Vesco, Kimberly K. MD, MPH1; Dietz, Patricia M. DrPH, MPH2; Rizzo, Joanne MPA1; Stevens, Victor J. PhD1; Perrin, Nancy A. PhD1; Bachman, Donald J. MS1; Callaghan, William M. MD, MPH2; Bruce, F Carol BSN, MPH2; Hornbrook, Mark C. PhD1
The prevalence of obesity (body mass index [BMI] 30 kg/m2 or more) among reproductive-aged women in the United States tripled between 1960–1962 and 2003–2004, increasing from 9% to 29%.1,2 Among pregnant women, the prevalence of overweight and obesity (prepregnancy BMI of greater than 26) has almost doubled in the last 20 years, from 24% in 1983 to 45% in 2007, with almost one half of pregnant women in the United States now beginning their pregnancies as either overweight or obese.3 Previous studies have found that weight gain during pregnancy is often associated with postpartum weight retention and the subsequent development of long-term obesity and obesity-related illness.4,5 Relative to women whose weight changed by less than 10 pounds between pregnancies, women who gained at least 10 pounds had a 1.5-fold increased risk of gestational diabetes in their next pregnancy, and women who lost at least 10 pounds between pregnancies had a 40% decreased risk.6 Weight gain above the Institute of Medicine Recommendations for pregnancy results in greater retained postpartum weight.7 Unfortunately, heavier women who gain excessive weight are more likely to retain it postpartum compared with lighter women.8,9 We sought to estimate the incremental effect of gestational weight gain above the Institute of Medicine (IOM) recommendation for obese women (at least 15 pounds for term pregnancy)10 on postpartum weight retention at 1 year.
MATERIALS AND METHODS
We conducted a retrospective cohort study using data from electronic medical records in Kaiser Permanente Northwest. Kaiser Permanente Northwest is a large not-for-profit, prepaid, federally-certified, Joint Commission-accredited, group practice health maintenance organization (HMO) with 470,000 members in northwestern Oregon and southwestern Washington. We selected all singleton gestations resulting in a live birth that occurred among obese women that began and ended between January 1, 2000, and December 31, 2005. Obesity was defined as a BMI of 30 kg/m2 or greater. Other inclusion criteria were a documented maternal weight in each of the following time periods: 1) between 6 months before pregnancy and 12 weeks gestation (–180 to +90 days of pregnancy onset), 2) within 2 weeks of delivery, and 3) between 8 to 18 months (243 to 547 days) postpartum. For women with more than one pregnancy during the study period, only their first pregnancy was included in this study. However, first pregnancies were excluded if the subsequent pregnancy overlapped within 1 year of completion of the first pregnancy. Other exclusion criteria included multifetal gestation and maternal height not available.
To identify pregnancies, we used a validated computer algorithm that accessed multiple Kaiser Permanente Northwest automated data systems. Details of this algorithm and examples of its applicability have been published previously.11 Briefly, the algorithm searches hospital discharge abstracts; prenatal outcome data, diagnostic and procedure codes from same-day surgery records, ambulatory encounters, and emergency department and home health visits; outside professional and facility claims and referrals; imaging procedures; laboratory test results; and pharmacy dispensing records. The resulting database identifies pregnancy “episodes” and includes all health care encounters with their attendant ancillary diagnoses, procedures, and other health services. After pregnancies are identified, a probabilistic method is used to match birth certificates to the Kaiser Permanente Northwest data to gather additional demographic information. This study was approved by the institutional review boards of the Centers for Disease Control and Prevention and Kaiser Permanente Northwest.
Our primary outcome variable was maternal weight retention at 1 year postpartum, evaluated as a continuous and categorical (less than 0 pounds, 0–10 pounds, greater than 10 pounds) variable (“Postpartum Weight Change”) and calculated as Weight at 1 year postpartum (closest weight measured 243–537 days postdelivery) minus weight at the onset of pregnancy (–180 to +90 days). We chose a wide eligibility window for the mother’s weight at the start of pregnancy to increase our research sample. Weight at pregnancy onset was defined as the weight recorded closest to “zero” weeks of gestation. The date of pregnancy onset or “zero” weeks of gestation was calculated retrospectively using the documented estimated date of delivery. Ninety-nine percent of women had a measured weight within 90 days of pregnancy onset. Weight measured before pregnancy (within 6 months) was available for 434 women (26%), for the remainder of the cohort we used the earliest available first trimester weight.
