Gingival changes during pregnancy have been well-documented.² Increases in both the rate of estrogen metabolism by the gingiva and in the synthesis of prostaglandins contribute to the gingival changes observed during pregnancy.⁸ Alterations in progesterone and estrogen levels have been shown to affect the immune system and the rate and pattern of collagen production in the gingiva, thus reducing the body’s ability to repair and maintain gingival tissue.^9,10 Periodontal infections, which can be a reservoir for inflammatory mediators, may pose a potential threat to the placenta and fetus, thereby increasing the likelihood of preterm delivery.⁴ Preliminary analysis of data from an ongoing study in Alabama shows that mothers with severe periodontitis have high levels of prostaglandin in their blood and are more likely to deliver preterm babies (Marjorie Jeffcoat, D.M.D., unpublished data, May 7, 2000). The preliminary analysis suggests that these levels of prostaglandins may be associated with early uterine contractions. In turn, increased levels of prostaglandins in the blood may increase sensitivity to irritants, causing additional inflammation.

Access to dental care during pregnancy is impeded, in part, by a limited window of opportunity for treatment.

Access to dental care during pregnancy is impeded, in part, by a limited window of opportunity for treatment. The ADA suggests that elective dental care should be avoided, if possible, during the first trimester and the last one-half of the third trimester.¹¹ This time frame apparently is widely recommended because it includes the periods of greater risk of harm to the developing embryo or fetus,⁸ as well as the least comfort for the mother.^12–14 During the first trimester, risks of birth defects associated with the use of teratogens are higher than in the other two trimesters.⁸ In addition, a large number of pregnancies undergo spontaneous abortion during the first trimester, and any dental procedures performed around the time of the spontaneous abortion could be perceived as causal.⁸ During the last one-half of the third trimester, the increased sensitivity of the uterus to external stimuli increases the risks associated with premature delivery.^8,11 Other factors that can limit access to dental care during pregnancy include lack of insurance coverage, women’s beliefs about dental treatment during pregnancy, and providers’ (both obstetricians’ and dentists’) attitudes and beliefs.

Until recently, state-level population-based data on the oral health of pregnant women have not been available. We gathered such data by using an existing information system and examined the prevalence and correlates of dental attendance during pregnancy. These data may be useful in developing new programs, and evaluating existing ones may help policy-makers and program planners design interventions that could improve general and oral health outcomes.

	METHODS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
We analyzed data from the Pregnancy Risk Assessment Monitoring System, or PRAMS, an ongoing, state-level (except for New York City, which is separate from New York state, as it is a separate vital records registration area), population-based survey of maternal behaviors and experiences before, during and shortly after a woman’s pregnancy. For this survey, all of the participating areas use standardized mail-telephone methodology for data collection.¹⁵ Each month, mothers who have given birth during the previous two to four months are sampled from state birth certificates and asked to participate in the survey. A questionnaire is sent to mothers chosen as the sample population; if they do not respond, two additional questionnaires are sent, and extensive telephone follow-up is done to increase questionnaire completion. The only mothers excluded from the sample are adoptive mothers, those who reside outside of the state where the birth occurred, and those who are mentally incapacitated or deceased. Data are statistically weighted to account for nonresponse, non-coverage and sample stratification of birth certificate files. Full details of the methodology have been described elsewhere.^16,17

Of the 25 areas that participate in PRAMS (as of Oct. 1, 2000) (Figure 1), four (Arkansas, Illinois, Louisiana and New Mexico) collected oral health information in 1997–1998 and offered a Spanish translation of the questionnaire. Each of these states asked women if they went to a dentist or dental clinic during their most recent pregnancy, and three (Illinois, Louisiana and New Mexico) also asked women if they needed to see a dentist for a problem during their pregnancy. We analyzed 1998 data from Arkansas, Illinois and Louisiana and data from July 1997-December 1998 from New Mexico. During this period, weighted response rates for each of these states exceeded 70 percent.

View larger version (94K): [in this window] [in a new window]   Figure 1. States participating in the Pregnancy Risk Assessment Monitoring System.

Compared with mothers whose prenatal care was paid for by private insurance or a health maintenance organization, mothers on Medicaid were 24 percent to 53 percent less likely to seek dental care.

