Disparity in dental attendance among older adult populations: a comparative analysis across selected European countries and the USA

Abstract

Background: The current study addresses the extent to which diversity in dental attendance across population subgroups exists within and between the USA and selected European countries. Method: The analyses relied on 2006/2007 data from the Survey of Health, Ageing and Retirement in Europe (SHARE) and 2004–2006 data from the Health and Retirement Study (HRS) in the USA for respondents ≥ 51 years of age. Logistic regression models were estimated to identify impacts of dental-care coverage, and of oral and general health status, on dental-care use. Results: We were unable to discern significant differences in dental attendance across population subgroups in countries with and without social health insurance, between the USA and European countries, and between European countries classified according to social welfare regime. Patterns of diverse dental use were found, but they did not appear predominately in countries classified according to welfare state regime or according to the presence or absence of social health insurance. Conclusions: The findings of this study suggest that income and education have a stronger, and more persistent, correlation with dental use than the correlation between dental insurance and dental use across European countries. We conclude that: (i) higher overall rates of coverage in most European countries, compared with relatively lower rates in the USA, contribute to this finding; and that (ii) policies targeted to improving the income of older persons and their awareness of the importance of oral health care in both Europe and the USA can contribute to improving the use of dental services.

INTRODUCTION

The proportion of older people is growing fast throughout western Europe and the USA1., 2., 3.. Older age brings many changes in people’s lives, including lower income from retirement and an increased likelihood of chronic diseases. Consequently, mobility can become limited, which can lead to decreasing social contacts and a worsening quality of life. Compounding this is the fact that common oral diseases are a problem in most industrialised countries, especially in adults, because better knowledge of the aetiology of dental diseases, self-care, various preventive measures and school dental care have resulted in improvements of oral health mostly in the young4., 5.. Recognising oral diseases is difficult for a layperson and thus, unlike general health care, all individuals, from children to the elderly, are advised to visit their dentist regularly6., 7., 8., 9.. Use of dental services is known to depend on several factors: patients’ perceived treatment needs; access to care; financial and practical resources; cost; and dentists’ recall traditions. In many countries, older people had their teeth extracted early in life because other treatment was not available or accessible and edentulous persons do not have much use for frequent dental services. The proportions of edentulous older adults have been shown to be high in some countries, whereas, in others, tooth loss is not as common or there has been a positive trend of reduction in edentulousness3., 4.. Paradoxically, rising numbers of dentate persons and remaining teeth mean greater need for dental treatment in the elderly. In addition, the oral health-care systems in all industrialised countries face big challenges because the type of dental treatment provided seems to be related more to a person’s education and income levels than is the case in general health care10., 11..

The Survey of Health, Ageing and Retirement in Europe (SHARE) and the Health and Retirement Study (HRS) in the USA are unique surveys allowing comparisons within and across several European countries, as well as between the USA and European countries. Recent studies using SHARE data have confirmed income inequalities in dental attendance from wave 2 (2006–2007), and socio-economic status inequalities throughout the life course from wave 3 (SHARELIFE), retrospective life histories10., 11.. Another study pooled wave 2 SHARE data across European countries to compare intercountry attendance rates in several logistic regressions¹². Our study is instead designed to compare socio-economic and demographic effects on dental attendance within and across the USA and the European countries in SHARE. Its uniqueness is established by: (i) using harmonised HRS and SHARE data; (ii) correctly estimating statistical standard errors by not pooling across European countries with different sample designs; and (iii) fully specifying theoretical models of dental utilisation as the basis for statistical analysis. Following the conceptual framework presented by Andersen¹³, we incorporate need and predisposing and enabling factors in our model of dental attendance. In making comparisons between the USA and selected European countries, we hypothesised that countries with Social Health Insurance (SHI) will have less variation between different population subgroups than countries without SHI, and that the USA will demonstrate more diversity in dental attendance than any European countries with or without SHI. In contrast to countries without SHI, countries with SHI were expected to offer better access to dental care for older populations by offering universal coverage to all socio-economic and demographic groups. In the USA, disparities in the use of dental care across population groups was attributed to the limited access to such care in non-urbanised and low-income parts of the nation14., 15..

METHODS

The data for our study were obtained from home-based face-to-face interviews conducted in wave 2 of SHARE and in wave 8 of the HRS in the USA for respondents ≥ 51 years of age. SHARE wave 2 was carried out in 2006/2007 and contains detailed information on the health, socio-economic status and family backgrounds of nearly 35,000 older Europeans from 14 countries (Denmark, Sweden, Austria, France, Germany, Ireland, Switzerland, Belgium, the Netherlands, Spain, Italy, Greece, the Czech Republic and Poland). As a panel survey, SHARE is designed to be similar to the HRS. More details about the methodology of SHARE are available in the literature and on the SHARE website16., 17.. Almost 19,000 older Americans were surveyed in wave 8 of the 2006 HRS regarding a wide range of health, socio-economic and demographic information. Administered by the Institute for Social Research at the University of Michigan and sponsored by the National Institute on Aging, the HRS provides data for the study of aging, retirement and health among older populations in the USA¹⁸.

SHARE wave 2 has been reviewed and approved by the Ethics Committee of the University of Mannheim. In most countries, respondents’ consent to participate in SHARE is verbal. The text for the verbal consent is found at the beginning of the cover screen questionnaire. In countries in which written consent is legally required, such as Poland, this is also obtained. After filling out a user statement, data are available and can be downloaded for free as long as the data are used for no purposes other than purely scientific. The Health and Retirement Study is under current Institutional Review Board (IRB) approval at the University of Michigan and the National Institute on Aging. No individual identifiers or links to individual identifiers are provided to researchers. Consent to participate in the HRS is verbal. Verbal consent is obtained for all participants in every wave before the interview starts. Separate written consent is obtained for the physical measurements and blood and saliva sample components of the enhanced face-to-face interview. The IRB at the University of Maryland has determined this study to be exempt (E4). To the best of the authors’ knowledge, this study has been conducted in full accordance with the World Medical Association Declaration of Helsinki.

Our study analysed self-reported dental insurance attendance and its correlates in the USA in wave 8 of the HRS and in European nations in wave 2 of SHARE. Persons in the wave 2 SHARE data responding ‘yes’ to the question, ‘During the last 12 months have you seen a dentist or dental hygienist?’, are considered to have had dental attendance in the past year. Persons in wave 8 of the HRS responding ‘yes’ to the question, ‘In the last 2 years have you seen a dentist for dental care, including dentures?’, are considered to have had dental attendance in the past 2 years.

