Income and wealth as correlates of socioeconomic disparity in dentist visits among adults aged 20 years and over in the United States, 2011–2014

Despite general improvements in oral health status due to advancement in social and living conditions, oral diseases remain among the most prevalent human diseases globally and a major public health problem. The 2015 Global Burden of Disease Study estimated that oral conditions (untreated dental caries, chronic periodontitis, and edentulism) ranked among the top ten health conditions, affecting 3.5 billion people worldwide [1]. The profile of oral diseases is not homogeneous between or within countries; the burden is substantially higher among poorer and disadvantaged populations in both high- and low-income countries, including the United States (US) [2–4]. In the US, people of lower socioeconomic position have a higher burden of oral diseases compared with those who are socioeconomically better off, and these disparities also apply to issues of access to oral health services. A study among US adults reported that people living in poverty and those with the least education had fewer dentist visits compared with more affluent and educated individuals [5].

Having a dentist visit in the past 12 months is one of the 17 Healthy People (HP) 2020 oral health objectives in the US [6]. The HP 2020 initiative contains health promotion and disease prevention national goals and objectives set by the US Department of Health and Human Services (HHS) for improving the health of all Americans. There are 12 Leading Health Indicators in HP 2020, and oral health is represented by the objective “to increase the proportion of children, adolescents, and adults who used the oral health care system in the past year.” Utilizing dental care in the past 12 months is correlated to higher levels of dental health satisfaction and overall quality of life [7]. Furthermore, differences in access to dental care still exist and a major reason for not having a dentist visit is financial circumstances [8]. For example, dental care utilization in the past 12 months among poor adults in the US was around 20% compared to approximately 50% among high income adults in 2014 [9].

Most studies in the US have used income [10, 11] and education [5, 12] as socioeconomic indicators to assess the association between lower socioeconomic position and lower rates of dentist visits, but there is limited information on other socioeconomic indicators, such as wealth, which measures accumulated assets rather than income alone. Yet there are arguments that health studies should include wealth as a socioeconomic indicator [13]. The wealth index has been widely used as a proxy of socioeconomic position in studies conducted in low- and middle-income settings. Based on methodology developed by Filmer and Pritchett [14], the index combines data on durable assets (car, refrigerator, and television), housing characteristics (dwelling floor and roof material), and access to services (drinking water source and electricity supply) and uses Principal Component Analysis to generate weights for household assets [14, 15]. However, in the US setting, where most households have durable assets and access to electricity and water supplies, it may be more appropriate to measure wealth differently, by combining income and assets such as cars, homes, savings, and stocks.

Even though income and wealth are positively correlated, they measure different things. For example, older adults may have little income but may have accumulated substantial wealth, through a combination of a lifetime of work and/or inheritance from ancestors. Recent immigrants and racial minorities, even those with high incomes, may be less likely to have significant familial or inherited wealth [16]. Income and occupation status among retired people lose their significance as measures of socioeconomic position and wealth becomes more important [17, 18]. Another example of the difference between income and wealth has been shown in studies using home ownership as a measure of wealth. Home owners may have more income to spend on health care than non-home owners because non-home owners still have to make rent payments [19]. Most recently, wealth rather than income was reported to be more sensitive as an indicator of socioeconomic disparity in health-related outcomes [13]. However, the best choice of socioeconomic measure will depend upon the purpose of the analysis and the policy context.

The purpose of this study is to improve understanding of income and wealth as correlates of socioeconomic disparity in dentist visits among adults in the US so that policy makers may be better informed about targeting interventions to improve access to dental care. The objective is to assess how two socioeconomic status measures, income and wealth, compare as correlates of socioeconomic disparity in dentist visits among adults in the US.

Table 1 also shows the prevalence of not having a dentist visit in the past 12 months among adults aged 20 years and over by sociodemographic characteristics. The overall prevalence was 39%, with a significant difference between participants who did or did not have a dentist visit. This difference held true for all groups except for the non-Hispanic Asian and middle income groups. An income gradient was observed whereby participants in the poorest group had the highest prevalence (58%) and those in the highest income group had the lowest prevalence (24%) of not having a dentist visit. A wealth gradient was also observed. The prevalence was highest in the lowest wealth group (57%) and lowest in the highest wealth group (25%). Not having a dentist visit was more prevalent among adults aged 20–44 years, males, Hispanics, those with the least education, current smokers, those who reported poor oral health, and those with the presence of untreated dental caries.