Our primary independent variable was total gestational weight gain evaluated as a continuous and categorical variable (“Pregnancy Weight Change”) and calculated as Last predelivery weight (within 2 wks of delivery) minus weight at onset of pregnancy (–180 to +90 days). The gestational weight gain categories (less than 0 pounds, 0–15 pounds, greater than 15 to 25 pounds, greater than 25 to 35 pounds, and greater than 35 pounds) were selected to correspond with the 1990 Institute of Medicine guidelines, which state that obese women should gain at least 15 pounds during pregnancy.10
We obtained important patient characteristics that included age (years), weight at onset of pregnancy (pounds), parity (0 compared with 1 or more), tobacco use during pregnancy (any), Medicaid (yes/no), race (white/other race or ethnicity), depression diagnosis during pregnancy or within 6 months postpartum, pregestational or gestational diabetes, chronic hypertension, time from weight at pregnancy onset to weight at delivery (days), and time from delivery to follow-up weight (days).
We used frequency plots of the primary independent and dependent variables to confirm normal distributions and identify spurious outliers. Although the Kolmogorov-Smirnov test of normality for pregnancy weight gain (D=0.05, P<.01) and for postpartum weight change (D=0.06, P<.01) were significant, the histograms appeared normally distributed with a small amount of right skew. Because significance tests of normality are often over powered with large sample sizes, and the distributions appear normal, the variables were not transformed. We used χ2 tests and analysis of variance with Bonferroni post-hoc tests to compare important patient characteristics and outcomes among the pregnancy weight change categories. Simple linear regression was used to estimate weight change from pregnancy onset to 1 year postpartum as a function of total gestational weight gain. The linear regression model was then adjusted for potential confounders. Multivariable logistic regression was used to estimate the risk of weight retention at 1 year postpartum (dichotomized as 10 pounds or less and greater than 10 pounds above baseline weight) as a function of pregnancy weight change after adjustment for important covariates. All of the patient characteristics listed above, as well as gestational age at delivery, were considered possible confounders of the relationship between pregnancy and postpartum weight change and were therefore included in the multivariable model. A backward, manual, step-wise selection method was used to create the most parsimonious model. A P<.05 was considered significant and was required for a variable to remain in the model. Two sensitivity analyses for the linear and logistic regression models were performed; one included only the women whose weight was assessed before pregnancy (n=434), and the other included only women who delivered at term (37 weeks or greater, n=1,528), and the results did not differ. We used Statistical Analysis Software 8.2 (SAS Institute Inc, Cary, NC) for all statistical programming.
Of the 3,273 obese women identified during the study period, 1,656 met eligibility criteria (Fig. 1). Compared with those who were included, women whose pregnancies were excluded were younger (mean age 27.2 compared with 28.4 years, P<.001), were more likely to be covered by Medicaid (12% compared with 7%, P<.001) and to be tobacco users (21% compared with 17%, P=.01), and were less likely to be parous (53% compared with 61%, P<.001), to have a depression diagnosis during pregnancy or within 6 months postpartum (18% compared with 23%, P<.001), or to have chronic hypertension (11% compared with 13%, P=.05).
Pregnancy weight change among eligible women ranged from a weight loss of 33.2 pounds to a total weight gain of 98.0 pounds. Nearly three-fourths of the cohort gained more than 15 pounds. Five percent lost weight and 24% gained 0 to 15 pounds (Table 1). The five weight change groups differed on age, with those gaining greater than 35 pounds being younger than those in the weight loss and 0–15 pounds weight gain groups (Bonferroni post-hoc test, P<.05). The groups also differed on baseline weight and BMI, with the weight loss and 0–15 pounds weight gain groups having both a higher mean weight and a higher mean BMI at baseline than the other groups (Bonferroni post-hoc test, P<.05). There were significant differences in depression and pregestational and gestational diabetes, with the highest rates in the weight loss group (Table 1). The proportion of women who had previously given birth to one or more children decreased across the weight gain categories from 69.9% in the weight loss group to 49.3% among those gaining greater than 35 pounds (P<.001).
