Self-Perceived Oral Health and Use of Dental Services by Pregnant Women in Surrey, British Columbia

Maintaining good oral health and preventing oral disease before, during and after pregnancy is an important aspect of general health for both mother and newborn.¹ Receiving oral health care — including a compressive dental examination and periodontal assessment, radiographs, pain medication, most dental treatments and professional scaling — is considered safe during pregnancy. In fact, it is recommended that pregnant women visit a dental professional if they have not done so in the last 6 months.²

Poor maternal oral health can not only affect the mother’s nutrition and oral health-related quality of life, but it is also believed to be associated with early childhood caries and long-term systemic disorders for the newborn.³ The harmful effects of poor oral health during pregnancy can be even more prominent among women who are vulnerable because of their race, ethnicity, economic disparities and other psychosocial and environmental factors.⁴

In 2014, there were 386 000 pregnancies in Canada, 43 700 of them in British Columbia.⁵ Yet, there are no baseline data describing the oral health status of Canadian pregnant women in general or of British Columbians in particular. Determining the oral health status and needs and the use of dental services by pregnant women, especially those who are vulnerable during pregnancy, may help inform future interventions for this patient group.³ Therefore, the aim of this study was to explore those issues in a sample of pregnant women living in British Columbia.

Methods

Study Participants and Data Collection

Fraser Health is one of 6 health authorities in British Columbia serving more than 1.6 million people, including a large population of Indo-Canadians, Korean-Canadians and Filipino-Canadians. Current oral health promotion initiatives include fluoride varnish application for children up to the age of 4 years, oral health education in elementary schools, preschool dental resource kits for children at age 3, and oral health care programs tailored for adults with developmental disabilities. We collaborated with the prenatal program called Best Beginnings, which is similar to programs offered by other provincial health authorities in British Columbia to support pregnant women and their newborns.

Study participants were pregnant women from the family planning and birth unit at Surrey Memorial Hospital (SMH), the largest hospital in the Fraser Health Authority area. SMH is enrolled in the Best Beginnings program, which is designed to provide public health services to pregnant women, new mothers, babies, children to age 2 years and their families.

SMH sees more than 5000 pregnancies a year, and the diversity and size of the target population offered us the opportunity to carry out data collection. Seven dental questions from the Canadian Health Measures Survey (CHMS) were adapted for this study. After pilot testing the adapted questions, data were collected using the Best Beginnings form (Appendix 1) and an attached list of 7 dental-focused questions (Appendix 2). These forms were distributed to all women attending prenatal registration at SMH from October 2012 to February 2013.

Those who agreed to participate in the study were advised to detach a cover page as the consent form for their records while filling out the Best Beginnings form and the dental survey. The Best Beginnings forms were kept in the reception area where a public health nurse collected them twice a week. An identification code was assigned to match the dental survey questions with the Best Beginnings form while de-identifying the forms by blacking out all personal information, such as name, address, postal code, care card number, date of birth, telephone number and other contact information. All forms were photocopied before the public health nurse picked them up. All data were numerically coded and entered into a password-protected spreadsheet for statistical analysis.

The 740 respondents provided enough statistical power to tease out the potential relationships and correlations between the different variables presented below.

Variable Construction

Andersen and Newman’s⁶ framework of health service utilization was used to inform the selection of outcome and independent variables. Self-reported oral health and use of dental services during pregnancy were the outcome variables; predisposing, enabling and need factors were independent variables. However, the focus of this paper is on predisposing and enabling factors only. The framework helped to determine the sociodemographic variables that would likely make these women vulnerable to inadequate oral health care during pregnancy.

Predisposing Factors

The predisposing factors in Andersen and Newman’s model comprise variables that represent the tendency to use available health services. In the context of this study, the included variables were age, education (high school education: yes/no), immigration status (born in Canada or elsewhere and number of years living in Canada), refugee status, Aboriginal heritage and smoking status (smoker or non-smoker).⁶

The model emphasizes that both financial and personal enabling factors must be present for most people to use available health services. Therefore, the enabling variables included were having difficulty living within total household income (yes or no), having dental insurance (yes or no), type of insurance, having children (yes or no), needing assistance with transportation (yes or no), feeling depressed during the last month (yes or no) and time of last dental visit.⁶

Vulnerability was assessed from the self-reported answers on the Best Beginnings form, grouped according to the Andersen and Newman model to identify the personal and societal factors that could affect the overall health of the pregnant women and their newborns, including difficulties with living and transportation, immigration status, smoking status, inadequate income and depression. Having 1 or more of these factors was an indicator of vulnerability.