In the parsimonious income model, adults in the poor group were close to twice as likely to not have a dentist visit (RR = 1.69; 95% CI: 1.51–1.90) and those aged 20–44 years were 16% more likely to not have a dentist visit than those aged 65 years and over (RR = 1.16; 95% CI: 1.04–1.30). In the parsimonious wealth model, adults in the low wealth group were close to twice as likely to not have a dentist visit compared with those in the high wealth group (RR = 1.68; 95% CI: 1.52–1.85) and unlike the income model, there were no significant differences within age subgroups in the multivariable wealth model. This result shows that age and income were independently associated with not having dentist visits but the age association attenuated to non-significance in the presence of wealth.

The effect of being in either the Total Hispanic or Non-Hispanic Black group attenuated from the unadjusted to the parsimonious income and wealth models. This means that relative to being in the Non-Hispanic White group, the likelihood of those in either the Total Hispanic or Non-Hispanic Black group having had no dentist visits decreased in the presence of other socio-demographic factors. In contrast, there was positive attenuation for the Asian group, relative to the Non-Hispanic White group, for both the income and wealth models. The parsimonious models showed increased and significant associations between being in the Asian group and having no dentist visits which suggests possible offsetting sociodemographic effects in the unadjusted models. Overall, the risk indicators in both models were: being male; Hispanic; non-Hispanic black; widowed or separated; least educated; a current smoker; not satisfied with oral health; and having untreated caries. The income and wealth models were compared using the AIC. The AIC values for the parsimonious income and wealth models were 10,829 and 10,895 respectively. These results indicate that the income models are marginally better than the wealth models (the smaller the AIC value, the better the model).

The analyses were stratified by race/ethnicity because of significant interactions. Figure 1 uses the parsimonious models to summarize the association (RRs). The figure presents income, wealth and other factors as binary variables to compare between the highest and lowest groups. In the income and wealth models, the poor and the low wealth respondents were more likely to not have a dentist visit in all four race/ethnic groups compared to the least poor and the high wealth groups. In the income model, age was a significant factor. Compared to participants aged 65 years and over, the non-Hispanic white and non-Hispanic Asian adults aged 20–44 years were more likely to not have a dentist visit, while the same association was not significant among non-Hispanic black and Hispanics. In the wealth model, age was not a significant factor in all race/ethnic groups.

This study is the first to describe both income and wealth as correlates of socioeconomic disparity in dentist visits among adults using NHANES data. Important differences were observed (especially in age groups) when each measurement was used. Age was not significant for wealth but significant for income. There were also significant interactions by race/ethnicity for income and wealth.

The overall prevalence of not having a dentist visit in the past 12 months among adults aged 20 years and over was 39%. This prevalence is consistent with other findings. For example, a report released in 2016 by the National Center for Health Statistics (NCHS) presented the proportion of adults aged 18 years and over without a dentist visit in 2014 at 38% [28]. However, studies using the Medical Expenditure Panel Survey (MEPS) data have reported higher percentages: 65% of adults aged 18–64 years did not see a dentist in 2013 [29]. Study design, sampling frame, reference periods, lead-in statements, question wording, and social desirability bias are some of the reasons for these different estimates [12].

Despite differences in these overall estimates of dentist visits, demographic and socioeconomic indicators such as sex, race/ethnicity and socioeconomic position have shown consistent associations across national US surveys. For example, persons in lower socioeconomic positions (commonly measured via income status and education) were significantly more likely than those in higher socioeconomic positions to not have a dentist visit in the past 12 months, as were non-Hispanic blacks; Hispanics were significantly more likely than non-Hispanic whites [12]. Similar findings were found in our study where participants in lower socioeconomic positions (measured by income status and wealth) were consistently more likely than those in higher socioeconomic positions to not have a dentist visit in the past 12 months. Other studies have also shown the same pattern in different contexts. A study among adults aged 50 years and over in 14 European countries reported higher rates of dental services utilization among the high-income group compared with the low-income group [30]. Income and education have also been reported as factors in not seeking oral health services in many other global contexts [31, 32].