Mean (±standard deviation) weight change from pregnancy onset to 1 year postpartum increased across the pregnancy weight change categories, ranging from –2.3 pounds (±16.3) for women who lost weight during pregnancy to +13.7 pounds (±20.9) among women gaining greater than 35 pounds during pregnancy (Table 1). There was an incremental increase across the pregnancy weight gain categories in the proportion of women retaining greater than 10 pounds at 1 year postpartum (Fig. 2).
The bivariable linear regression model demonstrated a significant linear relationship between pregnancy-related weight change and postpartum weight change (y=–4.9+0.39x, R2=0.11, P<.001; Fig. 3). For each pound gained during pregnancy, there was a 0.4-pound increase above baseline weight at 1 year postpartum. Adjustment for potential confounders did not change the significance of the model (P<.001).
Unadjusted logistic regression models demonstrated that, compared with women who gained 0–15 pounds during pregnancy, women who gained greater than 15 pounds were at a significantly increased risk of weighing greater than 10 pounds above their baseline weight at 1 year postpartum (Table 2). Risks increased with each 10-pound increment in gestational weight gain above 15 pounds. In adjusted models (Table 2), the risk of a postpartum weight greater than 10 pounds over baseline weight was twofold higher for women who gained greater than 15 to 25 pounds (odds ratio [OR] 2.18, 95% confidence interval [CI] 1.54–3.10) compared with women who gained 0–15 pounds, fourfold higher for women gaining greater than 25 to 35 pounds (OR 3.91, 95% CI 2.75–5.56), and almost eightfold higher for women gaining greater than 35 pounds (OR 7.66, 95% CI 5.36–10.97). Being insured under Medicaid was a significant predictor of a postpartum weight greater than 10 pounds above baseline weight, and the risk of a postpartum weight greater than 10 pounds above baseline weight was positively associated with BMI and inversely associated with age.
Our study demonstrates that greater than 70% of obese women gained more than the 15 pounds recommended by the IOM for obese pregnant women, and 21% gained greater than 35 pounds, which would be considered excessive for women with normal prepregnancy BMI. Furthermore, there was an incremental increase in risk of postpartum weight retention of greater than 10 pounds ranging from twofold among women gaining greater than 15 to 25 pounds to eightfold among women gaining greater than 35 pounds. We also identified a significant positive linear relationship between change in weight during pregnancy and change in weight from pregnancy onset to 1 year postpartum. It is not surprising that higher pregravid BMI, increasing parity, and diabetes were associated with decreased weight gain, because these associations have been demonstrated previously.12 Women with higher pregravid BMI and higher parity have greater fat stores, and the pregnancy weight gain required for fat stores would be less for these women. In addition, the trend toward less weight gain in women with higher prepregnancy weights and those with diabetes may be related to increased clinician focus on limiting weight gain for those patients. For example, in our health system, women with type 1, type 2, and gestational diabetes are entered into a case management program that includes dietary advice, home glucose monitoring, and weekly capillary blood glucose review.
Strengths of this study include the use of measured BMI and weight data rather than self-reported height and weight to assess the effect of incremental increases in gestational weight gain on postpartum weight retention among obese women. Additional strengths include our use of a well-validated, electronic algorithm to identify all pregnancies (and their associated comorbidities) in our health care system, which yielded a large cohort of obese women, more than 50% of whom had weight assessments at 1 year postpartum. In addition, the large proportion (more than 70%) of women in this cohort gaining greater than 15 pounds during pregnancy allowed us to evaluate the incremental effect of weight gain above 15 pounds (greater than 15 to 25 pounds, greater than 25 to 35 pounds, and greater than 35 pounds) on postpartum weight retention.
