Abstract
Introduction
Oral health is considered to be one of the significant factors that determine the general health and well-being of all populations (Bloom, Simile, Adams, & Cohen, 2012; Centers for Disease Control and Prevention [CDC], 2011; Ebersole, D’Souza, Gordon, & Fox, 2012; Institute of Medicine [IOM], 2011; U.S. Department of Health and Human Services [DHHS], 2000). Its importance is particularly pronounced in later years of life, because older individuals are particularly susceptible to oral health problems (Bloom et al., 2012; Dolan, Atchison, & Huynh, 2005; U.S. DHHS, 2000). Multiple medications that older adults take for their comorbid medical conditions often reduce salivary flow and cause dry mouth, which in turn puts them at a higher risk for cavities and soft tissue problems (Bloom et al., 2012; U.S. DHHS, 2000). Oral health problems not only limit older individuals’ basic ability to bite, chew, and swallow foods but also adversely affect their susceptibility to other medical conditions, as well as their self-image, social interactions, quality of life, and even mortality (Bloom et al., 2012; CDC, 2011; Ebersole et al., 2012; IOM, 2011; U.S. DHHS, 2000).
According to the Medical Expenditure Panel Survey, more than 2.7 million older Americans (of the civilian non-institutionalized population) were unable to obtain or experienced delays in getting needed dental care in 2007; this figure is higher than those reported for medical care and prescribed medicines (Chevarley, 2010). In particular, older ethnic immigrants are likely to have limited access to dental care (Kiyak & Reichmuth, 2005; Shelley, Russell, Parikh, & Fahs, 2011). This disparity in access is a cause for concern because the oral health of older minorities is poorer than that of age-comparable non-Hispanic Whites and of younger adults in the same ethnic group (Kiyak & Reichmuth, 2005; Shelley et al., 2011).
Andersen’s behavioral health model (Andersen & Newman, 1973) has been widely used to explain the utilization of various health services, including dental care (Andersen & Davidson, 1997). The model features predisposing, enabling, and need components. Predisposing factors include demographic characteristics that determine the propensity to utilize health care services. Health-related concerns and problems, in both objective and subjective form, represent the need for health care services. Objective indicators of dental care needs often include the results of clinical examination (e.g., the number of decayed, missing, and filled teeth); such clinical measures are known to be highly correlated with subjective ratings of oral health (Jones et al., 2001; Wu et al., 2011).
Enabling factors encompass a variety of resources that facilitate or impede individuals’ use of health care services. Health insurance coverage is a critical enabler for all types of health services across all age groups (Lillie-Blanton & Hoffman, 2005). However, the lack of dental care in the elderly population represents a particular burden, because many older individuals lose their employment-based dental insurance at retirement, and comprehensive dental coverage is not offered by social insurance programs (Bloom et al., 2012; Chevarley, 2010). For example, Medicare does not cover routine dental care, and Medicaid only offers limited dental coverage for eligible individuals in some states (Centers for Medicare and Medicaid Services, 2014; McGinn-Shapiro, 2008).
It has been recommended that when Andersen’s (1995) model is applied to racial/ethnic minorities, researchers should consider the sociocultural characteristics and life circumstances unique to the population. Examples of variables relevant to immigrant populations include the length of time since immigration, acculturation, and English proficiency. Because of their high correlations and potential overlap, the conceptualization of immigration-related variables requires special attention. The length of time since immigration is generally positively correlated with English proficiency and the individuals’ overall level of acculturation. However, language acquisition and cultural adaptation are not necessarily the products of time, and they reflect a wide range of individual variation. Acculturation refers to the process of cultural adaptation to a host society (Berry, 2002). Although proficiency in the host language is its key element, acculturation also includes knowledge of appropriate behavior in different contexts, use of media, friendship patterns, and many other indications of an individual’s ability to get along in the host culture (Berry, 2002; Chiriboga, 2004). Individuals with high levels of acculturation tend to have better access to resources and benefits (Berry, 2002; Chiriboga, 2004), including health services (Derose, Escarce, & Lurie, 2007; Ponce, Hays, & Cunningham, 2006). Having a family network is also known to be an enabler for health service use in older immigrants (Diwan, 2008; Leclere, Jensen, & Biddlecom, 1994). It is anticipated that older immigrants who are acculturated and/or have strong family ties may be able to use dental care services when needed, whereas those who lack such resources are likely to face barriers.
