Social support and dental visits

Abstract

Background.

The authors conducted a study to examine the influence of social support on dental visits among an adult population.

Methods.

Using 2003–2004 National Health and Nutrition Examination Survey data, the authors analyzed information pertaining to adults 40 years and older (N = 2,598) (with the exclusion of edentulous people), who represent about 108 million people in the United States. They weighted logistic regression models for dental visits, while controlling for demographic characteristics (age, race/ethnicity, sex, education, insurance, income), socioenvironmental characteristics (marital status, emotional and financial support, number of close family members and friends, years lived in the neighborhood) and physical and mental health status.

Results.

The study findings show that the odds of having had a timely dental visit (that is, within the preceding year), a self-care–related dental visit (that is, a visit initiated by the patient for a checkup, examination or cleaning more than one year previously but less than three years previously) or both were decreased significantly by not having had any financial help if needed and by having fewer close family members and friends. The authors did not find any association between marital status, emotional support or years lived in the neighborhood and having had a timely or self-care–related dental visit.

Conclusions and Clinical Implications.

Timely or self-care–related dental visits depend in part on financial support and the number of one’s close friends and family members. Clinicians should engage appropriate members of the patient’s social network to facilitate dental visits.

Former U.S. Surgeon General Dr. C. Everett Koop stated, “You are not a healthy person unless you have good oral health.”¹ Oral health is related to sustained and proactive health behavior, notably dental visits that are scheduled regularly and related to self-care.

In a study comparing perceptions of self-care among socioeconomically vulnerable versus nonvulnerable older adults, Clark and colleagues² found that keeping medical appointments was of primary importance to participants in the vulnerable group in their perception of self-care. To understand factors that contribute to timely dental visits, Osterberg and colleagues³ conducted a study among a sample of elderly Swedish people, the results of which showed that functional ability and general health were not as important as were socioeconomic, lifestyle and social support factors. In a study of people 65 years or older conducted in the United Kingdom, McGrath and Bedi⁴ used “living alone” as an indicator of social support and found that this was an important predictor of a participant’s reason for the last dental visit. Hanson and colleagues⁵ and Rickardsson and Hanson⁶ measured several aspects of social support and found various associations with dental care utilization. In an intervention study testing the association between four types of social support and dental care utilization among children of Latina immigrants in North Carolina, Nahouraii and colleagues⁷ found that some types of social support were associated with dental care visits.

Researchers generally consider social support to derive from social networks. Berkman⁸ defined it as “the emotional and instrumental assistance that is obtained from people who compose the individual’s social network.” A common explanation for the link between health and social support is that strong social ties provide a buffering effect from stress, reducing the vulnerability resulting from stress-related health problems, facilitating adaptation and speeding recovery, as well as encouraging health promotion activities.^8,9 In a study examining social support (that is, presence of a partner) and self-care among patients who experienced heart failure, Gallagher and colleagues¹⁰ found that patients with a high level of support reported significantly better self-care than did those with low or moderate levels of support.

Although the size of one’s social network is important, other aspects of supportive relationships, such as frequency of contact and type of available support, may be more important for dental visits. Preventing oral disease involves engaging in personal health practices, including timely dental visits. We conducted this study to examine the influence of social support on timely dental visits (that is, within the preceding year), self-care–related dental visits (that is, initiated by the patient not by the dentist) or both among an adult population, while controlling for sociodemographic and physical health characteristics.

METHODS

Sampling and data collection.

A number of publicly available national surveys contain oral health information. The nationally representative National Health and Nutrition Examination Survey (NHANES) is designed to assess the health and nutritional status of adults and children in the United States.¹¹ The survey involves the use of a stratified, multistage probability sampling design of the civilian noninstitutionalized U.S. population, with oversampling of low-income people, African Americans, Hispanics, people aged 12 through 19 years, and people 60 years and older. The NHANES offers comprehensive dental and oral health data sets, with both self-reported and clinical examination measures. An ongoing survey since 1999, NHANES has had a substantial history of collecting oral health data and has matured through four waves, with its results released in two-year waves. We used data from the 2003–2004 NHANES,¹¹ which offered the best available, although limited, information related to dental visits and social support. For NHANES 2003–2004, investigators selected 12,761 people for the sample; they interviewed 10,122 of these participants (79.3 percent) and examined 9,643 participants (75.6 percent) in a mobile examination center.¹¹

Inclusion criteria.