Socioeconomic and pregnancy-related information were obtained from responses to the PRAMS questionnaire. Annual household income was collected by Arkansas, Illinois and Louisiana; New Mexico chose to collect monthly household income. Entry into prenatal care, or PNC, was defined as "first trimester," "later" (after the first trimester), or "no prenatal care." Sources of PNC were defined as private physician, health maintenance organization—or HMO—or nurse or non-nurse midwife; health department, community health center or primary care center; hospital clinic or state hospital; or Indian Health Service.

For method of PNC payment, mothers could check multiple sources. When one payment method was chosen, women were classified accordingly: Medicaid, private health insurance or HMO, personal income or other sources (for example, military or special state-specific health programs). If women selected more than one payment method, the primary method was assigned according to the following hierarchy: Medicaid, private health insurance, other, personal income. Pregnancy was defined as "intended" if the woman had wanted the pregnancy then or sooner and as "unintended" if the woman had wanted the pregnancy later or not at all. As a proxy for persistent smoking throughout pregnancy, we categorized maternal smoking as "yes" if women reported smoking one or more cigarettes per day during the last trimester and "no" if they did not.

To accommodate the complex survey design of PRAMS, we used SUDAAN software, Release 6.40,¹⁸ to calculate estimates; standard errors; risk ratios, or RRs; and 95 percent confidence intervals, or CIs, for each state. RRs measure the magnitude of the bivariate association between specific characteristics and dental care–seeking among women who reported having a dental problem during pregnancy. An RR was considered significant if its 95 percent CI did not include "1." Variable categories associated with a lower risk of experiencing a poor birth outcome were selected as the referent category. Because of small numbers, we reclassified income level and method of PNC payment into more inclusive categories. Because of small numbers, we could not calculate RRs for mothers who did not receive PNC or who were not either black or white. We excluded responses coded as "missing," "blank" or "don’t know" from the analyses.

	RESULTS

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
The sampled population represented all recently delivered mothers in the four states. Even though state birth populations varied on some characteristics, we observed many similarities. Most mothers were younger than 30 years of age, had received at least a high school education, were white (except for in New Mexico where 50.1 percent of mothers were "not Hispanic") and were married (Table 1). Household income levels varied considerably across states. With the exception of Illinois, more than one-half the mothers in the other three states said their pregnancies were unintended. In all four states, a large majority of infants were of normal birthweight and were born at 38 weeks’ gestation or later. Although private health care insurance or HMO plans were the prevalent payment method for PNC in Illinois, state Medicaid plans were the most common methods in the other three states. In all four states, private physicians or HMO providers were the most frequently reported source of PNC, although hospital clinics provided PNC for a sizable proportion of women in these states.

View larger version (34K): [in this window] [in a new window]   Figure 2. Prevalence of reported dental experiences during pregnancy, Pregnancy Risk Assessment Monitoring System, 1998. Note that data were collected from July 1997 to December 1998 in New Mexico.

View larger version (29K): [in this window] [in a new window]   Figure 3. Prevalence of dental care seeking among mothers who reported having a dental problem during pregnancy, Pregnancy Risk Assessment Monitoring System, 1998. Note that data were collected from July 1997 to December 1998 in New Mexico.

Associations between maternal age and dental care during pregnancy were not consistent across states or categories. For example, 20- to 24-year-old New Mexico mothers were significantly less likely to seek dental care (51 percent, RR = 0.49) compared with older mothers, whereas 25- to 29-year-old Illinois mothers were significantly less likely to seek dental care (26 percent, RR = 0.74) compared with older mothers. In contrast, Louisiana mothers 15 to 19 years of age were more likely (33 percent, RR = 1.33) than older mothers to seek care, but the difference was not statistically significant. Neither maternal race nor smoking during the third trimester were associated with seeking dental care.

	DISCUSSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
This study provides the first multistate, population-based data that characterize the reported dental experiences and behaviors of pregnant women, a group with demonstrated oral health needs.

Our study results are similar to those from a survey of North Dakota mothers¹⁹ gathered in 1996. Relatively small proportions of new mothers had been to the dentist during their most recent pregnancy: 34.7 percent in the four PRAMS states in 1997 and 1998, and 43 percent in North Dakota in late 1995 and early 1996. Because most adults would be due for a routine dental visit during any nine-month period and because the need for periodontal care actually may increase during pregnancy,² our findings raise serious concerns about dental care–seeking behaviors during pregnancy. Even among those who reported having a dental problem during pregnancy, only about one-half said they had gone to a dentist or dental clinic during that interval. Although we do not know the nature and magnitude of these dental problems, the timing of the reported visits or the specific services provided, most dental problems that occur during pregnancy can be addressed and resolved during a dental visit.

