Factors associated with satisfaction with dental services among the elderly

OBJECTIVE: To analyze the prevalence of elderly users’ satisfaction with dental care and associated factors. METHODS: Cross-sectional study conducted with 495 elderly individuals aged 65-74 years, without cognitive impairment according to the Mini-Mental State Examination, who used dental services of the municipality of Montes Claros (Southeastern Brazil) from 2008 to 2009. A two-stage cluster, complex probability sample was used. Trained/ calibrated professionals and students conducted tests and household interviews using handheld computers with software developed for this purpose. In the statistical analysis, with correction by design effect, PASW® Statistics 18.0 was utilized. To identify associations between satisfaction and the factors of interest, Poisson regression was conducted, estimating the prevalence ratio and the 95% confi dence interval (PR / 95%CI). RESULTS: Overall, 91.4% of the elderly were satisfi ed. Greater satisfaction was found among users of services of the Sistema Único de Saúde (SUS – Brazil’s National Health Service) (PR = 1.07; 95%CI 1.01;1.12); those who had access to information on how to avoid oral problems (PR = 1.09; 95%CI 1.02;1.16); those who did not use medicines (PR = 1.05; 95%CI 1.00;1.10); and those who rated the appearance of their teeth and gums as excellent or good (PR = 1.13; 95%CI 1.00;1.28). CONCLUSIONS: The elderly who access the service offered by SUS have greater satisfaction with dental treatment, as well as those who do not use medicines, those with better self-perception of the aesthetics of teeth and gums and those who have more access to information about how to prevent oral problems. Therefore, health care policies in the SUS have been presenting positive results. DESCRIPTORS: Aged. Consumer Satisfaction. Dental Care for Aged. Dental Health Services. Cross-Sectional Studies. Original Articles Satisfaction of the elderly with dental services Rodrigues CAQ et al Evaluating the quality of healthcare is an important tool in the planning and management of health services and systems.23 Analysing satisfaction with health services from the service users’ perspective allows healthcare to be better adapted and promoted towards their real needs. Technical evaluations should be complemented by the perceptions of satisfaction of the individuals who receive care with the services provided.5,16 There has been little investigation into subjective aspects of the quality of the programmes and services, including evaluating consumer satisfaction, and this area includes issues that need to be discussed which may be peculiar to each place and setting.21 Interest is increasing in knowing the population’s expectations with regards to the services offered in the development of dental care programmes and strategies. The technical characteristics regarding care itself are used to measure results. Service user satisfaction with their relationship with health service providers should be considered.2,4 Donabedian3 (1988) proposes that health services be evaluated according to structure, the work process, results and user satisfaction. The determinants of user satisfaction with dental services, according to the theoretical model by Andersen & Davidson1 (1997), may infl uence the user’s perception of health services. One of the outcomes for oral health in this model is consumer satisfaction with access, communication and quality of services. Andersen & Davidson1 (1997) divide them into a) exogenous variables (immutable characteristics), belonging to a particular ethnic group or age group; b) primary determinants of oral health the external environment, the system of oral health care and personal characteristics; c) oral health habits, personal habits and use of offi cial dental services and d) oral health outcomes, normative and subjective oral health conditions, in addition to the consumer’s own satisfaction.1,11 Faced with the aspects which make up the evaluation of dental care from the users’ point of view, possibilities of transforming dental health care based on reorganising the work process appear. This includes soft technologies to mediate encounters of health professionals among themselves and with the population using these services. Historically, dental care in Brazil has focussed on treatment, on compartmentalised and fragmented knowledge of the act of healing17 without considering the service users’ point of view in evaluating the service. The dramatic aging process experienced by the Brazilian population and the National Policy for the Elderly highlight the need for expanding the debate on health care for this age group.13 The Statute for the Elderly emphasizes the need for comprehensive health INTRODUCTION a Ministério da Saúde (BR). Statute for the Elderly. Brasília (DF); 2003. care for the elderly through the Unifi ed Brazilian Health System (SUS), guaranteeing universal and equal access in integrated and continuous actions and services for the prevention, promotion, protection and recovery of health.a Dental care for the elderly in Brazil is not impressive, with individuals with precarious oral health conditions, high levels of tooth loss and increasing demand for dentures, which compromises this age group’s quality of life.11 Evaluating dental care from the perspective of the elderly may contribute to improvements in the availability of such services and in the quality of life of the service users. Therefore, this study aims to identify the prevalence and the factors associated with the elderly’s satisfaction with their dental care.