We included dental insurance coverage among the enabling explanatory variables included in our logistic regression models of dental attendance. Dental coverage in SHARE is measured by responses of ‘entirely or mostly paid or reimbursed by social insurance or respondent’s health insurance’ or ‘mostly paid by respondent’ to the question, ‘Who finally pays for dental care?’. Persons responding ‘entirely paid by respondent’ to this question are considered to lack dental coverage in SHARE. In the 2006 HRS, persons answering ‘yes’ to the question, ‘In the last 2 years have you seen a dentist for dental care including dentures?’, are considered to have coverage if they responded ‘completely, mostly, or partially covered’ to a follow-up question regarding how much of their dental expenses were covered by insurance. Persons who did not use dental care in the HRS were also considered covered if they responded ‘yes’ to the question: ‘If you did need to see a dentist, would you expect any of the costs to be covered by insurance?’.

Additional explanatory variables in our logistic models for dental attendance included RAND-harmonised versions of predisposing variables for age (51–64, 65–69, 70–74, 75–79, 80–84 and 85+ years of age), sex and marital status (married/partner; widowed, divorced, separated; never married); need-based variables for self-reported general health status (three categories: excellent/very good; good; fair/poor), objective oral health status (SHARE: binary measure indicating whether or not the respondent wears dentures; HRS: binary measure indicating whether or not the respondent has lost all permanent teeth), ability to bite/chew on hard foods (binary measure; only in SHARE); and additional enabling variables for retirement/labour force status (retired, not retired in the labour force; not retired, not in the labour force) and household income quintiles according to country. These variables were included to detect demographic, socio-economic and health-related differences in dental attendance.

There are subtle distinctions between HRS- and SHARE-harmonised variables. For example, SHARE collected after-tax income in wave 2 whereas only pre-tax income was collected in wave 8 of the HRS. Labour force status in SHARE is determined by a single question regarding which labour force status most generally describes the person’s situation; HRS adapts responses to multiple questions to determine this status. In both SHARE and HRS, household income quintiles were computed separately for couples and single persons within each country. The International Standard Classification of Education (ISCED) was used to harmonise three categories of educational attainment in European countries with those of older Americans. European subjects with no degrees, unknown level of education or ISCED levels 1 or 2 were considered equivalent to American subjects with less than a high-school degree; European subjects with ISCED level 3 were considered equivalent to American subjects with a General Education Diploma, high-school degree, or with some experience of college but not a college degree; and European subjects with ISCED levels 4–6 were considered equivalent to American subjects with a college degree or higher educational attainment. In the European countries, two explanatory variables were added, indicating whether the person uses dentures and has difficulty biting on hard foods. In the USA, an explanatory variable was added to indicate whether the person was missing all of their permanent teeth.

The wave 2 SHARE sample of 34,415 persons in 14 European countries was reduced by 2,217 persons without weights, including all 1,134 persons in Ireland. An additional 2,590 persons with data missing on one or more of the analytical variables were excluded to produce a final SHARE sample of 29,680 older Europeans for our study. The wave 8 HRS sample of 18,469 individuals was reduced by excluding 1,506 persons without sample weights and by excluding another 52 persons with data missing on at least one of the analytical variables used for the study. This produced a final HRS analytical sample of 16,911 older Americans. Unless stated otherwise, all estimates discussed in the text were statistically significant at least at the 5% level (P ≤ 0.05). The software packages SUDAAN and STATA were used to produce our estimates19., 20.. The complex sample designs for each country were incorporated into the estimated standard errors, although Denmark used a simple random-sample design, and Greece, Poland, Sweden and Switzerland adopted stratified sample designs.

To explore potential influences of public insurance subsidy, countries were categorised into those with or those without comprehensive SHI for dental care (see Table 1 for details on the SHI categorisation). Moreover, in line with recent research on welfare-state regimes in dentistry, European countries were classified as belonging to the Scandinavian, Bismarckian, Southern or Eastern welfare state regime21., 22..

Table 1.

Social Health Insurance (SHI) coverage of examined countries

	SHI status	Description
Austria	Yes	Full public coverage of restorative treatment and some surgical procedures. Partial coverage of prosthetic treatment27., 30., 31.
Belgium	Yes	Full public coverage of basic preventive treatment and extractions. Partial coverage for restorative, removable prosthetic, other preventive and other surgical treatments. No coverage for periodontal treatment, fixed prosthetic or implants27., 30., 32.
Czech Republic	Yes	Public support through SHI for basic dental treatments and restricted to the least expensive options. About 25% of expenditure on dental care is funded through out-of-pocket payments33
Denmark	Yes	Partial public refund (30–65%) for dental treatments received from private practitioners7., 30., 34.
France	Yes	Mandatory health insurance system, which reimburses about 70% of fees for standard dental treatment of conservative treatment (including sealing and scaling) and extractions. High proportion of patient co-payment for prosthodontic treatment27., 30.
Germany	Yes	Wide public coverage for diagnostics, fillings, conservative treatment, periodontal treatment, endodontic treatment; partial coverage for prosthetic treatment (50–60%). No coverage for implants27., 30.
Greece	No	Extensive differences in coverage between social insurance funds; large parts of the population are not covered for dental services35
Italy	No	Public coverage only for emergency treatment and for treatment of persons with disabilities, human immunodeficiency virus or rare disease27., 30., 36.
The Netherlands	No	For young people up to the age of 21 years, dental care is covered by basic health insurance. For people 22 years of age and above a complementary voluntary health insurance is needed for dental care37
Poland	Yes	Public coverage for basic dental treatment procedures that are reimbursed by the National Health Fund (NFZ); limited access to care because of restrictions in provider availability38
Spain	No	No public funding for dental care except extractions30., 39., 40..
Sweden	Yes	Fixed public subsidies for all types of general dental treatment, including prosthodontic and orthodontic treatments27., 41.
Switzerland	No	No public coverage; exceptions include treatment of very severe or unavoidable disease30., 42.
USA	No	Most dental treatment is funded privately; only 5% of expenditure is publicly funded43

SHI, comprehensive Social Health Insurance for the general population; ‘yes’ indicates that at least some public support exists for basic dental procedures (preventive, restorative, prosthodontic and emergency surgical care); ‘no’ indicates that public support, if any, exists only for some exceptional cases.

RESULTS

Table 2 shows descriptive statistics of dental attendance overall and by various covariates, without controlling for other variables, for the USA and for each European country clustered according to welfare state regime. Sample sizes for each country are shown in the last row of the table.

Table 2.