Some patterns of dental visits among the pregnant women we studied were similar to those regularly found among other population groups (for example, adults in 25 states,²⁰ in national surveys²¹ and among the North Dakota mothers¹⁹). For example, women with the highest household incomes or with education beyond high school were more likely than women with lower incomes or educational levels to have gone to the dentist during their pregnancies. North Dakota mothers reported having a higher level of education (67 percent reported having some college)¹⁹ compared with those in any of the four PRAMS states, which may explain why dental attendance among North Dakota mothers exceeded the proportion reported by PRAMS states.

The prevalence of dental visits, which varied among the four states, probably reflects factors such as prevailing attitudes toward dental care, provider availability and practice norms, and salient features of medical and dental care delivery within the state. For example, Medicaid coverage of dental care for adults in the four states ranges from minimal to comprehensive²²; Arkansas, which has the lowest reported rate of dental attendance, does not cover any routine adult dental services under its Medicaid program. In the four states we studied, a smaller proportion of women who relied on Medicaid for their PNC went to the dentist than did their peers who had private insurance coverage. Also, women entering PNC during their first trimester were more likely to have gone to the dentist than were those who entered PNC care later in their pregnancies. Although these two findings suggest possible avenues for development of program interventions, more comprehensive analysis of these pilot data—as well as data collected in future PRAMS surveys from other states—could be helpful. Such analyses may reveal factors associated with dental care that are open to change through education of patients and providers, systems modifications or incentives for collaboration among health care providers.

PRAMS is becoming a source of state-specific data used to monitor oral health in the context of broader surveillance of maternal and child health. Because PRAMS uses a standardized questionnaire and protocol, employs rigorous methods for data collection and has response rates greater than 70 percent, it is a source of high-quality data that are routinely used by state-based programs to improve maternal and child health. Among PRAMS states, awareness of the value of oral health during pregnancy has expanded: 14 of 25 participating areas have included one or more oral health questions on their year 2000 survey (Figure 1). For the 2000 survey, eight states asked a question about the recency of dental prophylaxis. With this information, the degree of preventive orientation toward oral health among respondents can be characterized.

Although advice to confine dental treatment to the second trimester remains common in the dental literature,^8,11–14 we were unable to find any professional group that has formal policies related to dental care during pregnancy or any studies that examined pregnancy outcomes related to dental visit behaviors during pregnancy. Shrout and colleagues²³ conducted a survey of obstetricians who directed obstetrics wards in hospitals or headed medical school obstetrics departments. Among the 50 percent who responded, 91 percent said that consultations with dentists before dental treatment were unnecessary, and 60 percent said that dentists were too cautious in their care of pregnant patients. Eighty-eight percent of respondents, however, were concerned that dentists may prescribe antibiotics without a consultation. Consideration of the current research by a multidisciplinary group of experts would be timely.

Our study has several limitations. First, because PRAMS relies on self-reported data, it is subject to the biases inherent in this method (for example, misclassification, unknown validity). Because most respondents completed the questionnaire within four months of delivery, however, recall bias should be minimal. Second, our findings cannot be generalized to states other than the four included in this analysis. In addition, the survey excluded mothers who no longer cared for their infants or who were nonresidents of the state in which they gave birth. Finally, the cross-sectional nature of PRAMS and the wording of the questions precluded establishing any cause-effect relationships among variables. Despite these limitations, PRAMS can serve as an important mechanism for monitoring reported oral health needs and behaviors, particularly as they relate to salient health outcomes of pregnancy.

	CONCLUSION

TOP ABSTRACT METHODS RESULTS DISCUSSION CONCLUSION REFERENCES

 
Including of oral health content in PRAMS offers an avenue for including oral health measures in the broader context of health, using state-specific information to plan and evaluate oral health interventions among pregnant women and monitoring trends in those measures over time. In short, PRAMS can be an invaluable resource for monitoring oral health and oral disease prevention.