This study does have a few limitations, including potential biases that may be associated with the presence or absence of electronic weight assessments before, during, and after pregnancy. For example, women with health insurance and women with comorbid illnesses may be more likely to seek routine medical care and therefore more likely to have recorded weights. Our analyses demonstrated that women who were excluded were more likely to have Medicaid and less likely to have depression and hypertension. Because having Medicaid insurance was a risk factor for postpartum weight retention, it is possible that excluding these women either had no effect on the results or that the strength of the association between gestational weight gain and postpartum weight retention was underestimated. Weight assessments for this cohort were obtained from multiple scales rather than one standardized scale; however, accuracy of weights recorded in our health plan records have been found to be quite close to those recorded in the research setting with a standardized scale.13 Additionally, we chose wide eligibility windows for weights at the start of pregnancy and at 1 year postpartum. Although this may be a limitation, 99% of women had a measured weight within 90 days of pregnancy onset, and, because women on average gain 1 to 5 pounds in the first trimester,14 one would expect that the weight gain 6 months before through the first 3 months of pregnancy would be relatively stable. Lastly, the population we evaluated, although well-representative of the surrounding population, was primarily white, limiting generalizability of the results to other racial/ethnic groups. Nevertheless, our findings are consistent with other postpartum weight retention studies, which have demonstrated that excessive weight gain during pregnancy is associated with postpartum weight retention and the subsequent development of overweight and obesity.9,15
In addition to long-term, obesity-related health risks with excessive weight gain during pregnancy, there are concerns about increased risk in subsequent pregnancies. A cohort study of 4,102 obese, nondiabetic women with two or more singleton live births between 1992 and 1998 revealed that, relative to women whose weight changed by less than 10 pounds, women who gained at least 10 pounds between pregnancies had a 1.5-fold increased risk of gestational diabetes in the subsequent pregnancy (relative risk 1.47, 95% CI 1.05–2.04).6 Women who lost at least 10 pounds between pregnancies had a decreased risk of gestational diabetes mellitus in the subsequent pregnancy (relative risk 0.63, 95% CI 0.38–1.02). Increased risk of future preterm delivery poses additional hazards among women with excessive gestational weight gain, because weight gain in excess of 0.65 kg/wk during pregnancy is associated with an increased risk of preterm delivery (less than 37 weeks of gestation).15 A cohort study of 45,245 women without gestational diabetes mellitus found that women who gained more than IOM recommendations were three times more likely (OR 3.05, 95% CI 2.19–4.26) to have an infant with macrosomia (in this study, birth weight more than 4,500 g) and 1.5 times more likely to have an infant with hypoglycemia (OR 1.38, 95% CI 1.01–1.89) or hyperbilirubinemia (OR 1.43, 95% CI 1.06–1.93) than women whose weight gain was in the recommended range.16
Although data are limited on the effects of minimal weight gain or weight loss during pregnancy on pregnancy outcomes in obese women, current evidence suggests that minimal weight gain may result in the most favorable pregnancy outcomes. A recent study of 298,648 singleton pregnancies in the Swedish Medical Birth Registry found that, in women with a prepregnancy BMI of 30 or more, a gestational weight gain of less than 13 pounds was associated with the lowest risk of adverse obstetric and neonatal outcomes.17 A population-based cohort study using data from the Missouri birth–death certificate registry evaluated outcomes of 120,251 pregnant obese women delivering full-term, liveborn, single infants and determined that increasing gestational weight gain was associated with an increased risk of gestational hypertension, cesarean delivery, and large for gestational age infants and a decreased risk of small for gestational age infants.18 Collectively, the minimal risk for all four outcomes corresponded to a gestational weight gain of 10–25 pounds for women with a prepregnancy BMI of 30–34.9 kg/m2, a gain of 0–9 pounds for women with BMI of 35–39.9, and a loss of 0–9 pounds for women with a BMI of 40 or greater.
Few trials have tested successful methods for helping women to avoid excessive weight gain during pregnancy or for losing weight in the postpartum period. In one study, a behavioral intervention included clinic visits where women were weighed and given printed information on appropriate weight gain, exercise, and healthy eating during pregnancy. A woman would receive positive feedback at her visit if she was within the guidelines or more nutritional information and behavioral counseling if she exceeded the IOM recommendations.19 In the second study, participants were mailed materials on healthy weight gain, nutrition, and exercise in pregnancy, and their weight gain was monitored by their providers using weight gain grids.20 Neither intervention was effective in preventing excessive weight gain among obese women.19,20 A third and more recent trial of 50 nondiabetic, white, obese women randomly assigned to either intensive dietary counseling (10 1-hour sessions) and energy intake restriction or to no intervention demonstrated significantly lower gestational weight gain (6.6 kg compared with 13.3 kg, P=.002) in the intervention group.21
Given the serious morbidities associated with obesity and excessive weight gain, there is an urgent need for the development of interventions that successfully help women avoid excessive weight gain during pregnancy. Furthermore, additional prospective studies are needed to determine the long- and short-term effect of weight maintenance or limited gestational weight gain for obese women and their offspring. Although the IOM recommended a gain of at least 15 pounds, it reports that many women who gain less than that do have good outcomes. The IOM pregnancy weight gain guidelines are currently undergoing re-evaluation, with updated guidelines anticipated this summer.22
1. Flegal KM, Carroll MD, Ogden CL, Johnson CL. Prevalence and trends in obesity among US adults, 1999–2000. JAMA 2002;288:1723–7.
