Evaluating the Impact of CenteringPregnancy Program Versus Individual Prenatal Care on Gestational Weight Gain

Abstract

Introduction:

The CenteringPregnancy (CP) program—proven to reduce preterm births—was modified to achieve more optimal gestational weight gain (GWG) by an intentional incorporation of nutrition education. We compared the effect of the modified CP program versus individual prenatal care (IPNC) on GWG.

Methods:

This observational study used linked birth certificate data and hospital discharge records of women who received prenatal care (PNC) in South Carolina Midlands' obstetric clinics between 2015 and 2019. Linear and multinomial logistic regressions were used to compare participants in CP (n = 568) versus IPNC on weight gain, measured by total GWG (delivery weight minus prepregnancy weight), weekly rate of weight gain, and meeting the Institute of Medicine's recommendations (inadequate, adequate, and excessive GWG). Nonrandom assignment to program was controlled by propensity scoring.

Results:

CP participants differed from IPNC participants in race, nulliparous, education, and type of health insurance, but not in parity or month PNC began (p-Value <0.05). CP and IPNC participants had a similar GWG experience: total GWG (coef(β) = −0.054; 95% confidence interval [CI] −0.78 to 0.6), total weekly weight gain (coef(β) = −0.004; 95% CI −0.03 to 0.03), total GWG category (inadequate GWG: RRR = 0.85, 95% CI 0.64–1.21, and excessive GWG: relative risk ratio (RRR) = 0.92, 95% CI 0.71–1.20 vs. adequate), and weekly weight gain category (inadequate GWG: RRR = 0.73, 95% CI 0.53–1.01, and excessive GWG: RRR = 0.83, 95% CI 0.61–1.13 vs. adequate).

Conclusion:

The CP program with an enhanced nutritional knowledge component was not associated with achieving recommended GWG. Further investigation is needed to explain the lack of impact.

Introduction

South Carolina maternal mortality rate is currently eighth highest among all states in the United States, at 26.2 maternal deaths per 100,000 live births.^1,2 The leading causes of maternal deaths include hemorrhage and cardiovascular and coronary conditions, which are associated with tissue injury, obesity, and excessive weight gain in pregnancy.^1,2 Gestational weight gain (GWG) is of particular interest because more than half (51.2%) of South Carolinian mothers had excessive GWG during pregnancy, while 24.4% had weight gain below the recommended guidelines.³ Maternal health and nutritional status, both before and during pregnancy, are important to appropriate fetal development and having a profound effect on birth weight, which is an important indicator of infant's well-being. Gain within recommended levels or range according to Institute of Medicine (IOM) 2009 guidelines has favorable pregnancy outcomes.^4
–6

Excessive and inadequate GWG in pregnancy are associated with increased risk of serious clinical conditions in the mother. For example, excessive GWG is associated with an increased risk of gestational diabetes, hypertensive disorders in pregnancy, complications during labor and delivery such as vaginal tears, excessive bleeding, nonelective cesarean section delivery, postdelivery weight retention, and subsequent maternal obesity.⁷ Weight gain below the recommended level has more detrimental effect on the unborn infant such as increased risk for delivery of small for gestational age infants, intrauterine growth restriction, low birth weight, poor fetal development, prematurity, and being prone to birth asphyxia leading to neonatal intensive care unit admission and increase in length and cost of hospital stay.^5,8,9

Prenatal care (PNC) is important and of great value because of its far and wide-reaching implications and long-term consequences on both the mother and the child. It identifies and manages clinical conditions that threaten the health of mother and/or unborn child and ensures women approach delivery in health. There are two approaches to PNC. Traditional or Individual type and Group PNC.¹⁰ The perceived shortcomings of individual prenatal care (IPNC) led to the start of group PNC, especially CenteringPregnancy (CP) program, in the 1990s in the United States to improve poor maternal and fetal health outcomes in its socioeconomically poorer populations.^11,12

South Carolina adopted the program in 2008 as a promising alternative to IPNC, based on its theory and evidence from published studies^13
–15; it was the first state to develop a coordinated statewide network with group PNC.¹² The CP program provides an integrated approach to PNC in the group setting by actively involving patients or expectant mothers of similar gestational ages or due date in their own health care management. It allows pregnant women to share learning and experiences, and provide social support to each other, leading to greater engagement, learning, and self-confidence.

Each group session typically has 8–12 women who meet every 2–4 weeks for ∼2 hours.^12,16 They are provided important information and resources relevant and specific for each trimester and gestational age. Through social support, the group format promotes psychological, social, and behavioral factors for change to promote healthy pregnancy and improved birth outcomes, as shown in prior studies.^17,18

In addition to the standard components of the program, the CP program in our study incorporates a unique nutrition component focused on eating healthy and breastfeeding, coupled with mild exercise and safe activities during pregnancy. Nutrition information is provided to program participants because of its potential influence on both the mothers' overall weight gain and fetal growth. This is stressed in all group sessions.