Percentage of older individuals with dental attendance according to population characteristic: USA and selected European countries (Health and Retirement Study, 2004–2006; Survey of Health, Ageing and Retirement in Europe, 2006–2007)

		European welfare state regime
Characteristic	USA‡	Bismarckian						Scandinavia		Southern			Eastern
		Austria§	Belgium§	France§	Germany§	The Netherlands‡	Switzerland‡	Denmark§	Sweden§	Greece‡	Italy‡	Spain‡	Czech Republic§	Poland§
Overall	66.02 (0.86)	51.45 (2.15)	50.79 (1.17)	48.76 (1.43)	73.18 (1.79)	66.26 (1.48)	72.65 (1.26)	79.84 (0.85)	81.91 (0.91)	36.92 (1.02)	35.93 (1.78)	30.33 (1.48)	58.78 (1.99)	23.09 (0.97)
Age
51–64 years	70.42* (0.98)	61.16* (2.41)	57.903, 4, 5, 6 (1.49)	51.894,6 (2.25)	78.014,5,6 (1.95)	76.07* (1.78)	78.164,5 (1.57)	86.343,4,5 (0.98)	82.136(1.28)	45.603, 4, 5, 6 (1.47)	40.713,4,5 (2.18)	39.44* (2.33)	68.60* (2.01)	29.99* (1.43)
65–69 years	65.354,5 (1.25)	51.284 (3.98)	53.833, 4, 5, 6 (2.57)	46.04 (3.56)	78.144,5,6 (2.44)	63.704 (3.42)	73.765 (3.32)	83.863,4,5 (2.14)	84.005,6 (2.00)	39.963, 4, 5, 6 (2.73)	39.453,4,5 (2.97)	27.924,5,6(3.18)	55.635,6 (3.55)	19.454,5 (2.52)
70–74 years	62.73 (1.52)	45.33 (4.20)	44.405,6 (3.23)	52.264,6 (3.48)	73.995,6 (3.62)	54.72 (3.16)	75.074,5 (3.62)	71.40 (2.80)	85.895,6 (2.33)	28.735,6 (2.44)	30.47 (3.44)	26.624,5,6(3.01)	45.575,6 (4.04)	15.824,5 (2.43)
75–79 years	58.84 (1.33)	38.46 (4.54)	41.295,6 (2.47)	41.23 (3.36)	65.955 (3.41)	49.17 (2.75)	64.37 (4.18)	69.30 (3.27)	83.606 (2.94)	23.746 (2.95)	26.81 (3.51)	15.87 (2.85)	45.705,6 (5.22)	9.07 (1.97)
80–84 years	58.88 (1.68)	27.12*,† (5.11)	29.52 (3.21)	43.01 (4.02)	45.50 (4.94)	39.75* (3.74)	61.14 (5.44)	63.12 (4.11)	75.58 (3.80)	18.07 (3.50)	27.22 (5.27)	17.01 (3.63)	23.23 (4.53)	6.09 (2.32)
85 years and older	52.50* (2.04)	22.52*,† (7.16)	28.01† (4.61)	38.57† (5.55)	49.40† (7.87)	21.56*,† (7.45)	33.43*,† (7.01)	40.65*,† (5.69)	69.87 (4.90)	12.20 (3.60)	27.64† (7.84)	10.12 (3.10)	13.74† (6.71)	9.57† (4.80)
Sex
Female	67.10 (0.92)	50.07 (2.31)	50.09 (1.43)	51.72 (1.60)	73.65 (2.04)	65.89 (1.86)	73.68 (1.69)	79.93 (1.17)	82.82 (1.22)	38.50 (1.41)	35.84 (1.91)	32.14 (1.80)	57.98 (2.15)	22.68 (1.26)
Male	64.72 (0.97)	53.10 (2.82)	51.60 (1.42)	45.211 (2.04)	72.63 (2.01)	66.66 (1.66)	71.44 (1.88)	79.74 (1.24)	80.89 (1.35)	35.09 (1.47)	36.03 (2.28)	28.14 (2.07)	59.78 (2.58)	23.61 (1.51)
Education (ISCED scores)¶
Category 1	36.09* (1.43)	39.97* (3.32)	40.48* (1.62)	42.27* (1.80)	57.19* (3.26)	54.58* (2.08)	62.17* (2.33)	59.29* (2.22)	77.94 (1.37)	31.962 (1.25)	30.24* (1.74)	25.74* (1.48)	49.81* (2.45)	13.35* (1.14)
Category 2	66.39* (0.83)	53.78* (2.79)	55.54* (2.01)	52.43 (2.66)	74.28* (1.93)	75.03* (2.44)	76.11 (2.00)	82.82* (1.24)	82.42 (2.15)	43.29 (2.29)	53.27 (3.71)	48.71 (4.68)	63.90* (2.32)	27.98* (1.62)
Category 3	86.71* (0.67)	61.44* (2.65)	65.31* (1.98)	59.09 (2.29)	80.40* (2.01)	82.75* (1.79)	80.78 (2.07)	88.88* (1.12)	87.79* (1.39)	49.23 (2.61)	47.46 (3.74)	53.96 (4.82)	72.28* (3.15)	42.70* (3.33)
Marital status
Married/ partner	70.87* (0.99)	55.902 (2.55)	53.00 (1.36)	49.51 (1.78)	77.06 (1.89)	70.652 (1.45)	74.98 (1.46)	84.042 (0.89)	84.56 (0.93)	38.35 (1.21)	37.36 (1.90)	31.43 (1.57)	63.91 (2.11)	25.84 (1.13)
Widowed/ divorced/ separated	55.85* (1.08)	42.67 (3.06)	42.76* (1.99)	47.54 (2.70)	62.841 (2.59)	53.43 (3.15)	64.41* (2.75)	68.90 (2.08)	74.701 (2.36)	31.71* (2.04)	32.92 (3.26)	26.07 (3.38)	48.771 (3.28)	17.681 (1.90)
Never married	62.42* (2.49)	45.77† (5.10)	55.86 (4.76)	44.25 (4.93)	70.58† (5.50)	61.79† (4.88)	80.57† (4.64)	76.19 (4.23)	81.43† (3.69)	41.61 (4.50)	30.56 (6.20)	30.62 (5.82)	51.51† (11.40)	20.43† (5.24)
Health status
Excellent/ very good	77.87* (0.86)	58.58 (3.33)	59.99* (1.89)	53.00 (2.57)	74.30 (4.05)	74.02* (1.94)	77.30* (1.79)	85.11 (1.05)	86.78* (1.25)	44.33* (1.78)	41.51 (3.33)	29.08 (3.30)	68.59* (2.97)	32.25 (4.01)
Good	65.34* (1.06)	52.55 (3.00)	50.73* (1.51)	52.36 (2.29)	78.47 (1.88)	68.37* (1.76)	69.70 (2.11)	81.56 (1.64)	80.77 (1.66)	37.10* (1.63)	38.66 (2.26)	33.81 (2.30)	59.94 (2.67)	27.71 (1.88)
Fair/poor	47.32* (1.09)	43.94* (2.74)	41.99* (2.24)	42.36* (2.38)	67.172 (2.31)	55.98* (2.39)	67.19 (3.13)	66.44* (2.08)	76.36 (1.90)	27.11* (1.81)	31.60* (1.93)	27.712 (1.89)	53.39 (2.72)	19.70* (1.16)
Objective oral health status (Europe: denture use; USA: tooth loss)
No impairment	74.79 (0.82)	65.69 (2.56)	63.57 (1.67)	51.12 (1.90)	78.00 (2.28)	87.42 (1.13)	81.57 (1.32)	89.81 (0.77)	84.92 (0.90)	42.37 (1.22)	38.47 (2.04)	33.18 (1.94)	68.78 (1.93)	25.59 (1.57)
Impairment	20.981 (0.91)	41.481 (2.56)	39.801 (1.53)	44.541 (2.10)	68.501 (1.99)	41.401 (1.81)	53.621 (2.51)	55.211 (1.96)	61.781 (3.19)	22.161 (1.71)	31.121 (2.26)	26.491 (2.06)	43.681 (2.91)	21.171 (1.21)
Chewing ability (Europe only)
Can bite/chew on hard foods	NA	55.16 (2.04)	54.85 (1.18)	50.89 (1.60)	75.53 (1.94)	69.70 (1.43)	74.99 (1.30)	82.99 (0.89)	83.08 (0.92)	39.88 (1.18)	37.80 (1.98)	30.93 (1.63)	62.54 (2.05)	25.56 (1.23)
Cannot bite/chew on hard foods		36.301 (3.79)	38.781 (2.21)	41.031 (2.61)	63.281 (2.95)	46.561 (3.58)	55.151 (4.14)	65.901 (2.35)	67.851 (4.19)	27.551 (1.90)	31.131 (2.48)	27.96 (3.05)	44.851 (3.08)	17.701 (1.49)
Retirement/labour force status
Retired	60.89* (1.15)	48.53 (2.39)	47.41 (1.51)	47.54 (1.78)	71.98 (2.02)	57.95 (1.77)	68.74 (1.98)	73.99 (1.31)	80.89 (1.20)	29.81* (1.44)	37.05 (2.02)	24.15 (1.98)	53.79* (2.50)	17.83* (1.08)
Not retired, in labour force	73.66* (0.89)	66.65* (3.32)	59.80* (1.95)	50.82 (2.41)	76.28 (2.16)	78.97* (1.58)	77.89* (1.81)	87.55* (1.09)	83.33 (1.42)	47.70* (1.98)	41.05 (3.06)	42.82* (2.92)	67.99 (2.21)	35.53* (2.54)
Not retired, not in labour force	53.43* (1.89)	39.85 (4.57)	46.85 (2.25)	49.47 (4.11)	70.17 (4.50)	62.20 (2.84)	68.85 (3.47)	74.58 (3.39)	81.51† (5.88)	38.41* (2.00)	29.64* (2.62)	28.86 (2.14)	69.69† (7.81)	27.72* (2.73)
Income quintile**
1st quintile	41.32* (1.48)	44.185,6 (3.45)	37.46* (2.37)	37.66* (2.95)	60.723,4,5 (3.50)	52.973,4,5 (2.85)	64.743,4,5 (3.02)	58.16* (2.32)	74.913,4,5 (2.26)	27.853,4,5 (2.08)	28.83 (2.81)	25.66 (3.17)	51.364 (3.30)	19.74 (2.11)
2nd quintile	53.70* (1.06)	41.015,6 (3.46)	46.634,5 (2.29)	48.764 (2.88)	69.183,4,5 (2.68)	58.874,5 (2.73)	63.203,4,5 (3.04)	77.59* (1.98)	77.193,4,5 (2.17)	30.464,5 (2.18)	30.75 (2.95)	21.97 (2.62)	48.624 (4.04)	13.10* (1.72)
3rd quintile	62.18* (1.25)	47.115,6 (3.48)	48.604,5 (2.08)	51.58 (3.54)	78.27 (2.70)	66.645 (3.41)	73.755 (2.77)	86.98 (1.61)	85.82 (1.78)	34.704,5 (2.24)	28.29 (2.93)	26.82 (2.89)	57.23 (3.62)	19.87 (2.01)
4th quintile	74.70* (1.15)	60.18 (3.34)	59.97 (2.18)	58.14 (2.92)	79.50 (2.60)	74.56 (3.08)	80.52 (2.52)	87.14 (1.59)	87.00 (1.75)	44.79 (2.30)	41.41* (2.95)	35.25* (3.18)	59.53 (2.36)	28.30* (2.38)
5th quintile	85.94* (0.72)	68.87 (3.29)	61.47 (2.71)	49.804 (2.59)	79.36 (2.57)	79.91 (2.60)	81.92 (2.49)	89.35 (1.51)	86.98 (1.85)	47.70 (2.38)	52.40* (2.82)	44.92* (3.54)	71.90* (3.23)	36.14* (2.46)
Dental insurance coverage
Not covered	55.34 (1.00)	57.88† (5.03)	48.93† (6.96)	61.51† (3.66)	72.98† (5.28)	52.84 (3.31)	72.41 (1.42)	64.12 (3.55)	80.11 (1.57)	39.00 (1.49)	38.78 (1.73)	31.79 (1.78)	59.03 (9.03)	37.81 (2.26)
Covered	77.611 (0.75)	51.03 (2.24)	50.82 (1.20)	48.501 (1.42)	73.19 (1.80)	68.191 (1.45)	73.55 (2.70)	81.271 (0.87)	82.91 (1.11)	35.32 (1.35)	24.951 (5.38)	27.00 (2.36)	58.76 (1.99)	18.591 (1.03)
Observations	16,911	1,236	2,769	2,493	2,345	2,333	1,332	2,329	2,172	2,878	2,783	1,988	2,495	2,135