We included only adults 40 years and older whose interviews included social support and oral self-care–related questions (n = 3,008), representing 120,455,464 people in the U.S. population. In the analysis and modeling of this study, we excluded those who were edentulous, resulting in a final sample size of 2,598, representing 107,569,688 people in the United States.

Dependent variable.

We created the self-care proxy variable by combining two questions from the survey. The first question asked participants to specify about how long it had been since they had last visited a dentist. Respondents were instructed to include all types of dentists, such as orthodontists, oral surgeons and all other dental specialists, as well as dental hygienists. The second question instructed participants to indicate the main reason for their last visit to the dentist. We dichotomized responses by using a binary measure. We recorded a code of “1” for patients who had visited the dentist for a checkup, an examination or a cleaning within the preceding year, or who had visited the dentist on their own for a checkup, an examination or a cleaning more than one year previously but less than three years previously; we considered these patients to have had a timely dental visit, a self-care–related dental visit or both. We coded as “0” all other reasons for the last dental visit, including the patient’s feeling that something was wrong, bothering him or her or causing pain, as well as treatment of a condition that a dentist discovered at an earlier checkup or examination.

Covariates.

NHANES investigators collected all covariate data from study participants in the home interview. We selected specific covariates for their known association with dental visits, on the basis of results from previous studies. These covariates include sociodemographic variables (age, ethnicity, sex, education, insurance, income), socioenvironmental variables (marital status, emotional and financial support, number of close family members and friends, years in the neighborhood) and health status variables (number of days of inactivity due to poor physical or mental health and general health status).

We used five NHANES variables to measure social support, as it was the main focus of this study:

• What is your current marital status?

• Can you count on anyone to provide you with emotional support such as talking over problems or helping you make a difficult decision?

• If you need some extra help financially, could you count on anyone to help you, for example, by paying any bills, housing costs or hospital visits or providing you with food or clothes?

• In general, how many close friends do you have? By “close friends” I mean relatives or nonrelatives with whom you feel at ease, can talk to about private matters and can call on for help.

• How many years have you lived in the neighborhood?

Data and statistical analysis.

We used nationally representative oral health data from NHANES 2003–2004 for our analyses.¹¹ This wave of NHANES data includes a total of 10,122 participants older than 20 years, 3,008 of whom were 40 years or older and answered both self-care–related questions and social support questions. We excluded those who were edentulous, resulting in a final sample size of 2,598. The measures used in our analysis included demographics, number of years living in the neighborhood, emotional and financial support, and health behaviors. The dependent variable measured the nature of dental visits as timely, self-care–related or both (dental visit = 1) and other types of dental visits (dental visit = 0).

We analyzed the data at three levels. First, through univariate analysis, we calculated the marginal distribution of each of the outcome measures, predictors and covariates. For the continuous variables, we calculated the weighted mean, standard deviation and minimum and maximum range. For categorical variables, we calculated the weighted frequency distributions, modes and percentages in the population. We used a weighted bivariate analysis to examine the association between each of the continuous or categorical outcome measures and each of the predictors and covariates. Finally, we built weighted logistic regression models for the binary categorical outcomes, adjusted for demographic characteristics such as age group, race/ethnicity, sex, education and income. We used statistical software (SAS Version 9.2, SAS Institute, Cary, N.C.) for all analyses.

RESULTS

Forty-nine percent of participants reported having had a timely dental visit, a self-care–related visit or both (Table 1, page 189). As expected, all covariates indicated in the literature to be associated significantly with dental visits were statistically significant at the bivariate level (Table 2, pages 190–191). In a logistic regression model including all statistically significant relationships at the bivariate level, the odds of having had a timely dental visit, a self-care–related visit or both were predicted (Table 3).

TABLE 1.

Weighted frequency of having had a dental visit.^*

DENTAL VISIT	WEIGHTED FREQUENCY	STANDARD DEVIATION OF WEIGHTED FREQUENCY	PERCENTAGE	STANDARD ERROR OF PERCENTAGE
Yes	52,575,294	4,696,012	48.88	1.9526
No	54,994,394	4,170,712	51.12	1.9526
TOTAL	107,569,688	7,841,482	100	—†

^*Dental visit defined as timely, self-care related or both. Source: Centers for Disease Control and Prevention.¹¹

^†Not applicable.