The program provides a daily guide to meals that include grains, dairy, protein, vegetables, fruit fat/oil, and water with serving size range per day. CP participants are held accountable for their daily diet by recording their food-group servings or drinks in a handbook, with the information used in discussions with the provider/coordinator as to how well their diet meets recommendations for healthy eating (Food Diary in Fig. 1). Tips for reducing calories include eating high-fiber foods, and baked, broiled, or roasted foods instead of fried foods, and limiting sauces, dressings, and gravies.

FIG. 1.

Food diary. Source: Prisma-USC Medical Group Centering Pregnancy Program Training Workshop Materials.

With the help of a nutritionist available during meetings, modifications were made to recommended food options to make it easier for patient adherence. Centering mothers were made to realize that they did not require extra calories in the first trimester, but needed only an extra 340 calories per day during the second trimester and an additional 450 calories per day in the third trimester.¹⁹ Participants were made to understand that they need to eat for two, need not eat double the amount, and need to avoid or save snacks and desserts high in sugar and or fat for treats. The health care staff encourages vegetarians and vegans to take vitamin supplements aside from prenatal vitamins, especially vitamin B12, which is not found in plant foods, and other accessories that might be helpful.

The nutritionist or provider teaches centering moms how to size up their meal components with practical guides, for example, three ounces of meat is about the size and thickness of a smartphone, one cup of mashed potatoes or broccoli is about the size of their fist, one teaspoon of butter or peanut butter is about the tip of their thumb, and one ounce of cheese is about four stacked pieces equaling the average size of a thumb. A reasonable amount of time (about 30 minutes) is spent on this topic in each session with a healthy food demonstration. Supposed calorie intake or weight gain is above or below recommendations according to Food Diary; in that case, the nutritionist/program coordinator makes changes during the meeting to reduce calorie intake or adjust the recommended food options to make it easier for patients to follow.

The CP providers are trained by the Centering Healthcare Institute (CHI), which provides consultation and support to clinical practice sites for system change, training in centering group facilitation, site approval for quality assurance, policy, and advocacy work to support scale-up and spread. The program was administered by OBGYN specialists, physicians, nurses, nutritionists/registered dieticians, and social workers for our sites. Most people (health program providers) are the same from session to session, but the specialists or physicians may differ. CP providers receive special training to develop practical facilitation skills. Two facilitators, typically a provider and a support staff member, lead discussions and direct activities to the best of group functioning.

The effect of the nutrition-enhanced CP program on GWG has never been evaluated. Studies of group PNC generally, including the CP (without nutrition enhancement) program, report mixed results. Some indicate group PNC is associated with increased GWG,^20
–22 while one study found reduction in excessive weight gain among group care participants,²³ and other studies report no statistically significant difference between participants of individual vis-a-vis group PNC.^14,24,25

Previous studies highlight the limitations to assessing GWG such as lack of awareness and monitoring of calorie content of daily food intake during pregnancy,²² assessing or measuring the actual GWG without giving consideration to prepregnancy weight or gestational age in accordance to IOM guidelines,^14,25
–27 not adjusting for their actual level of participation in the group care program,^24,28,29 and not controlling for self-selection into the program^27,29 or differences in matching criteria.^14,21,23 All these limitations may have affected the estimates of group PNC effect on GWG and could explain the variability in findings among studies. Therefore, given the existing literature and recognizing these limitations, this study examined the effects of the nutrition-enhanced CP program versus IPNC on GWG. We hypothesized that women who received CP group care will have either a higher likelihood or a higher rate of adequate GWG than women who received IPNC.

Methods

Study design, population, and data source

This is a retrospective study of women (de-identified from record) who registered between 2015 and 2019 for the CP program (hereafter referred to as centering group) at any of three (out of five) obstetric sites operated by a single medical group in Midland region of South Carolina. A comparison group (hereafter referred to as individual prenatal care or IPNC) was constructed of women who registered for individual prenatal care during the same time period at these PNC locations.

The participants in both groups were followed from registration until delivery. Two of the small practice sites were not considered because a large proportion of their patient populations included immigrant women who came in as an emergency and their electronic medical records were missing much of the information necessary for the study. The protocol was reviewed, approved, and documented by the Prisma Health Medical Group and the University of South Carolina Institutional Review Board (IRB-Pro00088616) for Human Research.

For centering and IPNC women, de-identified birth certificate data and hospital discharge records were received from the South Carolina Revenue and Fiscal Affairs Office (SC RFA) The SC RFA received identifiers from the medical group to identify and link the centering women's data in their database. About 98% of centering women were successfully matched. Reasons for failure to match include delivery outside South Carolina, at a birthing center versus a hospital, or delivery outside the study timeframe.

The same inclusion and exclusion criteria were applied to all participants (Flow chart in Fig. 2). The CP program is usually offered to women who meet the following inclusion criteria: PNC initiating within the first 20 weeks of pregnancy, 18–49 years of age (reproductive age group), and expected to have a singleton birth. Women were not excluded if they had previous low-birth-weight infants, mental health issues, prior cesarean birth, prepregnancy hypertension, or chronic hypertension. Women with data values relating to height, prepregnancy weight, prepregnancy body mass index (BMI), or infant birth weight that were unlikely to be valid were excluded.³⁰

FIG. 2.