Standard errors are in parentheses and account for sample design. Results are weighted. Estimates for the USA refer to dental-care use within the past 24 months; estimates for European countries refer to dental-care use within the past 12 months.

¹Indicates significantly different from the first characteristic listed in the group (P ≤ 0.05).

²Indicates significantly different from the second characteristic listed in the group (P ≤ 0.05).

³Indicates significantly different from the third characteristic listed in the group (P ≤ 0.05).

⁴Indicates significantly different from the fourth characteristic listed in the group (P ≤ 0.05).

⁵Indicates significantly different from the fifth characteristic listed in the group (P ≤ 0.05).

⁶Indicates significantly different from the sixth characteristic listed in the group (P ≤ 0.05).

^*Indicates significantly different from all other characteristics listed in the group (P ≤ 0.05).

^†Indicates that estimate is based on fewer than 100 observations.

^‡Country does not have comprehensive social health insurance benefits incorporating dental coverage for older persons.

^§Country has comprehensive social health insurance benefits incorporating dental coverage for older persons. See table A.1 (Appendix) for further country-specific details about SHI systems.

^¶Category 1, International Standard Classification of Education (ISCED) = 1,2 (Europe) or <High School degree (USA); Category 2, ISCED = 3 (Europe) or High School degree or some college (USA); Category 3, ISCED = 4,6 (Europe) or college degree (USA).

^**Arranged lowest (1st quintile) to highest (5th quintile).