TABLE 2.

Bivariate analysis of dental visits.^*

COVARIATE	DENTAL VISIT		NO DENTAL VISIT		TOTAL WEIGHTED FREQUENCY, IN MILLIONS
	Weighted Frequency, in Millions	% of Participants	Weighted Frequency, in Millions	% of Participants
Age, Years
< 54	29.64	48.10	31.99	51.90	61.63
55–64	11.17	50.39	11.00	49.61	22.17
65–74	7.36	48.87	7.70	51.13	15.07
≥ 75	4.40	50.52	4.31	49.48	8.70
TOTAL	52.58	48.88	54.99	51.12	107.57
Race/Ethnicity †
Hispanic	3.05	34.98	5.68	65.02	8.73
White	43.31	52.50	39.19	47.50	82.50
African American	4.12	37.19	6.95	62.81	11.07
Other	2.09	39.76	3.17	60.24	5.26
TOTAL	52.58	48.88	54.99	51.12	107.57
Sex †
Female	29.35	51.53	27.60	48.47	56.95
Male	23.23	45.88	27.39	54.12	50.62
TOTAL	52.58	48.88	54.99	51.12	107.57
Married ‡
Yes	39.72	52.73	35.60	47.27	75.33
No	12.75	39.87	19.24	60.13	31.99
TOTAL	52.47	48.90	54.84	51.10	107.31
Education ‡
Less than 9th grade	1.62	23.73	5.19	76.27	6.81
9th-11th grade	2.93	30.20	6.78	69.80	9.71
High school	13.52	45.97	15.89	54.03	29.40
Some college	16.06	46.80	18.26	53.20	34.32
College degree	18.35	67.71	8.75	32.29	27.09
TOTAL	52.47	48.89	54.87	51.11	107.34
Annual Family Income, $ ‡
Refuse to answer/missing data	1.44	44.99	1.77	55.01	3.21
< 20,000	5.32	26.72	14.60	73.28	19.92
20,000 to 64,999	23.85	47.71	26.14	52.29	49.99
≥ 65,000	21.96	63.74	12.49	36.26	34.45
TOTAL	52.58	48.88	54.99	51.12	107.57
Insurance ‡
Medical and dental	2.87	21.50	10.46	78.50	13.33
No dental	16.42	46.95	18.56	53.05	34.98
No medical or dental	32.26	57.61	23.74	42.39	56.00
TOTAL	51.55	49.42	52.75	50.58	104.30
Emotional Support †
Have support	50.92	50.08	50.75	49.92	101.68
Do not have support	1.17	25.10	3.50	74.90	4.67
Do not need support	0.47	39.58	0.71	60.42	1.18
TOTAL	52.56	48.88	54.97	51.12	107.53
Financial Support ‡
Have support	43.53	51.94	40.28	48.06	83.82
Do not have support	7.79	36.59	13.51	63.41	21.30
Support offered, not accepted	1.07	55.02	0.87	44.98	1.94
TOTAL	52.39	48.94	54.66	51.06	107.05
Close Friends ‡
None	1.00	34.00	1.95	66.00	2.95
Few (1–3)	9.99	36.00	17.75	64.00	27.74
Some (4–9)	25.32	53.05	22.41	46.95	47.73
≥ 10	16.23	56.40	12.55	43.60	28.78
TOTAL	52.55	49.02	54.65	50.98	107.20
Years in Neighborhood ‡
Less than 1	2.47	33.08	5.00	66.92	7.47
1–2	4.45	38.37	7.15	61.63	11.60
3–5	8.02	46.51	9.23	53.49	17.25
6–10	10.45	54.70	8.65	45.30	19.11
> 10	26.81	52.57	24.19	47.43	50.99
TOTAL	52.21	49.05	54.22	50.95	106.43
General Health ‡
Excellent	30.19	58.51	21.41	41.49	51.60
Good	15.32	44.07	19.44	55.93	34.76
Poor	7.00	33.11	14.15	66.89	21.15
TOTAL	52.51	48.85	54.99	51.15	107.51
Inactive Days ‡ §
None	39.23	51.76	36.57	48.24	75.80
Less than 1 week	5.47	51.57	5.14	48.43	10.62
More than 1 week	1.96	28.88	4.83	71.12	6.80
TOTAL	46.67	50.07	46.54	49.93	93.21

^*Source: Centers for Disease Control and Prevention.¹¹ Some numbers have been rounded.