The goal of the present study was to identify factors associated with dental care utilization and unmet dental needs. In the present assessment, any dental visit, regardless of the nature of the visit (preventive care or treatment) or the source of concern (teeth or gums), was considered dental care utilization. Unmet dental need was defined as the experience of being in need of dental care but unable to receive service. Older Korean Americans were the target population in this study. Koreans are the fifth largest Asian American subgroup (National Asian Pacific Center on Aging, 2013), and their unfavorable physical and mental health status and underutilization of health services have been well documented. For example, the rate of probable depression among older Korean Americans (prevalence estimates in the middle-to-upper 30% range) is particularly high when compared with the 9% to 16% found in community-dwelling older Whites and African Americans (Jang, Chiriboga, Kim, & Phillips, 2008). Korean Americans are also highly likely to be uninsured and to lack a usual place for health care (Barnes, Adams, & Powell-Griner, 2008). However, no information is available pertaining to their oral health status and dental care utilization.
Following the general structure of the Andersen’s model (Andersen & Newman, 1973), the current study aimed to explore the contributions of predisposing (age, gender, marital status, living arrangement, education, length of stay in the United States, and chronic medical conditions), enabling (dental insurance, acculturation, and family network), and need (self-rated oral health) variables to dental care utilization and unmet dental needs in older Korean Americans.
Method
Participants
After obtaining approval from the university’s Institutional Review Board, we conducted a survey of older Korean Americans (aged ≥60) during the spring of 2013 in Central Texas. Because immigrant populations are often hard to identify by any single approach, we used a variety of sources for recruitment. The research team contacted potential sources of older Korean Americans and arranged for surveys to be conducted. The sites and organizations contacted include 11 Korean churches, 2 other religious groups, 1 senior educational center, 4 meetings of the Korean Elderly Association, and 3 social clubs. Efforts were made to reach less mobile or socially isolated individuals by soliciting active referrals from a variety of sources.
The survey instrument consisted of a standardized questionnaire in Korean, developed through a back-translation and reconciliation method. Although the survey was designed to be self-administered, trained interviewers were available for anyone who needed assistance. Data collection was conducted in locations convenient to the participants, such as meeting rooms and cafeterias in churches and community centers. Respondents were paid US$10 for their participation. A total of 209 individuals participated in the survey. All participants were included in the analysis because none of their responses was missing more than 5% of the required information.
Measures
Outcome variables
Dental care utilization was measured by asking participants how many dental visits they had had in the past year. No restrictions were given in terms of the nature of the visit (preventive care or treatment) or the source of concern (teeth or gums).
Unmet dental need was assessed with a single item asking participants whether they had needed dental care but could not receive service at any time within the past year. An open-ended question was also asked to determine the main reason for the unmet needs.
Predisposing variables
Demographic information included age (in years), gender (0 = male, 1 = female), marital status (0 = married, 1 = unmarried), living arrangement (0 = living with others, 1 = living alone), education (0 = <high school graduation, 1 = ≥high school graduation), and length of stay in the United States (in years). In addition, the total number of chronic medical conditions was assessed using a checklist of nine diseases and conditions common in older populations (e.g., diabetes, cancer, arthritis, heart disease, and high blood pressure).
Enabling variables
For dental insurance coverage, participants were asked whether they had insurance that covered the cost of any dental visit. Acculturation was assessed with a 12-item acculturation inventory (Jang, Kim, Chiriboga, & King-Kallimanis, 2007). The inventory included two items for each of six domains (language use, media consumption, food consumption, social relations, sense of belonging, and familiarity with culture). The total score could range from 0 to 36, with higher scores indicating a greater level of acculturation to mainstream American culture. Internal consistency of the scale in the present sample was high (α = .93).
Family network was measured with three items from the Lubben’s Social Network Scale (Lubben, 1988). The items consisted of the number of family/relatives seen at least once a month (0 = “none” to 5 = “9 or more”), frequency of contact (0 = “less than monthly” to 5 = “daily”), and the number of family/relatives the participant felt close to (0 = “none” to 5 = “9 or more”). Internal consistency of the scale in the present sample was satisfactory (α = .75).
Need
Self-rated oral health was used as an indicator of dental care need. Participants were asked how they would rate their overall oral health status on a four-point scale: “excellent,” “good,” “fair,” or “poor.” Responses were coded into a binary variable of “excellent/good” (0) and “fair/poor” (1). This single item, validated as an efficient indicator of overall oral health status, is known to be highly correlated with multiple-item subjective oral health measures and the results of clinical examination (Jones et al., 2001; Wu et al., 2011).