Flow chart of centering and noncentering women for study selection.

Women with prepregnancy diabetes were also excluded because their physicians would have recommended modifying lifestyle-related factors, which would likely have affected their weight gain. Similarly, women who had multiple gestation were excluded from the study as their weight gain recommendation would be different from the singletons. After applying the above inclusion and exclusion criteria, the study population included 28,109 women with singleton births (568 births in centering group and 27,541 births in noncentering group or IPNC) (Fig. 2).

Study measures

Dependent variables

The primary outcome of interest was GWG, which was defined in three ways. As one definition, total GWG in kilograms was calculated as the difference between weight at delivery and prepregnancy weight, as documented on the birth certificate. Alternatively, using 2009 IOM recommendations, total weight gain was categorized into inadequate, adequate, and excessive weight gain based on prepregnancy BMI: adequate is defined as GWG of 12.5–18 kg for BMI <18.5 kg/m², 11.5–16 kg for BMI 18.5–24.9 kg/m², 7–11.5 kg for BMI 25.0–29.9 kg/m², and 5–9 kg for BMI ≥30 kg/m².⁵

The third definition was the weekly rate of weight gain over the second and third trimester, calculated as follows: total weight gain − expected first trimester GWG)/(gestational age at birth in weeks −13 weeks). This average weekly rate of weight gain for second and third trimesters was used as a continuous variable and categorized as inadequate, adequate, and excessive weight gain based on their prepregnancy BMI category using IOM guidelines. This last measure took into account the participants' pregnancy duration, which is related to GWG, even though preterm birth rate was low (11%) in our study population. Adequate average weekly rate of weight gain is defined as 0.5, 0.4, 0.4, and 0.2 kg/week for underweight, normal weight, overweight, and obese prepregnancy BMI categories, respectively. Gestational age at delivery is categorized into preterm (≤37 weeks) and term (>37 weeks) birth.^31
–33

Independent variables

The independent variable of interest is PNC program: CP program or individual prenatal care. Group assignment was determined from medical records and the birth certificate.

Covariates

As covariates, we used variables linked to GWG in the literature, including maternal age, race/ethnicity, level of education, smoking status during pregnancy, parity, gestational age at entry or time PNC began, and prepregnancy BMI. Covariate categorization is shown in Table 1. Other maternal characteristics that were assessed include number of PNC visits, Kotelchuck index, health insurance type, and Women, Infant, and Children (WIC) nutrition program participation.

Table 1.

Characteristics of Women Who Had CenteringPregnancy and Individual Prenatal Care Between 2015 and 2019 (N = 28,109)

Maternal characteristics	CP (N = 568)	IPNC (N = 27,541)	p
Maternal age, n (%)			<0.0001
<20	52 (9.15)	1,377 (5.00)
20–24	172 (30.28)	6,362 (23.10)
25–29	185 (32.57)	8,654 (31.42)
30–34	104 (18.31)	7,074 (25.69)
≥35	55 (9.68)	4,074 (14.79)
Race, n (%)			<0.0001
White	143 (25.18)	10,961 (39.80)
Black-African American	322 (56.69)	11,954 (43.41)
Asian (Indian, Chinese, Japanese, Hawaiian)	103 (18.13)	4,623 (16.79)
Level of education, n (%)			<0.0001
<High school	83 (14.64)	3,210 (11.71)
High school graduate	321 (56.61)	13,910 (50.72)
College graduate	114 (20.11)	7,189 (26.21)
Postgraduate	49 (8.64)	3,115 (11.36)
Health insurance type, n (%)			<0.0001
Private insurance	348 (61.27)	13,811 (50.21)
Government/Medicaid	154 (27.11)	10,299 (37.45)
Self-pay	31 (5.46)	1,289 (4.69)
Other pay type	35 (6.16)	2,105 (7.65)
Parity, n (%)			<0.0001
No child	372 (65.46)	11,866 (43.08)
One child	108 (19.01)	8,224 (29.86)
Two children	59 (10.39)	4,574 (16.61)
≥Three children	29 (5.11)	2,877 (10.45)
Prepregnancy BMI, mean ± SD	28.43 ± 7.24	28.46 ± 7.71	0.9227
Prepregnancy BMI, n (%)			0.472
Underweight	15 (2.64)	905 (3.39)
Normal	194 (34.15)	9,968 (36.19)
Overweight	160 (28.17)	7,122 (25.86)
Obese	199 (35.04)	9,546 (34.66)
Month PNC began, n (%)			<0.0001
0–3 Months	400 (70.42)	20,284 (73.68)
4–6 Months	147 (25.88)	5,326 (19.35)
7–9 Months	21 (3.70)	1,919 (6.97)
Smoking prepregnancy, n (%)			0.886
No	540 (95.07)	26,219 (95.20)
Yes	28 (4.93)	1,322 (4.8)
Smoking in pregnancy, n (%)			0.225
No	556 (97.89)	26,719 (97.02)
Yes	12 (2.11)	822 (2.98)
Total number of prenatal visits, mean ± SD	12.87 ± 4.61	11.70 ± 4.39	<0.0001
WIC participation, n (%)			<0.0001
No	257 (45.415)	14,504 (53.26)
Yes	309 (54.59)	12,726 (46.74)
Gestational age at delivery, mean ± SD	38.35 ± 2.16	38.11 ± 2.41	0.0055
Gestational age at delivery, n (%)			0.126
Preterm (≤37 weeks)	61 (10.74)	3,556 (12.91)
Term (>37 weeks)	507 (89.26)	23,982 (87.09)
Total GWG at delivery, mean ± SD	12.61 ± 8.41	12.68 ± 7.99	0.8465
Total weight gain categories at delivery according to IOM recommendation, n (%)			0.967
Inadequate	162 (28.52)	7,861 (28.54)
Adequate	132 (23.24)	6,518 (23.67)
Excess	274 (48.24)	13,162 (47.79)
Estimated weekly rate of weight gain (kg/week), mean ± SD	0.45 ± 0.33	0.46 ± 0.32	0.7171
Estimated weekly rate of weight gain according to IOM recommendation, n (%)			0.847
Inadequate	159 (27.99)	8,011 (29.09)
Adequate	81 (14.26)	3,906 (14.18)
Excessive	328 (57.75)	15,624 (56.73)