^†P < .05.

^‡P ≤ .001.

^§Defined as inactive days due to poor physical or mental health during the past 30 days. Source: Centers for Disease Control and Prevention.¹¹

TABLE 3.

Weighted logistic regression model.

COVARIATE	ESTIMATE	ODDS RATIO	STANDARD ERROR	P VALUE
Intercept * †	1.1944	Not applicable	0.1761	< .001
Demographics
Age, years (reference: ≤ 54)
55–64	0.1599	1.173	0.132	.226
65–74	0.1695	1.185	0.1551	.275
≥ 75†	0.5281	1.696	0.1653	.001
Race/ethnicity (reference: white)
African American	−0.3141	0.73	0.1634	.055
Hispanic	−0.1106	0.895	0.3144	.725
Other	−0.2901	0.748	0.3927	.460
Sex (reference: male)
Female†	0.4309	1.539	0.1081	< .001
Education (reference: college degree)
Less than 9th grade‡	−0.9579	0.384	0.32	.003
9th-11th grade†	−0.9242	0.397	0.1955	< .001
High school	−0.5075	0.602	0.266	.056
Some college†	−0.639	0.528	0.1381	< .001
Insurance (reference: medical and dental)
No dental‡	−0.3669	0.693	0.1168	.002
No medical or dental†	−1.0005	0.368	0.2271	< .001
Income
Annual family income, $ (reference: ≥ 65,000)
Refuse to answer/missing data	0.3489	1.418	0.5054	.49
< 20,000‡	−0.687	0.503	0.1942	< .001
20,000 to 64,999‡	−0.2696	0.764	0.1335	.043
Social Support
Married (reference: yes)
Not married	−0.1675	0.846	0.1071	.118
Emotional support (reference: have support)
No support	−0.5439	0.58	0.2998	.07
No support needed	0.6299	1.877	0.7439	.397
Financial support (reference: have support)
No support†	−0.5148	0.598	0.1453	< .001
Support offered, not accepted	0.2297	1.258	0.3002	.444
Close friends (reference: ≥ 10)
Few (1–3)†	−0.3698	0.691	0.1095	< .001
None	0.0372	1.038	0.6019	.951
Some (4–9)‡	−0.0127	0.987	0.1069	.905
Years in neighborhood (reference: > 10)
Less than 1	−0.4701	0.625	0.2969	.113
1–2	−0.2332	0.792	0.202	.248
3–5	0.1008	1.106	0.1585	.525
6–10	0.0794	1.083	0.1819	.663
Health Status
Inactive days§ (reference: none)
Less than 1 week	0.0867	1.091	0.2495	.728
More than 1 week‡	−0.6858	0.504	0.2375	.004
General health (reference: excellent)
Good	−0.3276	0.721	0.2276	.150
Poor‡	−0.4331	0.648	0.18	.016

^*Represents the probability of having had more dental visits if all of the risk factors are absent.

^†P ≤ .001.

^‡P < .05.

^§Defined as inactive days due to poor physical or mental health during the past 30 days. Source: Centers for Disease Control and Prevention.¹¹

With regard to sociodemographic characteristics, all bivariate relationships except race/ethnicity remained statistically significant in the multivariate model. Compared with participants 54 years and younger, those older than 75 years were significantly more likely to have had a timely or self-care–related dental visit during the previous year. Female sex significantly increased the odds of having had a timely or self-care–related dental visit. The odds of having had a timely or self-care–related dental visit were significantly decreased for participants who did not have a college degree compared with those who had a college degree. The estimated odds of having had a timely dental visit, a self-care–related dental visit or both for those who had medical insurance but no dental insurance and for those who had no insurance at all were 0.69 (P = .002) and 0.37 (P < .001) times, respectively, the estimated odds for participants who had both medical and dental insurance. We tested the interactions between insurance status and social support with regard to having had timely or self-care–related dental visits and found that only financial support exhibited a slightly significant interaction with dental insurance. The odds of having had a dental visit were significantly lower for participants with an annual family income of less than $20,000 compared with the odds for those with an annual family income of $65,000 or more.