Overall attendance rates

There was considerable variation across countries in the overall percentage of respondents reporting dental attendance; the highest percentage of dental attendance in Europe was observed for the Scandinavian countries Denmark (80%) and Sweden (82%), followed by Germany (73%), Switzerland (73%), the Netherlands (66%) and the Czech Republic (59%). The other Bismarckian countries (Austria, Belgium and France) each had attendance rates of about 50%. All of these countries, with the exception of the Netherlands and Switzerland, have comprehensive SHI. None of the Southern countries has comprehensive SHI, and none has attendance rates in excess of 37%. In contrast, Poland, which does have comprehensive SHI, had the lowest dental-attendance rate of all the selected European countries, at 23%. The 2-year attendance rate in the USA, of 66%, is not strictly comparable with the 1-year European rates and would be lower on an annual basis.

Demographic variables

For the predisposing demographic variables age, sex and marital status, we generally found fairly consistent disparity, or lack thereof, across countries. In all countries, dental-attendance rates declined with age. Without exception, dental-attendance rates for those 51–64 years of age were higher than for older age cohorts. In all European countries, with the exception of Germany, Switzerland, Denmark and Sweden, attendance rates were, at most, 49% for those ≥75 years of age. With the exception of Sweden, access and edentulous issues may be responsible for dental-attendance rates of no higher than 49% for non-institutionalised persons ≥85 years of age in European countries.

Among all countries, significantly different attendance rates between the sexes were found only in France: the female dental-attendance rate in France was 7% points higher (52%) than the male dental-attendance rate (45%).

With only a few exceptions (France, Italy and Spain), married persons had higher dental-attendance rates than divorced or separated persons. Only in the USA, with its much larger sample size, were dental-attendance rates significantly higher for married persons than for those never married. In the USA, Switzerland and Greece, dental-attendance rates for divorced or separated persons were also significantly lower than those for never-married individuals.

Socio-economic status variables

With regard to the socio-economic status enabling variables for education, household income and labour market status, we also generally found similar disparities in dental attendance across the countries. For example, in every country, dental use was significantly higher in the highest income quintile than in the lowest income quintile. With only one exception (Greece) was dental attendance higher at the highest educational level than at the lowest one. Only in the SHI countries France, Germany and Sweden were there no disparities in attendance across retirement/labour force status. Typically in the other countries, non-retired persons in the labour force had higher dental-attendance rates than did those persons retired or otherwise out of the labour force.

Health status variables

With regard to need variables measured according to self-reported general or oral health status, we also found similar disparities in dental attendance across countries in our study. Regardless of country of residence, persons in fair or poor health had lower dental attendance than did those in good or very good/excellent health. In every European country, persons with dentures had lower dental attendance than did those without dentures. In the USA, edentate persons had much lower dental-use rates than did dentate persons. With the exception of Spain, those in European countries with difficulty chewing or biting on hard foods had lower dental-attendance rates than did those without this problem.

Dental coverage

In the SHI countries Austria, Belgium, Germany, Sweden and the Czech Republic, parity in dental attendance was found between those with and without dental coverage. Disparity in dental attendance was found between those with and without coverage in the SHI countries France, Denmark and Poland. Unexpectedly, those reporting no coverage in France and Poland had higher dental attendance than those reporting coverage. In the non-SHI countries, the Netherlands and the USA, dental use was higher for those with coverage than for those without coverage. However, there was no evidence of a disparity in attendance in the non-SHI countries Switzerland, Greece and Spain. By contrast, those in Italy reporting no coverage had higher dental-attendance rates than those reporting coverage.

Table 3 shows estimated odds ratios from country-specific logistic regression models adjusted by controls for the demographic, socio-economic status and health-status covariates. Disparity was measured by comparing dental attendance in each population subgroup with that of the reference subgroup, rather than with all other subgroups within a variable category, as in Table 2. To aid in interpreting the table, the odds ratio estimate for men in the USA, of 0.61, indicates that the odds of an elderly male having a dental visit are nearly 40% less than those of an elderly female, after adjusting for other covariates, where the odds are defined as the probability of a dental visit divided by the probability of not having a dental visit.

Table 3.

Adjusted odds ratios from logistic regressions for likelihood of dental attendance for older individuals: USA and selected European countries (Health and Retirement Study, 2004–2006; Survey of Health, Ageing and Retirement in Europe, 2006–2007)