Explanations for observed socioeconomic disparities have been put forward in the literature. The Commission on Social Determinants of Health (CSDH) produced a landmark report on the impact of social determinants of health (via unfair economic arrangements and poor social policies and programs) on the unequal distribution of health experiences, where health and illness invariably follow a social gradient [33]. Bartley [34] has also reviewed the four theories proposed to lie behind inequality in health and access to health care, namely behavioral, psychosocial, material, and life-course approaches [34]. Importantly, even well-defined material and neo-material resources (income, living conditions) and health behaviors (smoking, diet) do not fully explain health inequality, let alone the social gradient, which may be much more difficult to characterize. The persistent health inequality observed at all levels of the social gradient may be explained by the psychosocial model and the concept of buffering resources such as social capital, social support, social relations, social participation, and self-efficacy. Whatever the case, health inequalities are affected by a complex interaction of these pathways during a person’s life-course depending on the context and the population [34].

Our study shows that both income and wealth have strong associations with dentist visits, similar to findings in previous studies. For example, a study among US adults aged 51 years and older in 2008 showed the separate effects of income and wealth on dental utilization [35]. Likewise, a recent study among Japanese adults aged 50 to 75 years showed that both wealth-related and income-related inequalities in dental care use existed, with greater impact shown by wealth [36]. Our findings also show that results differ when measurements of income or wealth are used. For instance, age was significant in the income model but was not significant in the wealth model. Possible explanations for the observed differences may be that wealth reflects net accumulation of advantage and disadvantage over the life course while income reflects the direct and immediate impacts of a lack of resources. Moreover, without adjusting for wealth, participants aged 20–44 years were more significantly likely to not have a dentist visit but after adjusting for wealth and other sociodemographic factors the association was not significant. One possible explanation is that there are other cultural factors impacting on dental utilization.

There are some limitations. This is a cross-sectional study and causation cannot be inferred. We show that income and wealth are associated with dental visits. Furthermore, the study analyzed self-reported data, which might be subject to information and social desirability biases. For example, while respondents may report poor oral health, they could be misclassified based on clinical examination, or respondents might be more prone to provide socially acceptable responses or unable to correctly recall an event or to accurately evaluate their caries severity. Our outcome variable – not having a dentist visit in the past 12 months – may also capture a heterogeneous group of people who had various reasons for their answers to the question “about how long is it since you last visited a dentist?” Some people may have poor oral health because they had not visited the dentist in the past two or 3 years, whereas others may not have visited the dentist in the past 4 years, but have good oral health. We acknowledge that the outcome variable is not comprehensive but we are confident of its relevance, given that in the Healthy People 2020 objectives, one of the oral health objectives (OH-7) is to increase the proportion of children, adolescents, and adults who used the oral health care system in the past 12 months [6].

This is the first study of its kind to analyze wealth as a socioeconomic variable using NHANES data. We used two indicators - monthly family income and home ownership - to construct a wealth variable. Nevertheless, the results of our study should be interpreted with caution because capturing an individual’s wealth is a complex undertaking. One common method of measuring wealth is using Principal Component Analysis to generate weights indicative of household wealth. For this method to work, it requires gathering large amounts of information from respondents, such as ownership of material goods, savings, consumption of goods and services, etc. Unfortunately, due to data limitations we were not able to derive a wealth index using this method, but we think our approach may open doors for future surveys to incorporate more questions regarding household assets to create a wealth index that can give more precise measures of economic well-being. Lastly, we are unable to account for the role of environmental and psychosocial factors such as social capital, social support, geo-locality (urban/rural residence), as well as dental insurance and out of pocket payments. Studies have shown the impact of these determinants in oral health services utilization [37–39] but unfortunately, information on such factors was not collected in the NHANES.

In the US, adults in lower socioeconomic positions (low wealth or income) are less likely to have annual dentist visits yet the socioeconomic patterning varies by age, race/ethnicity and other factors. This study showed that income may be a better measure of socioeconomic disparity than wealth, although wealth may well be a more suitable socioeconomic measure in older adult populations. There is clearly a need for further research into ways of more precisely measuring socioeconomic disparities in dentist visits in the US. Importantly, these findings strengthen the call to action regarding policy based on socioeconomic inequalities in oral health.