Self-reported physical and mental health status also was associated significantly with timely or self-care–related dental visits in the logistic regression model. Compared with participants who reported having had no days of inactivity as a result of physical or mental health issues, those reporting more than one week of inactivity were significantly less likely to have had a dental visit. Similarly, compared with those who reported having excellent general health, those who reported having poor general health were significantly less likely to have had a timely or self-care–related dental visit.

The logistic regression model revealed a complex picture of the relationship between social support and having had timely or self-care–related dental visits. We found no statistically significant association between marital status or having someone available to provide emotional support (such as talking over problems or helping make a difficult decision) and the odds of having had a dental visit. Social connection, as measured by the number of years lived in a neighborhood, also was not a significant predictor of having had a timely or self-care–related dental visit in this model. On the other hand, not having anyone to provide financial help if needed significantly decreased the odds of having had timely or self-care–related dental visits. In addition, compared with having a large number (10 or more) of close friends and family members, having only a few (one through three) close friends and family members significantly decreased the odds of having had a dental visit (Table 3).

DISCUSSION

Our study findings show that timely or self-care–related dental visits were influenced by some, but not all, components of social support. The relationship between social support and dental visits is affected by who is providing the support and the type of support provided. We found that marital status was not associated with dental visits. As Gallagher and colleagues¹⁰ reported, it is not simply the presence of a spouse that influences self-care, but the quality of the relationship and the functional support provided by the spouse that matters. In a clinical setting, it is not enough for a practitioner to ask a patient if he or she has a partner or is living with someone. The more important questions pertain to the quality of the relationship.

Because the NHANES data set includes both relatives and nonrelatives in the category of “close friends,” interpreting the significant odds ratios was challenging. It may be that among those who had only a few close friends or family members, these one, two or three people were less available or less inclined to help facilitate a dental visit compared with a larger group of family members or friends or compared with no family members or friends. This varying support also was reported by Lim and colleagues¹² in their study of patients’ responses to a course of instruction in plaque control. These authors found that gingivitis levels decreased in participants who had a higher number of reported discussions with friends and parents, but gingivitis levels actually increased among those who reported having had discussions with spouses. Sabbah and colleagues¹³ found marital status and number of friends, but not emotional support, to be associated with oral health in terms of the extent of loss of periodontal attachment.

Engage family members and friends.

Optimal oral health depends on the timely use of oral health services. One practical clinical intervention toward this goal is to identify and engage appropriate members of the patient’s social network to facilitate timely self-care–related dental visits. Patients, especially older adults, often are accompanied to dental appointments by a close family member or friend. This provides an opportunity for the dentist or a staff member to have a brief conversation with the companion and acknowledge his or her role in supporting the patient’s oral self-care activities. Such acknowledgment of support by health care professionals sends the message that family members, friends and even the community are important to a patient’s self-care.

By reviewing with patients the specific challenges related to attending dental visits and who might be able to assist them, practitioners can help patients begin to formulate their own strategies for receiving timely dental care. Gironda and Lui¹⁴ presented a sample of assessment tools appropriate for clinical settings to help practitioners determine social support needs. Ideally, the practitioner should conduct an assessment at the first patient encounter to ensure that the best possible social support is in place to facilitate timely self-care–related dental visits. Practitioners also should consider socioeconomic barriers because limited education, low income, lack of dental insurance and not having a source of financial assistance if needed significantly reduced the odds of a patient’s having had a dental visit.

Study limitations.

Although NHANES data were limited with regard to the number and type of social support variables, NHANES is one of the few large data sets with both social support and oral health behavior variables.¹¹ Our ability to measure the full range of social support and to identify those who specifically provide the support was constrained by the limited number of variables and the way in which they were worded in the NHANES. For example, combining support from family members and friends into one question about the number of close friends limits our ability to interpret and translate the findings. Investigators in future studies of the relationship between social support and dental visits should use some of the well-validated measures of social support found in the literature.

CONCLUSION

Timely or self-care–related dental visits depend on some, but not all, elements of one’s social support network. From a clinical standpoint, it may be worthwhile to identify those members of the patient’s network who either facilitate or impede timely or self-care–related dental visits. Engagement of appropriate members of the patient’s social network may help clinicians facilitate timely dental visits. ■

ABBREVIATION KEY.

NHANES

National Health and Nutrition Examination Survey