BMI, body mass index; CP, CenteringPregnancy; GWG, gestational weight gain; IOM, Institute of Medicine; IPNC, individual prenatal care; PNC, prenatal care; SD, standard deviation; WIC, Women, Infant and Children.

Statistical analysis

Maternal sociodemographic and medical history variables were compared for CP women versus those in individual prenatal care using descriptive statistics such as two-sample t-test for continuous variables and chi square (χ ²) test for categorical variables. To assess the effect of the program on GWG, whether as a continuous and categorical variable (with reference group consisting of women with adequate GWG), we used linear and multinomial logistic regressions in unadjusted and adjusted format.

To control for self-selection by the participant into type of PNC, propensity score analysis (PSA) was performed using the above-mentioned covariates. There are several approaches to propensity score method, but no clear consensus as to which is preferred. Two PSA methods were selected that best matched the array of approaches to PSA from simplest to complex, allowing us to also assess the robustness of the results: propensity score used in covariate adjustment, and Inverse Probability of Treatment Weighting (IPTW), which is more robust with heteroscedasticity in observational study data and all participants are retained in analysis. The propensity score, that is, the probability to enroll in the CP program, was derived from a logistic regression model. The effect of CP versus individual prenatal care on GWG (measured as continuous or categorical variables) was then estimated, adjusting for derived propensity scores.

In that regard, linear and multinomial logistic regressions were used for the first propensity score method to obtain effect estimates, standard errors, and confidence intervals (CIs). For the second method (IPTW), propensity score weights (weighted predicted probabilities) were generated and were used in linear and multinomial logistic regression models to obtain effect estimates, standard errors, and CIs.³⁴ The effect estimates from the two propensity score methods were compared. Supplementary sensitivity analysis was performed to determine robustness of results by restricting the study to those women who began PNC in first trimester only. A second sensitivity analysis was performed by excluding those women who delivered at preterm (<37 weeks gestation). All statistical analyses were performed using Stata/SE statistical software version.

Results

Of the 28,109 women who had PNC between 2015 and 2019 and met inclusion and exclusion criteria, 27,541 received individual prenatal care, while 568 received group PNC through the CP program. As shown in Table 1, women participating in the CP program were more likely to be of Black-African race, younger, and with lower educational attainment than their counterparts in individual prenatal care (p < 0.0001).

CP participants were also more likely to have private health insurance (61.27%) compared to those in individual prenatal care (50.21%) (p < 0.0001). Almost two-thirds (65.46%) of pregnancies among centering women were first births compared with 43.08% among those who received individual prenatal care (p < 0.0001). Average total number of PNC visits, and average gestational age at delivery were higher among centering women (p < 0.0001). No difference was observed between the groups with respect to delivering at term, smoking during pregnancy, prepregnancy BMI, or gestational age categories. Adjustment for nonrandom (self) selection of PNC type did not modify these relationships (Table 2).

Table 2.