		European welfare state regime
(a)	USA†	Bismarckian
		Austria‡	Belgium‡	France‡	Germany‡	The Netherlands†	Switzerland†
Age
51–64 years (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
65–69 years	1.31* (1.16–1.48)	0.87 (0.59–1.27)	1.10 (0.85–1.43)	0.79 (0.53–1.17)	0.75 (0.49–1.15)	0.77 (0.50–1.19)	0.85 (0.50–1.44)
70–74 years	1.43* (1.24–1.66)	0.86 (0.59–1.26)	0.84 (0.61–1.17)	1.04 (0.71–1.52)	0.65* (0.43–1.00)	0.67 (0.43–1.05)	1.04 (0.60–1.80)
75–79 years	1.40* (1.24–1.59)	0.68 (0.43–1.06)	0.79 (0.61–1.04)	0.76 (0.50–1.14)	0.49* (0.32–0.75)	0.71 (0.46–1.10)	0.64 (0.37–1.12)
80–84 years	1.64* (1.36–1.97)	0.42* (0.24–0.71)	0.53 (0.37–0.76)	0.81 (0.50–1.29)	0.24* (0.14–0.42)	0.49* (0.32–0.76)	0.72 (0.38–1.37)
85 years and older	1.50* (1.23–1.83)	0.47* (0.23–0.98)	0.48* (0.26–0.89)	0.72 (0.39–1.33)	0.31* (0.14–0.67)	0.20 (0.08–0.46)	0.23* (0.11–0.50)
Sex
Women (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Men	0.61* (0.57–0.66)	0.75 (0.56–1.01)	0.86* (0.74–1.00)	0.69* (0.56–0.85)	0.58* (0.45–0.73)	0.78 (0.61–1.00)	0.59* (0.44–0.80)
Education (ISCED scores)§
Category 1	0.54* (0.47–0.62)	0.79 (0.55–1.15)	0.52* (0.41–0.66)	0.57* (0.44–0.74)	0.42* (0.30–0.60)	0.45* (0.30–0.67)	0.60* (0.41–0.88)
Category 2	0.28* (0.23–0.34)	1.06 (0.78–1.45)	0.77* (0.60–1.00)	0.77* (0.60–1.00)	0.72* (0.56–0.93)	0.77 (0.56–1.06)	0.88 (0.61–1.27)
Category 3 (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Marriage status
Married/partnered (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Widowed/divorced/ separated	0.56* (0.50–0.62)	0.88 (0.65–1.19)	0.81 (0.64–1.03)	0.96 (0.73–1.27)	0.67* (0.50–089)	0.67* (0.50–0.93)	0.66* (0.48–0.91)
Never married	0.67* (0.51–0.87)	0.83 (0.51–1.34)	1.07 (0.72–1.61)	0.78 (0.50–1.22)	0.67 (0.39–1.15)	0.51* (0.31–0.84)	1.21 (0.62–2.36)
Health status
Excellent/very good (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Good	0.73* (0.65–0.82)	0.98 (0.70–1.37)	0.79* (0.66–0.94)	1.04 (0.80–1.35)	1.55* (1.05–2.30)	1.03 (0.78–1.36)	0.86 (0.63–1.17)
Fair/poor	0.55* (0.50–0.62)	0.89 (0.62–1.29)	0.69* (0.54–0.89)	0.79 (0.55–1.13)	1.17 (0.72–1.89)	0.81 (0.64–1.03)	0.95 (0.65–1.40)
Objective oral health status (Europe: denture use; USA: tooth loss)
No impairment (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Impairment	0.13* (0.12–0.15)	0.54* (0.41–0.73)	0.51* (0.42–0.62)	1.00 (0.80–1.25)	0.91 (0.68–1.20)	0.14* (0.11–0.19)	0.35* (0.26–0.48)
Chewing ability (Europe only)
Can bite/chew on hard foods (Ref.)	N/A	1.00	1.00	1.00	1.00	1.00	1.00
Cannot bite/chew on hard foods		0.79 (0.54–1.15)	0.76* (0.62–0.92)	0.78* (0.61–1.09)	0.82 (0.61–1.09)	0.76 (0.51–1.15)	0.71 (0.48–1.04)
Retirement/labour force status
Not retired, in labour force (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Not retired, not in labour force	1.10 (0.93–1.31)	0.55* (0.33–0.90)	1.09 (0.87–1.37)	1.21 (0.84–1.73)	1.04 (0.65–1.66)	1.21 (0.89–1.64)	0.91 (0.56–1.46)
Retired	1.32* (1.18–1.48)	0.75 (0.53–1.06)	1.04 (0.82–1.33)	1.24 (0.89–1.72)	1.67 (1.15–2.45)	1.30 (0.92–1.84)	1.68* (1.02–2.76)
Income quintile¶
1st quintile	0.63* (0.52–0.76)	0.55* (0.36–0.83)	0.61* (0.45–0.82)	0.78 (0.54–1.13)	0.53* (0.35–0.81)	0.72 (0.51–1.01)	0.66 (0.41–1.06)
2nd quintile	0.41* (0.35–0.48)	0.50* (0.33–0.75)	0.86 (0.64–1.17)	1.20 (0.88–1.62)	0.86 (0.58–1.28)	0.84 (0.57–1.22)	0.51* (0.32–0.81)
3rd quintile	0.35* (0.30–0.41)	054* (0.35–0.83)	0.83 (0.62–1.12)	1.25 (0.88–1.79)	1.22 (0.77–1.93)	0.87 (0.54–1.39)	0.78 (0.48–1.26)
4th quintile	0.26* (0.22–0.31)	0.89 (0.58–1.37)	1.03 (0.78–1.36)	1.49* (1.09–2.03)	1.03 (0.67–1.58)	0.96 (0.67–1.39)	0.95 (0.59–1.53)
5th quintile (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Dental insurance coverage
Not covered (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Covered	2.09* (1.90–2.30)	1.00 (0.64–1.56)	1.02 (0.54–1.92)	0.47* (0.23–0.95)	0.73 (0.33–1.65)	2.21* (1.60–3.04)	1.01 (0.72–1.40)

	European welfare state regime
(b)	Scandinavia		Southern			Eastern
	Denmark‡	Sweden‡	Greece†	Italy†	Spain†	Czech Republic‡	Poland‡
Age
51–64 years (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
65–69 years	1.01 (0.62–1.65)	1.30 (0.84–2.01)	1.07 (0.79–1.43)	0.96 (0.73–1.27)	0.71 (0.46–1.08)	0.64* (0.43–0.95)	0.94 (0.64–1.38)
70–74 years	0.63 (0.38–1.04)	2.07* (1.25–3.45)	0.79 (0.57–1.10)	0.72 (0.51–1.02)	0.69 (0.45–1.06)	0.48* (0.32–0.73)	0.73 (0.48–1.11)
75–79 years	0.74 (0.43–1.25)	2.21* (1.28–3.84)	0.65* (0.44–0.96)	0.64* 0.43–0.97)	0.36* (0.20–0.62)	0.56* (0.38–0.81)	0.42* (0.25–0.71)
80–84 years	0.71 (0.1–1.24)	1.49 (0.84–2.65)	0.50* (0.29–0.87)	0.72 (0.41–1.26)	0.41* (0.22–0.79)	0.22* (0.13–0.36)	0.33* (0.14–0.78)
85 years and older	0.33* (0.16–0.68)	1.50 (0.76–2.94)	0.34 (0.16–0.76)	0.84 (0.39–1.83)	0.24* (0.10–0.58)	0.12* (0.04–0.39)	0.57 (0.17–1.86)
Sex
Women (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Men	0.62* (0.48–0.79)	0.75* (0.58–0.96)	0.74* (0.59–0.93)	0.80* (0.64–1.00)	0.73* (0.55–0.97)	0.82 (0.63–1.07)	0.80 (0.62–1.03)
Education (ISCED scores)§
Category 1	0.41* (0.29–0.58)	0.65* (0.46–0.91)	0.81 (0.62–1.07)	0.66* (0.46–0.97)	0.41* (0.25–0.66)	0.46* (0.33–0.64)	0.38* (0.26–0.54)
Category 2	0.85 (0.62–1.16)	0.74 (0.49–1.09)	0.88 (0.65–1.17)	1.36 (0.91–2.02)	0.79 (0.43–1.43)	0.69* (0.48–0.98)	0.60 (0.43–0.83)
Category 3 (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Marriage status
Married/partner (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Widowed/divorced/separated	0.61* (0.46–0.82)	0.67* (0.48–0.93)	0.99 (0.77–1.27)	1.17 (0.88–1.53)	1.04 (0.67–1.62)	0.67* (0.49–0.92)	0.98 (0.71–1.34)
Never married	0.65 (0.37–1.14)	0.92 (0.53–1.60)	0.99 (0.66–1.49)	0.63 (0.32–1.23)	1.24 (0.68–2.25)	0.58 (0.25–1.36)	0.81 (0.39–1.70)
Health status
Excellent/very good (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Good	1.26 (0.93–1.70)	0.72* (0.52–0.99)	0.93 (0.75–1.16)	0.93 (0.68–1.27)	1.55* (1.02–2.33)	0.94 (0.68–1.30)	0.94 (0.68–1.30)
Fair/poor	0.72* (0.53–0.97)	0.62* (0.44–0.86)	0.79 (0.61–1.03)	0.83 (0.62–1.13)	1.77* (1.14–2.74)	1.01 (0.72–1.44)	0.87 (0.56–1.34)
Objective oral health status (Europe: denture use; USA: tooth loss)
No impairment (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Impairment	0.25* (0.19–0.33)	0.35* (0.24–0.49)	0.62* (0.48–0.81)	0.94 (0.74–1.20)	1.09 (0.79–1.49)	0.55* (0.44–0.70)	0.85 (0.67–1.08)
Chewing ability (Europe only)
Can bite/chew on hard foods (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Cannot bite/chew on hard foods	0.73* (0.54–0.98)	0.72 (0.46–1.11)	0.90 (0.71–1.15)	1.04 (0.80–1.35)	1.07 (0.76–1.49)	0.62* (0.47–0.80)	0.85 (0.65–1.10)
Retirement/labour force status
Not retired, in labour force (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Not retired, not in labour force	1.01 (0.61–1.67)	1.24 (0.55–2.79)	0.91 (0.68–1.23)	1.03 (0.70–1.53)	0.76 (0.50–1.15)	1.45 (0.77–2.73)	1.16 (0.80–1.68)
Retired	1.83* (1.13–2.96)	1.13 (0.76–1.69)	0.78 (0.60–1.02)	1.45* (1.06–1.97)	0.79 (0.53–1.18)	1.20 (0.88–1.64)	0.78 (0.57–1.07)
Income quintile¶
1st quintile	0.46* (0.29–0.73)	0.66 (0.42–1.04)	0.58* (0.42–0.79)	0.56* (0.39–0.80)	0.64* (0.41–0.99)	0.71 (0.48–1.06)	0.67* (0.46–0.98)
2st quintile	0.85 (054–1.35)	0.63* (0.40–0.99)	0.72* (0.52–1.00)	0.57* (0.38–0.84)	0.62 (0.38–1.01)	0.79 (0.50–1.26)	0.55* (0.37–0.83)
3rd quintile	1.23 (0.78–1.96)	1.10 (0.70–1.74)	0.80 (0.59–1.09)	0.48* (0.33–0.69)	0.71 (0.47–1.07)	1.04 (0.69–1.58)	0.76 (0.52–1.09)
4th quintile	0.98 (0.63–1.53)	1.04 (0.66–1.64)	1.11 (0.84–1.48)	0.74 (0.53–1.02)	0.82 (0.53–1.27)	0.79 (0.56–1.12)	1.03 (0.73–1.44)
5th quintile (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Dental insurance coverage
Not covered (Ref.)	1.00	1.00	1.00	1.00	1.00	1.00	1.00
Covered	2.03* (1.39–2.95)	1.09 (0.84–1.42)	0.85 (0.71–1.02)	0.58* (0.35–0.97)	0.91 (0.68–1.21)	0.83 (0.43–1.60)	0.49* (0.38–0.63)