Analytic Strategy
Descriptive and bivariate analyses were conducted to understand the characteristics of the sample and the underlying associations among study variables. For the count variable (the number of dental visits in the past year), Poisson regression analysis was used. Poisson regression is an optimal method for estimating count data with a high level of skewness (Cohen, Cohen, West, & Aiken, 2003). For the binary outcome (unmet dental need), logistic regression analysis was used. In both models, sets of predictors were entered in the following sequence: (a) predisposing variables, (b) enabling variables, and (c) needs. The number of dental visits was entered as the fourth and final step for the model of unmet dental need. Analyses were performed using IBM SPSS Statistics 21.
Results
Descriptive Information
Table 1 summarizes descriptive information concerning the sample and study variables. The sample was composed of 209 participants, aged 60 to 95, with an average age of 69.6 (SD = 7.50). More than half of the participants were females (64.4%). More than a quarter (26.4%) were unmarried, and 12% were living alone. Because of the high correlation between marital status and living arrangement (r = .58, p < .001), a composite variable (1 = not married and living alone, 0 = others) was created and used for analysis.
Descriptive Characteristics of the Sample (N = 209).
Approximately 60% had received at least a high school education. All participants were foreign-born immigrants, and the average length of stay in the United States was 26.9 years (SD = 10.8). The number of chronic medical conditions averaged 1.20 (SD = 1.15). The overall demographic characteristics of the sample were similar to those observed in samples of older Korean Americans in California, New York, Maryland, and Florida (e.g., Han, Kim, Lee, Pistulka, & Kim, 2007; Jang & Chiriboga, 2010; Kim et al., 2010; Roh et al., 2011).
Approximately 19% of the sample reported that they had insurance that covered the cost of any dental visit. The scores for acculturation averaged 10.9 (SD = 7.36), out of a potential range of 0 to 36, indicating a relatively low level of acculturation. For instance, 78% of the sample rated their English proficiency as less than “very well.” Family network scores averaged 8.35 (SD = 2.73), with a range of 0 to 15.
More than 61% of the sample rated their oral health status as “fair” or “poor.” The reported number of dental visits in the past year ranged from 0 to 10, and the score distribution was non-normal (skewness = 1.58, standard error [SE] = 0.17). The average number of dental visits was 1.68 (SD = 1.64). More than 44% reported an unmet dental need (the experience of being in need of dental care but unable to receive service) in the past year. In the open-ended responses, a vast majority (98%) reported the lack of dental insurance or financial resource as a major reason for the unmet need. Some participants also mentioned issues associated with transportation, difficulty with scheduling an appointment, and lack of time.
Predictive Models of Dental Care Utilization and Unmet Needs
Prior to analyzing multivariate models, we assessed bivariate correlations among the study variables. The variables were correlated in the expected directions, and no concern about collinearity was identified. The highest coefficient was found in the relationship between education and acculturation (r = .43, p < .001). The coefficient for the relationship between the length of stay in the United States and acculturation (r = .35, p < .001) was at a moderate level.
Table 2 summarizes the results of the multivariate analyses. When applied to the count of dental visits, the Poisson regression model did not generate a significant effect for the predisposing variables. In the subsequent step with the enabling variables, family network emerged as a significant factor. A higher number of dental visits was predicted by a stronger family network. In the final model, fair/poor self-ratings of oral health were shown to increase the number of dental visits.
Regression Models of Dental Care Utilization and Unmet Dental Needs.
Note. The number of dental visits was analyzed by Poisson regression analysis, and unmet dental needs by logistic regression analysis. In each step, the previous blocks were controlled. B = unstandardized coefficient; SE = standard error; OR = odds ratio; CI = confidence interval.
p < .05. **p < .01. ***p < .001.
In the initial logistic model of unmet dental needs, education and length of stay in the United States were statistically significant. The odds of having an unmet need increased when individuals had less than a high school education and had stayed in the United States for a shorter period of time. The subsequent step revealed a significant contribution of all three of the enabling variables that were included. The lack of dental insurance coverage, limited acculturation, and fewer family ties all increased the likelihood of an unmet dental need. In the subsequent models, both the presence of need (fair/poor self-ratings of oral health) and lower levels of dental care utilization (fewer numbers of dental visits) independently contributed to the odds of having an unmet need.