Odd Ratios from Propensity Score Estimation to Predict Participation in CenteringPregnancy Versus Individual Prenatal Care (N = 27,974)

Covariates	Odd ratio (95% CI)	p
Maternal age
<20	Reference
20–24	0.97 (0.70–1.34)	0.854
25–29	1.11 (0.79–1.56)	0.551
30–34	0.90 (0.62–1.33)	0.611
≥35	0.93 (0.60–1.44)	0.738
Race
White	Reference
Black-African American	2.04 (1.65–2.51)	0.0001
Asian (Indian, Chinese, Japanese, Hawaiian)	1.72 (1.32–2.23)	0.0001
Parity
No child	Reference
One child	0.40 (0.32–0.51)	0.0001
Two children	0.37 (0.28–0.50)	0.0001
≥Three children	0.26 (0.18–0.41)	0.0001
Prepregnancy BMI
Underweight	Reference
Normal	1.35 (0.79–2.30)	0.267
Overweight	1.57 (0.92–2.69)	0.101
Obese	1.39 (0.81–2.37)	0.229
Level of education
<High school	Reference
High school graduate	0.73 (0.57–0.95)	0.019
College graduate	0.56 (0.41–0.77)	0.0001
Postgraduate	0.58 (0.39–0.86)	0.007
Month PNC began
0–3 Months	Reference
4–6 Months	1.23 (1.03–1.53)	0.022
7–9 Months	0.51 (0.33–0.81)	0.003
Smoking in pregnancy
No	Reference
Yes	0.83 (0.46–1.49)	0.531

CI, confidence interval.

GWG, whether defined in terms of total weight gain or rate of weekly gain, was similar among participants, regardless of type of PNC received (Table 3). This held with and without adjustment for self-selection into type of PNC. Use of IOM categories yielded similar findings (Table 4).

Table 3.

Comparing Total Weight Gain and Weekly Rate of Weight Gain Between CenteringPregnancy and Individual Prenatal Care Participants: Results from Linear Regression Models

CP versus IPNC (total weight gain in kg)	Coefficients (95% CI)	p
Unadjusted model	−0.069 (−0.73 to 0.61)	0.839
Adjusted model^a	−0.041 (−0.68 to 0.61)	0.9
Propensity score adjusted	−0.044 (−0.71 to 0.62)	0.897
Propensity score weight adjusted (IPTW)	−0.054 (−0.78 to 0.68)	0.883
CenteringPregnancy versus IPNC (weekly weight gain in kg)
Unadjusted model	−0.005 (−0.032 to 0.022)	0.711
Adjusted model^a	−0.006 (−0.032 to 0.021)	0.671
Propensity score adjusted	−0.006 (−0.033 to 0.021)	0.679
Propensity score weight adjusted (IPTW)	−0.004 (−0.033 to 0.025)	0.798

Covariate adjusted includes maternal age, race-ethnicity, level of education, parity, smoking status, prepregnancy BMI, and time at initiating prenatal care.

IPTW, Inverse Probability of Treatment Weighting.

Table 4.

Comparing Categorical Total Weight Gain and Weekly Rate of Weight Gain Between CenteringPregnancy and Individual Prenatal Care Participants: Results from Multinomial Regression Models

	RRR (95% CI)	p
CP versus IPNC (total weight gain in kg at delivery according to IOM recommendation)
Unadjusted model
Adequate	Reference
Inadequate	1.02 (0.78–1.24)	0.883
Excessive	1.03 (0.83–1.23)	0.797
Adjusted model^a
Adequate	Reference
Inadequate	0.97 (0.81–1.31)	0.832
Excessive	0.97 (0.81–1.21)	0.769
Propensity score adjusted
Adequate	Reference
Inadequate	0.97 (0.76–1.22)	0.778
Excessive	0.95 (0.82–1.21)	0.624
Propensity score weight adjusted (IPTW)
Adequate	Reference
Inadequate	0.85 (0.64–1.13)	0.259
Excessive	0.92 (0.71–1.20)	0.557
CenteringPregnancy versus IPNC (weekly rate of weight gain in kg according to IOM recommendation)
Unadjusted model
Adequate	Reference
Inadequate	0.96 (0.73–1.25)	0.751
Excessive	1.01 (0.87–1.28)	0.922
Adjusted model^a
Adequate	Reference
Inadequate	0.92 (0.70–1.21)	0.561
Excessive	0.96 (0.74–1.22)	0.706
Propensity score adjusted
Adequate	Reference
Inadequate	0.92 (0.71–1.21)	0.547
Excessive	0.95 (0.74–1.22)	0.673
Propensity score weight adjusted (IPTW)
Adequate	Reference
Inadequate	0.73 (0.53–1.02)	0.064
Excessive	0.83 (0.61–1.13)	0.232

Covariate adjusted includes maternal age, race-ethnicity, level of education, parity, smoking status, prepregnancy BMI, and time at initiating prenatal care.

RRR, relative risk ratio.

Confining the study population to women who started PNC in the first trimester in both groups, CP participation was associated with lower odds of inadequate GWG versus adequate GWG compared to those women who had individual prenatal care, but achieved statistical significance only for the weekly rate of weight gain under the IPTW method (data not shown). The program (compared to IPNC) was also associated with lower odds of excessive GWG versus adequate GWG, but, again, lacked statistical significance. By restricting the study population to those with term births, similar results were observed.

Discussion

This retrospective study examined the association between a nutrition-enhanced CP program versus individual prenatal care program on GWG using PSAs to account for nonrandom assignment to either type of PNC. We found that participants in the CP program enhanced with a nutrition component had a similar GWG experience compared to those in individual prenatal care. Sensitivity analyses suggest that the findings cannot be explained by either fewer prenatal visits or factors affecting fetal growth.