95% confidence intervals are in parentheses. The adjusted value refers to the inclusion in the regression of the control variables listed in the rows of the table. The adjusted odds ratio is an estimate of (probability of dental coverage/probability of no dental coverage) for persons with row characteristic divided by (probability of dental coverage/probability of no dental coverage) for persons in the reference group. The SHARE sample is allocated across the 13 countries for the regressions as follows: Austria (1,208), Belgium (2,753), Czech Republic (2,494), Denmark (2,325), France (2,465), Germany (2,326), Greece (2,858), Italy (2,695), The Netherlands (2,331), Poland (2,135), Spain (1,823), Sweden (2,478) and Switzerland (1,325).

Ref. indicates reference group.

^†Country does not have comprehensive social health insurance benefits incorporating dental coverage for older persons.

^‡Country has comprehensive social health insurance benefits incorporating dental coverage for older persons. See table A.1 (Appendix) for further country-specific details about social health insurance systems.

^§Category 1, International Standard Classification of Education (ISCED) = 1,2 (Europe) or <High School degree (USA); Category 2, ISCED = 3 (Europe) or High School degree or some college (USA); Category 3, ISCED = 4,6 (Europe) or college degree (USA).

^¶Arranged lowest (1st quintile) to highest (5th quintile).

^*Indicates statistical significance at least at the 5% level.

Predisposing demographic variables

Similarly to Table 2, most countries show at least one older age group with lower odds of dental attendance than persons 51–64 years of age, but no age effect was found in France, and unexpectedly higher odds of dental attendance for older age groups were discovered in the USA and Sweden in the adjusted results.

Despite finding a gender effect only in France (Table 2), adjusted results show that male subjects had lower odds of dental attendance than female subjects in most countries. Only in Austria, the Netherlands and the Eastern countries were no gender effects found (Table 3).

Similarly to the results presented in Table 2, widowed, divorced and separated older persons had lower odds of dental attendance than older married persons, but Austria, the Netherlands, Greece and Poland joined France, Italy and Spain in showing no effect of marital status on dental use. The Netherlands joined the USA as the only countries with lower odds of dental attendance for never-married persons compared with married individuals.

Enabling socio-economic status variables

Like the unadjusted results, in most countries we continued to find lower odds of dental attendance in the lowest or next to lowest income quintile compared with the highest quintile, but we found no such effect of income on dental use in the Netherlands, France or the Czech Republic.

In the adjusted results, Austria joined Greece, from Table 2, in showing no effect of education on dental attendance. Otherwise in Table 3, older persons in the lowest education category continued to exhibit lower odds of dental attendance than did those in the highest category.

In Table 3, no longer were any countries found with lower odds of dental attendance for retired individuals compared with older persons still in the labour force. In fact, these results, from Table 2, were either reversed, as in the USA, Switzerland, Denmark and Italy, where retired persons were more likely to visit the dentist, or were not found in Table 3.

Health status need variables

Unlike the unadjusted results for all countries, a lower likelihood of dental attendance for persons in fair or poor health compared with those in very good or excellent health was found only in four countries (USA, Belgium, Denmark and Sweden) (Table 3). In Spain, those in fair or poor health even had higher odds of dental use than those with the best health.

The lack of a difference in the likelihood of dental attendance between older persons with and without dentures was no longer found in France, Germany, Sweden and Poland and in all the Southern countries after adjusting for other covariates. Dentate persons in the USA continued to have higher odds of dental use than edentate persons (Table 3).

Only Belgium, France, Denmark and the Czech Republic, among the European countries, continued to show lower odds of dental attendance for older persons with difficulty biting or chewing on hard foods in Table 3.

Dental coverage

The dental-coverage effects found in Table 2 were unchanged in Table 3 after adjusting for the other covariates in the model. Only in the USA, the Netherlands and Denmark were the odds of dental attendance higher for those reporting dental coverage compared with those without coverage. The opposite findings occurred in France, Poland and Italy, with higher odds of dental use for those without coverage, whereas there continued to be no further evidence of disparity in dental use based on dental coverage within the remaining countries in our study.