Statistical Analysis

Continuous variables were compared by using two sample t tests, while categorical variables were tested using Fisher’s exact test and the χ² test of association. Multivariable logistic regression was used to estimate the odds ratio and 95% confidence interval (CI) for self-reported oral health and use of dental services. A receiver operating characteristic analysis was conducted to determine which of either the predisposing or enabling factors was most associated with dental service use. Results were considered significant for p < 0.05. Data analysis was carried out using SPSS software, v. 22 (SPSS, Inc., Chicago, Ill.).

Ethical Approval

This study was approved by the Research Ethics Boards of both the Fraser Health Authority and the University of British Columbia.

Results

The study reached 835 registrants over the 4-month period. Of these, 740 (87%) responded fully to the survey. Given that the prenatal registration program is also voluntary, the actual number of pregnant women seen at SMH during this period was probably larger.

Participants included a high proportion of high school graduates (85.2%) who were able to live within their income (85.8%) (Table 1). Few reported recent depression (9.3%) or lack of a social network, i.e., “someone to talk with” (8.1%). Although some of the pregnant women (5.6%) identified themselves as smokers, almost one-quarter (22.7%) were exposed to secondhand smoke.

Surrey has a large immigrant population, with almost 75% born in Asia.⁷ Of the women who participated in the survey, almost two-thirds were immigrants (Table 1), with the largest proportion born in India (50%). In all, 18 birth countries were identified. Few women identified as a refugee (5.5%) or of aboriginal heritage (5.5%). When asked to describe their oral health, most respondents reported it as excellent or good, while 16.7% rated it as fair (14%) or poor (3%) (Table 2).

Almost 50% had seen a dental professional in the last year and 23% reported they visited a dental professional for emergency purposes only. Nine percent of the respondents had never visited a dental professional; of these, almost two-thirds came from India (data not shown). Women who visited a dental professional within the last 2 years were more likely to rate their oral health as excellent or good (89%) than women who hadn’t consulted a dental professional in more than 2 years (72%) (p < 0.001) (data not shown).

Tables 2–4 show the relation of predisposing and enabling factors to self-reported oral health and dental service use, respectively. Of all the predisposing factors, only decreasing exposure to secondhand smoke was associated with better self-reported oral health (Table 2). Women exposed to secondhand smoke on a daily or weekly basis were at least 60% less likely to report their oral health as excellent or good. For enabling factors, financial issues were the dominant limiting factor; lack of dental insurance, type of dental insurance and difficulty living within their income were all associated with fair or poor self-reported oral health (Table 3). Women with dental insurance were 3 times more likely to report their oral health as excellent or good. However, women on government programs or First Nations/Inuit insurance programs were up to 90% less likely to report good oral health. Women who used dental care more recently rated their oral health higher than women who had not seen a dental professional in more than 2 years (Table 3).

Lack of financial means also appeared to limit dental visits (Table 4). Women who reported difficulty living on their income and/or who had no dental insurance reported a greater elapsed time since their last dental visit. Almost 46% of the women had visited a dental professional within the last year for regular care (data not shown).

Multiple logistic regression analysis was used to evaluate the effect of the psychosocial variables mentioned by Anderson and Newman⁶ on self-reported level of oral health and use of dental services (Table 5). All variables that were significant in univariate analysis were included in the multivariate model. The presence of dental insurance significantly predicted both better self-reported oral health (p < 0.001) and receiving dental care within the previous 2 years (p < 0.001). To determine which predisposing or enabling factors were most associated with dental service use, a receiver operating characteristic curve was plotted (Fig. 1). The area under the curve is greater for enabling factors (72.5%) versus predisposing factors (63.6%) and perceived need of care (69.9%).