The role of unobserved confounding factors cannot be ruled out; statistical adjustment may have been insufficient. Evidence from prior research, including meta-analysis, demonstrates mixed results or inconsistent findings: previous retrospective studies show that group care is associated with increased GWG among low-income African American women, Latina-Spanish speaking women, and non-Hispanic population,^20
–22 reduction in excessive weight gain among group care participants,²³ or no statistically significant difference between GWG among participants in either type of PNC, as shown in this study.^14,24,25

Several important limitations identified in previous studies, which may have led to this inconsistency, were addressed in our study, including lack of awareness and monitoring of calorie content of daily food intake during pregnancy,²² assessing or measuring GWG without consideration for prepregnancy BMI and gestational age in accordance to IOM guidelines, as seen in most studies,^14,26,27 and not controlling for self-selection into the program^27,29 coupled with differences in matching criteria in few studies.^14,21,23 This study used PSAs to account for or reduce the problem of confounding of patient characteristics and lack of randomization to treatment (CP program).

This study examined a CP program that had a nutrition-specific contents/curriculum, which has not been assessed previously. The medical group offering the group PNC specifically designed the CP program to target nutrition and GWG among participants. The health providers, nutritionist, or health educator provided guidance to daily meal-serving, made available some healthy food choices for demonstration, and eating with emphasis placed on how to attain adequate GWG based on prepregnancy BMI (desired result) during sessions.

Possible explanations for lack of effect include varying fidelity to the program in its delivery or participants were not sufficiently engaged and activated in following recommendations. Neither fidelity to the program in its delivery nor uptake by the participants was assessed. Further studies are recommended to look at the treatment effect in the nutrition-enhanced CP program, preferably using a randomized controlled trial format that specifically evaluates GWG outcomes and includes components to maximize program fidelity in delivery and in uptake.

A strength of this study is that the accuracy of GWG records on birth certificate was validated by comparing it with data in the electronic medical record for a subsample of centering women (30%). Previous studies that looked at the outcome as measure of effectiveness of CP versus individual prenatal care highlighted that GWG records or reported values on birth certificates may be inaccurate.^22,35

Overall, the study found birth certificate data of height, prepregnancy BMI, prepregnancy weight, delivery weight, and GWG categories to be similar to that of the electronic medical record, and thus, the birth certificate can provide reasonable prevalence estimates for GWG and relevant variables. Using linked birth certificate and hospital discharge data was another strength that provided linkage for 95% of patients with valuable information on diagnoses and clinical variables for the analyses. PSA was another strength as it provided some control for the problem of confounding from lack of randomization to treatment (CP).

Several limitations to this study are recognized. The study could not account for the actual level of participation among the CP women. It would have been ideal to use the total number of CP prenatal visits to address this, but the variable was not available in birth certificate data or hospital discharge records used for analyses.

The total number of prenatal visits was available, but patients may have attended the clinic for different reasons other than for scheduled CP group meetings. The same concern applies to the Kotelchuck index variable that assesses the adequacy of PNC utilization based on the time PNC began and the number of prenatal visits from PNC initiation until delivery. Further studies should consider the use of medical record or other data to obtain a more precise characterization of exposure to PNC, which may impact the effectiveness of CP on pregnancy-related outcomes. We also recognize the unavailability of data on fidelity to the program in its delivery and can only assume that it was followed as prescribed. Further investigation through more rigorous study designs, including rigorous evaluation of program fidelity, is warranted.

A relatively low percentage of pregnant women participated in the CP program, and a greater proportion were Black women, carrying their first pregnancy (nulliparous) compared to those in individual prenatal care. Although the study attempted to minimize these limitations through PSA, some residual bias may have remained. While CP program was originally adopted to improve poor maternal and fetal health outcomes among socioeconomically poorer populations, our data showed that women with private insurance were more likely to participate in the CP program. Future studies should examine the cause for this disparity in program enrollment and assess CP program's effectiveness among underserved population specifically.

Conclusion

In conclusion, the study found no significant difference in GWG for women who enrolled in the CP program enhanced with a nutrition component compared to individual prenatal care. Theoretically, the special nutritional knowledge component of the program along with its other elements should have assisted participant women to achieve appropriate or adequate weight gain in pregnancy, as recommended by IOM guideline. Further studies of the program are recommended with attention to fidelity in its delivery and the uptake by the participants.

Footnotes

Acknowledgments

I would like to express my sincerest gratitude to God almighty who has given me strength and empowerment in all the challenging moments of completing the doctoral dissertation. I would also like to specially thank my darling husband (Mr. Shaka Momodu) for his unflinching support, my loving children (Alfred, Isabella, and Merilyn), and all well-wishers, who one way or the other have been there for me throughout my academic journey. I am so thankful to my dissertation chairs and committee members, who are co-authors (Dr. Jihong Liu, Dr. Elizabeth Crouch, Dr. Brian Chen, and Dr. Ronnie D. Horner), for their immense support, guidance, and commitment toward the success of the dissertation project.