Discussion
Responding to the recognized need to address oral health disparities in older ethnic immigrants (Bloom et al., 2012; Dolan et al., 2005; IOM, 2011; U.S. DHHS, 2000), the present study identified predictors of dental care utilization and unmet dental needs in older Korean Americans. Using Andersen’s behavioral health model (Andersen & Newman, 1973), attention was given to the predictive roles of predisposing (age, gender, marital status, living arrangement, education, length of stay in the United States, and chronic medical conditions), enabling (dental insurance, acculturation, and family network), and need (self-rated oral health) variables.
As anticipated, the present sample of older Korean Americans demonstrated a poor status with respect to oral health, utilization of dental services, and insurance coverage. More than 61% of the sample rated their oral health status as “fair” or “poor.” The average number of dental visits in the past year was 1.68 (SD = 1.64); however, more than 44% of the participants reported that they had needed dental care but could not receive service. Cost and lack of dental insurance coverage were most frequently reported as a major reason for an unmet need, as has been shown in previous studies using national data (e.g., Chevarley, 2010; Qiu & Ni, 2003).
A substantial proportion (81.2%) of the sample had no dental insurance coverage; this figure is higher than the 66% reported for a national sample of older adults (Wu et al., 2011). Although the specific type of dental insurance was not assessed in the present study, about 17% of the participants were Medicaid enrollees. Despite the fact that Texas is one of the few states that offers comprehensive dental care coverage for elderly Medicaid enrollees (Centers for Medicare and Medicaid Services, 2014), 80% of our participants with Medicaid reported no dental coverage. This finding reveals a serious gap in Medicaid enrollees’ knowledge about their dental coverage. A similar finding has been reported for older Chinese immigrants in New York, another exceptional state in which Medicaid includes full coverage for adult oral health care (Shelley et al., 2011). These findings call attention to the urgent need for systematic efforts to educate new immigrant groups of older adults about their benefits and coverage. Such knowledge is likely to promote their utilization of dental services for both treatment and prevention.
Although the personal level of resources (dental insurance and acculturation) had no impact on dental service use in our study, family network was found to be a critical enabler. This finding is in line with literature showing the importance of family support and networks in older ethnic minorities’ pursuit of health care (Diwan, 2008; Leclere et al., 1994) and suggests that interventions to promote dental care use should incorporate family components. After controlling for the predisposing and enabling variables, we found that the perceived need for dental care need, as indicated by a fair/poor self-rating of oral health, was a driving force for the use of services.
In the predictive model of unmet dental need, all three enabling variables turned out to be significant. Not only the lack of dental insurance but also the two enablers viewed as having special relevance to immigrant populations—limited levels of acculturation and family network—increased the odds of having an unmet dental need. These findings move beyond the conventional service barrier of a lack of health insurance by documenting the importance of social and cultural factors for dental health care. In concordance with the findings from other studies of ethnic immigrants (Derose et al., 2007; Diwan, 2008; Leclere et al., 1994; Ponce et al., 2006), our study highlights the vulnerability of individuals who are culturally and linguistically isolated and lack family resources.
Some limitations to the present study should be noted. The use of a regionally defined convenience sample and a cross-sectional design potentially limits the generalizability of the findings and causal inferences. Although the estimated models considered multiple key factors, a substantial proportion of the variance remains unexplained. Future efforts should incorporate a broader range of variables, including oral health-related knowledge, beliefs, and behaviors as well as objective oral health indicators from clinical assessment. It would also be helpful to consider the function of different types of support (e.g., instrumental, informational, and emotional support) and satisfaction with dental services received. Furthermore, consideration should be given to system-level variables such as local dental service resources and particularly the availability of culturally and linguistically competent dental care professionals.
Despite these limitations, the present study contributes to improving knowledge on dental care in the target population. The enabling role of dental insurance and sociocultural factors needs to be incorporated into interventions designed to bridge the gap between dental need and services. National policies, as reflected in both the Affordable Care Act (ACA) and the DHHS Office of Minority Health’s (OMH; 2013) Culturally and Linguistically Appropriate Services (CLAS) standards, have great potential to improve the oral health and utilization of services among older Korean immigrants. However, more remains to be accomplished. Future efforts should include financial assistance programs for those ineligible for federal health insurance programs, interpreter services, and educational outreach for these older adults and their families.
Footnotes
Declaration of Conflicting Interests
The authors declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.
Funding
The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: This work was supported by the St. David’s Center for Health Promotion and Disease Prevention Research (CHPR) Pilot Grant Program (30-2142-4351, principal investigator [PI]—Yuri Jang, PhD).