DISCUSSION

Despite finding more disparity in dental coverage with regard to population characteristics in non-SHI countries compared with SHI countries in a previous study, we did not find similar disparities in dental attendance across population characteristics between SHI and non-SHI countries²³. For example, dental attendance for those in the lowest education category was lower than for those in the highest education category in all but one SHI country (Austria) and in all but one non-SHI country (Greece), after adjusting for other influences on dental use. The adjusted results also show lower dental attendance at lower family-income levels in all but one non-SHI country (the Netherlands) and in all but two SHI countries (France and the Czech Republic).

In general, we found stronger, more pervasive, associations between dental use and income and education than between dental use and dental insurance across the European countries in our study. Because of the generally high percentage of persons with dental coverage in the SHI countries, it is not surprising that perceived coverage, as reported in SHARE, had little influence on dental attendance. It is surprising that those persons in the SHI countries France and Poland, who were aware of having dental coverage, had lower dental attendance than those who reported not having coverage in those countries. A similar result was found in Italy, a non-SHI country. Only in the non-SHI countries, the USA and the Netherlands, and in the SHI country, Denmark, did we find a positive influence of coverage on dental attendance, as one might expect. A possible explanation for these results might be that the type of dental care generally sought (treatment procedures, not preventive care, in France and Poland and non-emergency care in Italy) was not covered by insurance—see Table 1.

We did find some puzzling results after introducing covariates into our logistic regression models of dental attendance. Unlike the unadjusted results, older American and Swedish subjects became more likely to use dental care than those between 51 and 64 years of age. Lower rates of dental attendance for male subjects compared with female subjects were found in all but three SHI countries and in all but one non-SHI country, yet were virtually absent in the unadjusted results. Surprisingly, we found that older European persons with difficulty biting or chewing on hard foods were never more likely to use dental care than were those with no such difficulty, and in some cases were even less likely. Anxiety about receiving dental services and ease of access to dental providers are not directly measured in our data. Reasons for not seeking regular oral health care over a lifetime are collected in SHARELIFE wave 3, but no comparable data were available from the HRS for comparative purposes²². Absence of these relevant covariates may contribute to these unexpected findings.

In general, diversity in dental attendance with regard to population characteristics was not markedly different between the USA and European countries, but was often not as widespread among the latter. For example, the lower dental use observed for non-married older Americans was found for older non-married Europeans only in the SHI countries Germany, Denmark, Sweden and the Czech Republic and only in the non-SHI countries Switzerland and the Netherlands. However, the income and education patterns among older Americans were found in nearly every European country.

Compared with the Special Eurobarometer interview study among subjects 15+ years of age in 27 European countries, the overall dental-attendance rates in this study were somewhat lower in most countries except Denmark and Sweden, confirming earlier results of declining utilisation rates when age increases²⁴. This study also indicated that edentulousness explained part of the non-attendance, as could be expected from national epidemiological studies. This was, however, not true for Sweden, where a local study revealed that the proportion of edentulous individuals between 40 and 70 years of age decreased from 16% in 1973 to 1% in 2003²⁰. Furthermore, the Dental Register shows that only 0.9–2.2% of those between 60 and 79 years of age are edentulous in southern Sweden²⁶. These good results are the result of early public service starting in 1938 and generous insurance, since 1975, covering all kind of prosthetics.

The unexpected finding, of no significant differences in the degree of diverse dental attendance across population subgroups of older persons with and without social health insurance, can have several explanations. Self-reported assessments of service use, insurance and cost coverage, and certain demographic information, may suffer from cultural bias when compared across countries. In this study rather crude questions were used. All insurance systems in European countries are different²⁷. In some countries, the ‘insurance systems’ have undergone significant changes during past years regarding coverage of different population groups and treatment and amount of costs covered; in others, they have been relatively stable and may be also generous for decades. There are considerable differences between countries regarding coverage of prosthetic treatment²⁷. Thus, differences in resources and social history are likely to influence the results of this study.

Unobserved variables, correlated with dental use and observed covariates, may also limit, and possibly explain, some of our discrepant findings, as discussed above. It has been suggested that use of dental services in middle and older ages is related to corresponding habits in childhood²⁸. Use of services is also related to a wide range of dentist and dental practice-related factors²⁹. Health fund rebate periods may also have a steering effect. In Denmark and the Netherlands it is also possible to visit clinical dental technicians instead of dentists. The SHARE data did specifically ask about visits to a dentist or dental hygienist, although the HRS data only asked about visits to the dentist.

After harmonising covariates of dental attendance between two surveys of older persons – the HRS for the USA and the SHARE for Europe – we conclude that predisposing, enabling and need correlates of dental use have much in common across countries. Greater dental attendance is associated with older persons who are younger, female, married, dentate, well-educated and in higher-income households compared with respective older, male, unmarried, edentate, not well-educated counterparts in lower-income households. No differences in these effects were found between countries with and without social health insurance coverage. In fact, income and education stood out as more prominent correlates of dental use than self-reported dental coverage, which was only found to be a significant correlate of dental attendance in the USA, Denmark and the Netherlands. Our study implies that there are important attributes, other than dental insurance coverage, accounting for differences in dental attendance among an older population. For example, governmental policies designed to lessen disparities in education and income may also produce reductions in dental-use disparities as a population ages, regardless of the degree of public provision of dental health insurance within a country.

Acknowledgements

We gratefully acknowledge support from Michael Moldoff (RAND Corporation, Santa Monica) who helped with constructing the harmonised data file used for this study. The present paper uses data from RAND HRS and RAND SHARE. RAND HRS is a user-friendly version of HRS, and the HRS public-use data set is produced and distributed by the University of Michigan with funding from the National Institute on Aging (grant number NIA U01AG009740), Ann Arbor, MI. RAND SHARE is a user-friendly version of SHARE, created with funding from the National Institute on Aging (grant number R01AG030153). RAND SHARE used SHARE wave 4 release 1.1.1, as of March 28th 2013 or SHARE wave 1 and 2 release 2.5.0, as of May 24th 2011 or SHARELIFE release 1, as of November 24th 2010. The SHARE data collection has been primarily funded by the European Commission through the 5th Framework Programme (project QLK6-CT-2001-00360 in the thematic programme Quality of Life), through the 6th Framework Programme (projects SHARE-I3, RII-CT-2006-062193, COMPARE, CIT5- CT-2005-028857, and SHARELIFE, CIT4-CT-2006-028812) and through the 7th Framework Programme (SHARE-PREP, No. 211909, SHARE-LEAP, No. 227822 and SHARE M4, No. 261982). Additional funding from the USA National Institute on Aging (U01 AG09740-13S2, P01 AG005842, P01 AG08291, P30 AG12815, R21 AG025169, Y1-AG-4553-01, IAG BSR06-11 and OGHA 04-064) and the German Ministry of Education and Research as well as from various national sources is gratefully acknowledged (see www.share-project.org for a full list of funding institutions).

Statement of sources of funding for the study

This investigation was supported by the National Institute of Dental and Craniofacial Research of the USA National Institutes of Health (3R01DE021678-06S1).

Disclosure of conflicts of interests

The authors declare that there are no conflicts of interest.