I am also grateful to the CenteringPregnancy Manager, Ms. Beverly Huegel at Prisma Health-University of South Carolina OBGYN department, who provided support to the conception and logistics to data acquisition. I also extend my appreciation to Ivory Harding, Allyson Malbouf, and other members of the OBGYN department and information technology department for their assistance with research logistics and participants' data gathering. I could not have successfully gathered without their help and support. I acknowledge Dr. James Hardin, Interim Department Chair, Department of Health Services Policy and Management, Arnold School of Public Health, University of South Carolina, for his support and willingness to provide statistical insights and guidance from his wealth of experience when I needed it.

Authors' Contributions

All authors have made substantial contributions to all aspects of the research. O.A.M.: Conception, design of the work, acquisition of data, formal data analysis, interpretation of data, original drafting/article preparation, editing of work, and substantial revision. J.L. and R.D.H.: Conception, design of the work, acquisition of data, interpretation of data, editing of work, substantial revision, and supervision. E.C. and B.C.: Conception, design of the work, interpretation of data, editing of work, substantial revision, and supervision.

Disclaimer

The research presented in this article is that of the authors and does not reflect the official policy of the NIH or other research funding agencies.

Author Disclosure Statement

No competing financial interests exist.

Funding Information

No funding was received for this article.

References

South Carolina Law Review. Preventing the Preventable: A Review of Maternal Mortality Rates in South Carolina; Published August 8, 2021. Available from: https://sclawreview.org/preventing-the-preventable-a-review-of-maternal-mortality-rates-in-south-carolina/ [Last accessed: May 29, 2023 ].

Latest South Carolina

Infant

, Maternal Mortality Reports Reveal Alarming Trends |

SCDHEC

. Available from: https://scdhec.gov/news-releases/latest-south-carolina-infant-maternal-mortality-reports-reveal-alarming-trends [Last accessed: May 29, 2023].

Wende

, Liu

, Mclain

, Wilcox

. Gestational weight gain disparities in South Carolina: Temporal trends, 2004–2015. Paediatr Perinat Epidemiol, 2021; 35(1):37–46; doi: 10.1111/ppe.12706

Rode

, Kjærgaard

, Ottesen

, et al. Association between gestational weight gain according to body mass index and postpartum weight in a large cohort of Danish women. Matern Child Health J, 2012; 16(2):406–413.

Report Brief—Weight Gain During Pregnancy.pdf. Available from: https://www.nationalacademies.org/hmd/∼/media/Files/Report%20Files/2009/Weight-Gain-During-Pregnancy-Reexamining-the-Guidelines/Report%20Brief%20-%20Weight%20Gain%20During%20Pregnancy.pdf [Last accessed: March 9, 2019].

Goldstein

, Abell

, Ranasinha

, et al. Association of gestational weight gain with maternal and infant outcomes: A systematic review and meta-analysis. JAMA, 2017; 317(21):2207–2225; doi: 10.1001/jama.2017.3635

McDowell

, Cain

, Brumley

. Excessive gestational weight gain. J Midwif Womens Health, 2019; 64(1):46–54; doi: 10.1111/jmwh.12927

Premature-birth-report-card-south-carolina.pdf. Available from: https://www.marchofdimes.org/materials/premature-birth-report-card-south-carolina.pdf [Last accessed: March 8, 2019].

Consequences of Gestational Weight Gain for the Mother. Weight Gain During Pregnancy. NCBI Bookshelf. Available from: https://www.ncbi.nlm.nih.gov/books/NBK32818/ [Last accessed: July 21, 2019 ].

10.

American Academy of

Pediatrics

, American College of Obstetricians and

Gynecologists

, eds. Guidelines for Perinatal Care. 7th ed. American Academy of Pediatrics; American College of Obstetricians and Gynecologists; 2012; p 580.

11.

Carter

, Temming

, Akin

, et al. Group prenatal care compared with traditional prenatal care: A systematic review and meta-analysis. Obstet Gynecol, 2016; 128(3):551–561; doi: 10.1097/AOG.0000000000001560

12.

Van De Griend

, Billings

, Frongillo

, et al. Core strategies, social processes, and contextual influences of early phases of implementation and statewide scale-up of group prenatal care in South Carolina. Eval Program Plann, 2020; 79:101760; doi: 10.1016/j.evalprogplan.2019.101760

13.

Hale

, Picklesimer

, Billings

, et al. The impact of Centering Pregnancy Group Prenatal Care on postpartum family planning. Am J Obstet Gynecol, 2014; 210(1):50.e1–50.e7; doi: 10.1016/j.ajog.2013.09.001

14.

Zielinski

, Stork

, Deibel

, et al. Improving infant and maternal health through CenteringPregnancy: A comparison of maternal health indicators and infant outcomes between women receiving group versus traditional prenatal care. Open J Obstet Gynecol, 2014; 4(9):497–505; doi: 10.4236/ojog.2014.49071

15.

