respectively.

Physical activity during first 20 weeks of pregnancy

Women engaged in recreational physical activity during the first 20 weeks of pregnancy experienced a 48% reducd risk of GDM, adjusting for confounding factors, when compared with those women who were inactive during this period (OR =.52; 95% CI .33-.8).

As with physical activity performed before pregnancy, greater time spent was associated with decreased risk, with a flattening of effect seen at about 4/week.  Energy expenditure was also associated with a decrease in risk of GDM, with a plateauing of effect seen beyond about 10 MET-h/week.

Active women participating in moderate recreational physical activity expereinced a 37% reduced risk fo GDM (OR =.63; 95% CI .39-1.02).  Women participating in vigorous intensity physical activity experienced an even greater reduction (66%) in risk of GDM as compared with physically inactive women (OR =.34; 95% CI .19-.63).

There was no clear evidence of a linear trend in risk of GDM with respect to the number of miles walked daily during this time period (P for linear trend =.518).  There was also no clear pattern related to distance and pace walked.

Stair climbing during pregnancy was inverserly related to risk of GDM after adjusting for potential confounding factors (P for linear trend .011).  When compared with women who did not climb stairs during this time period, those who climbed 1-4, 5-9, and >10 flights daily experienced reduction in risk of 49% (OR = .51;95% CI .3-.86), 78% (OR=.22;95% CI .10-.47) and 74% (OR = .26;95% CI .13-.52), respectively.

Women who climbed stairs, irrespective of whether they participated in no, moderate or vigorous recreational physical activities, experienced a reduced risk of GDM when compared with those who did not climb stairs.

Other Findings

To assess any differences by maternal pre-pregnancy weight, the analysis for risk of GDM were repeated within subgroups of normal weight, BMI <25, and overweight, BMI >25.  Women in either BMI group who participated in any recreational physical activity as compared to those who did not experienced a reduced risk of GDM of a magnitude similar to that seen in combined analyses.  Likewise, moderate and vigorous intensity levels of physical activity were associated with reduced risk of GDM in both normal and overweight women.  Also, women in both weight groups who climbed stairs daily experienced a reduction in risk of GDM when compared with those women who did not.

Women who participated in recreational physical activity during the year before and during the first 20 weeks of pregnancy experienced a 60% reduction in risk of GDM (OR = .40; 95% CI .23-.68). 

 

Regular participation in any recreational physical activity during the year before and/or during the first 20 weeks of the index pregnancy was associated with an approximate halving of risk for GDM in this case-control study. 

This risk reduction was further increased for women who were physically active during both of these time periods. 

In addition, the number of hours spent performing recreational activites and the energy expended were related to a decrease in risk of GDM. 

Finally the authors found that women who climbed stairs, irrespective of their participation in recreational physical activities and pre-pregnancy body mass index, also experienced a reduction in risk of GDM.

With respect to walking and pace and the questions that were asked of participants, ie do you walk <2, 2, 3-4 and >4 miles per h combined with pace (casual or strolling, average or normal, fairly brisk, brisk or striding), and how the subsequent categorization and reporting of walking as <2miles/casual, >2 miles/casual, < 2 miles brisk, >2miles/brisk) was derived is not clear.  It is also not clear why MET values were not assigned with walking pace instead given that it was used with all other activities.

There appears to be missing data with respect to background variables, but no information as to why this information is missing is provided.

The authors describe in detail several limitations of this study.  Some of these are due to self-report and retrospective data collection.  With respect to the later, they refer readers to their prospective study, Dempsey JC et al, Am J Epidemiol;159:663-670.  Some of these limitations are inherent to studies of physical activity and unavoidable.  Also in many instances, the bias worked against the authors such that the findings would likely have been stronger without these limitations.