2. Ogden CL, Carroll MD, Curtin LR, McDowell MA, Tabak CJ, Flegal KM. Prevalence of overweight and obesity in the United States, 1999–2004. JAMA 2006;295:1549–55.
4. Rooney BL, Schauberger CW. Excess pregnancy weight gain and long-term obesity: one decade later. Obstet Gynecol 2002;100:245–52.
5. Rooney BL, Schauberger CW, Mathiason MA. Impact of perinatal weight change on long-term obesity and obesity-related illnesses. Obstet Gynecol 2005;106:1349–56.
6. Glazer NL, Hendrickson AF, Schellenbaum GD, Mueller BA. Weight change and the risk of gestational diabetes in obese women. Epidemiology 2004;15:733–7.
7. Keppel KG, Taffel SM. Pregnancy-related weight gain and retention: implications of the 1990 Institute of Medicine guidelines. Am J Public Health 1993;83:1100–3.
8. Olson CM, Strawderman MS, Hinton PS, Pearson TA. Gestational weight gain and postpartum behaviors associated with weight change from early pregnancy to 1 y postpartum. Int J Obes Relat Metab Disord 2003;27:117–27.
9. Gunderson EP, Abrams B, Selvin S. Does the pattern of postpartum weight change differ according to pregravid body size? Int J Obes Relat Metab Disord 2001;25:853–62.
10. Institute of Medicine (U.S.) Subcommittee on Nutritional Status and Weight Gain During Pregnancy. Nutrition in Pregnancy. Washington (DC): National Academies Press; 1990.
11. Hornbrook M, Whitlock E, Berg C, Callaghan WM, Bachman D, Gold R, et al. Development of an algorithm to identify pregnancy episodes in an integrated health care delivery system. Health Serv Res 2007;42:908–27.
12. Institute of Medicine. Influence of pregnancy weight on maternal and child health. Washington (DC): National Academies Press; 2007.
13. Stevens VJ, Wagner EL, Rossner J, Craddick S, Greenlick MR. Validity and usefulness of medical chart weights in the long-term evaluation of weight loss programs. Addict Behav 1988;13:171–5.
14. Carmichael S, Abrams B, Selvin S. The pattern of maternal weight gain in women with good pregnancy outcomes. Am J Public Health 1997;87:1984–8.
15. Sarwer D, Allison K, Gibbons L, Markowitz J, Nelson D. Pregnancy and obesity: a review and agenda for future research. J Womens Health (Larchmt) 2006;15:720–33.
16. Hedderson M, Weiss N, Sacks D, Pettitt D, Selby J, Quesenberry C, et al. Pregnancy weight gain and risk of neonatal complications: macrosomia, hypoglycemia, and hyperbilirubinemia. Obstet Gynecol 2006;108:1153–61.
17. Cedergren MI. Optimal gestational weight gain for body mass index categories. Obstet Gynecol 2007;110:759–64.
18. Kiel DW, Dodson EA, Artal R, Boehmer TK, Leet TL. Gestational weight gain and pregnancy outcomes in obese women: how much is enough? Obstet Gynecol 2007;110:752–8.
19. Polley BA, Wing RR, Sims CJ. Randomized controlled trial to prevent excessive weight gain in pregnant women. Int J Obes Relat Metab Disord 2002;26:1494–502.
20. Olson CM, Strawderman MS, Reed RG. Efficacy of an intervention to prevent excessive gestational weight gain. Am J Obstet Gynecol 2004;191:530–6.
21. Wolff S, Legarth J, Vangsgaard K, Toubro S, Astrup A. A randomized trial of the effects of dietary counseling on gestational weight gain and glucose metabolism in obese pregnant women. Int J Obes (Lond) 2008;32:495–501.
22. Institute of Medicine. Food and Nutrition Board. Reexamination of IOM Pregnancy Weight Guidelines. July 2009. Available at: http://iom.edu/CMS/3788/48191.aspx
. Retrieved August 11, 2009.