Picklesimer

, Billings

, Hale

, et al. The effect of CenteringPregnancy group prenatal care on preterm birth in a low-income population. Am J Obstet Gynecol, 2012; 206(5):415.e1–415.e7; doi: 10.1016/j.ajog.2012.01.040

16.

Schindler Rising

, Kennedy

, Klima

. Redesigning prenatal care through CenteringPregnancy. J Midwif Womens Health. 2004; 49(5):398–404; doi: 10.1016/j.jmwh.2004.04.018

17.

Ickovics

, Kershaw

, Westdahl

, et al. Group prenatal care and perinatal outcomes: A randomized controlled trial. Obstet Gynecol, 2007; 110(2, Part 1):330–339; doi: 10.1097/01.AOG.0000275284.24298.23

18.

Ickovics

. Group prenatal care and preterm birth weight: Results from a matched cohort study at public clinics. Obstet Gynecol, 2003; 102(5):1051–1057; doi: 10.1016/S0029-7844(03)00765-8

19.

Nutrition During Pregnancy. Available from: https://www.acog.org/Womens Health/FAQs/Nutrition During Pregnancy [Last accessed: December 6, 2020].

20.

Klima

, Norr

, Vonderheid

, et al. Introduction of CenteringPregnancy in a public health clinic. J Midwifery Womens Health, 2009; 54(1):27–34; doi: 10.1016/j.jmwh.2008.05.008

21.

Trudnak

, Arboleda

, Kirby

, et al. Outcomes of Latina women in CenteringPregnancy group prenatal care compared with individual prenatal care. J Midwif Womens Health, 2013; 58(4):396–403; doi: 10.1111/jmwh.12000

22.

Kominiarek

, Crockett

, Covington-Kolb

, et al. Association of group prenatal care with gestational weight gain. Obstet Gynecol, 2017; 129(4):663–670; doi: 10.1097/AOG.0000000000001940

23.

Tanner-Smith

, Steinka-Fry

, Gesell

. Comparative effectiveness of group and individual prenatal care on gestational weight gain. Matern Child Health J, 2014; 18(7):1711–1720; doi: 10.1007/s10995-013-1413-8

24.

Brumley

, Cain

, Stern

, et al. Gestational weight gain and breastfeeding outcomes in group prenatal care. J Midwifery Womens Health, 2016; 61(5):557–562; doi: 10.1111/jmwh.12484

25.

Walton

, Shaffer

, Heaton

. Group prenatal care outcomes in a military population: A retrospective cohort study. Mil Med, 2015; 180(7):825–829; doi: 10.7205/MILMED-D-14-00273

26.

Mazzoni

, Hill

, Webster

, et al. Group prenatal care for women with gestational diabetes. J Matern Fetal Neonatal Med, 2016; 29(17):2852–2856; doi: 10.3109/14767058.2015.1107541

27.

Trotman

, Chhatre

, Darolia

, et al. The effect of centering pregnancy versus traditional prenatal care models on improved adolescent health behaviors in the perinatal period. J Pediatr Adolesc Gynecol, 2015; 28(5):395–401; doi: 10.1016/j.jpag.2014.12.003

28.

Harden

, Beauchamp

, Pitts

, et al. Group-based lifestyle sessions for gestational weight gain management: A mixed method approach. Am J Health Behav, 2014; 38(4):560–569; doi: 10.5993/AJHB.38.4.9

29.

Holbrook, Helene

. Exploring the Value of Group and Traditional Obstetrical Appointments to Reduce Health Disparity. Doctor of Nursing Practice (DNP) Projects. 4; 2010. Available from: https://repository.usfca.edu/dnp/4 [Last accessed: December 3, 2020 ].

30.

Definitions | Data, Trends and Maps | CDC; Published July 8, 2020. Available from: https://www.cdc.gov/nccdphp/dnpao/data-trends-maps/help/npao_dtm/definitions.html [Last accessed: April 18, 2021].

31.

Bodnar

, Siega-Riz

, Simhan

, et al. Severe obesity, gestational weight gain, and adverse birth outcomes. Am J Clin Nutr, 2010; 91(6):1642–1648; doi: 10.3945/ajcn.2009.29008

32.

Bodnar

, Siega-Riz

, Arab

, et al. Predictors of pregnancy and postpartum haemoglobin concentrations in low-income women. Public Health Nutr, 2004; 7(6):701–711; doi: 10.1079/phn2004597

33.

Bodnar

, Hutcheon

, Abrams

. Biases in studying gestational weight gain and infant mortality in U.S. birth certificates. Matern Child Health J, 2012; 16(4):745–746; doi: 10.1007/s10995-012-0987-x

34.

Olmos

, Govindasamy

. A practical guide for using propensity score weighting in R. Pract Assess Res Eval, 2019; 20:13; doi: 10.7275/jjtm-r398

35.

Dietz

, Bombard

, Mulready-Ward

, et al. Validation of selected items on the 2003 U.S. standard certificate of live birth: New York City and Vermont. Public Health Rep, 2015; 130(1):60–70; doi: 10.1177/003335491513000108