Evaluating the quality of healthcare is an important tool in the planning and management of health services and systems. 23 Analysing satisfaction with health services from the service users' perspective allows healthcare to be better adapted and promoted towards their real needs. Technical evaluations should be complemented by the perceptions of satisfaction of the individuals who receive care with the services provided. 5,16 There has been little investigation into subjective aspects of the quality of the programmes and services, including evaluating consumer satisfaction, and this area includes issues that need to be discussed which may be peculiar to each place and setting. 21 Interest is increasing in knowing the population's expectations with regards to the services offered in the development of dental care programmes and strategies. The technical characteristics regarding care itself are used to measure results. Service user satisfaction with their relationship with health service providers should be considered. 2,4 Donabedian 3 (1988) proposes that health services be evaluated according to structure, the work process, results and user satisfaction. The determinants of user satisfaction with dental services, according to the theoretical model by Andersen & Davidson 1 (1997), may infl uence the user's perception of health services. One of the outcomes for oral health in this model is consumer satisfaction with access, communication and quality of services. Andersen & Davidson 1 (1997) divide them into a) exogenous variables (immutable characteristics), belonging to a particular ethnic group or age group; b) primary determinants of oral health the external environment, the system of oral health care and personal characteristics; c) oral health habits, personal habits and use of offi cial dental services and d) oral health outcomes, normative and subjective oral health conditions, in addition to the consumer's own satisfaction. 1,11 Faced with the aspects which make up the evaluation of dental care from the users' point of view, possibilities of transforming dental health care based on reorganising the work process appear. This includes soft technologies to mediate encounters of health professionals among themselves and with the population using these services.
Historically, dental care in Brazil has focussed on treatment, on compartmentalised and fragmented knowledge of the act of healing 17 without considering the service users' point of view in evaluating the service. The dramatic aging process experienced by the Brazilian population and the National Policy for the Elderly highlight the need for expanding the debate on health care for this age group. 13 The Statute for the Elderly emphasizes the need for comprehensive health care for the elderly through the Unifi ed Brazilian Health System (SUS), guaranteeing universal and equal access in integrated and continuous actions and services for the prevention, promotion, protection and recovery of health. a Dental care for the elderly in Brazil is not impressive, with individuals with precarious oral health conditions, high levels of tooth loss and increasing demand for dentures, which compromises this age group's quality of life. 11 Evaluating dental care from the perspective of the elderly may contribute to improvements in the availability of such services and in the quality of life of the service users. Therefore, this study aims to identify the prevalence and the factors associated with the elderly's satisfaction with their dental care.

METHODS
This is a cross-sectional study of 495 elderly people aged between 65 and 74 years old, in the municipality of Montes Claros, MG, between 2008 and 2009. The study was carried out as part of a survey of the population's oral health from 2008 to 2009 ("SBMOC Project"), with a two-stage cluster complex probability sample. The methodology suggested by the World Health Organisation (WHO) was used, with the alterations proposed in the epidemiological survey of the Brazilian population's oral health. 11 We deemed the estimate for incidence of events of illness to be 50% of the population, error of 5.5%, no-response rate to be 2.0, ensuring proportionality by sex and deff (design effect) to be 2.0. 11 The clusters were randomly selected. We estimated a sample of 740 elderly people. Of those who agreed to take part, the study was composed of those who reported using dental care and who showed no defi ciency in cognitive capacity, tested using the Mini Mental test version validated for Brazil, with total scores adjusted for the level of Education of the testee. 8 The household interviews and examinations were conducted by trained interviewers and examiners. The data collection was carried out by 24 dental-surgeons who participated in special training (Kappa inter/intra examiners and intra-class correlation coeffi cient ≥ 0.60). This step involved the use of handheld computers with a program designed to collect data and simultaneously construct the database, called the Program in Health Data Collector (PHDC). 12 The examinations took place in a spacious, well-illuminated environment, using sterile mouth mirrors and probes approved by the Community Periodontal Index (CPI). The dependent variable (evaluation of dental services) was obtained through the questions: Have you ever visited the dentist? How long ago? Where? Why did you go to the dentist? How would you rate the care? The variable was dichotomized based on the responses to this last question: dissatisfi ed (bad, poor or regular) or satisfi ed (good or excellent).
The main covariate considered was the type of service (SUS and other services -private / covered by health agreement / charity). The other independent variables were placed into three groups, with their respective sub-categories, defi ned in the theoretical model previously mentioned: primary determinants of oral health (external environment, oral health systems, personal characteristics), oral health habits (personal habits and use of offi cial dental services and oral health outcomes (normative and subjective oral health conditions. 1,11 The exogenous variables in the Andersen & Davidson 1 (1997) theoretical model, which concern the age group and the ethnic group to which interviewees stated they belonged (self-reported race), were considered as personal characteristics in the primary determinants of oral health group.
In the external environment area, reported general state of health (reported chronic illness or being on medication), the infl uence of general health on quality of life -using the 12-Item Short -Form Health Survey (SF12) version validated for Brazil, with weighted scores for physical and mental health -and satisfaction with life. Individuals with SF12 higher than 50 and 42 respectively in the physical and mental areas were considered, according to this instrument, to have satisfactory quality of life. b We analysed the type of dental service used in the system of oral health care and statements of the household being included in the Family Health Strategy (FHS).
We deemed personal characteristics to be: age (dichotomised based on the median -68 years old), selfreported race, sex, marital status, level of education (number of years of studies, categorised according to the education system at the time of the study), per capita income and reason for needing dental care.
Personal habits included information on oral hygiene (frequency and method used), oral self-examination and current and past tobacco and alcohol consumption.
The analysis of the use of official dental services covered access to information about preventing dental problems and frequency of visits to the dentist. The covariates of the normative condition of oral health included changes in oral soft tissues (mucous membrane), edentulism, using dentures, The DMFT (decayed, missing and fi lled teeth) index and normative need for dental treatment, created from the variables: need for dental treatment (restoration of a tooth surface, restoring two or more tooth surfaces, crown for any reason, veneers, pulp treatment and restoration, extraction, remineralisation of white spot lesions and dental sealant); need for periodontal treatment, defi ned based on the Community Periodontal Index (CPI) ≥ 1 (bleeding on probing) and Periodontal Attachment Loss Index (PALI) ≥ 1 (4 to 5 mm); and the need for dentures (fi xed or removable, partial or total). Those elderly people suffering from one or more of the above were considered to have a normative need for dental treatment. 24 The subjective condition of oral health were evaluated by self-perceived oral health, mastication, the appearance of the teeth and gums, speech as affected by the teeth and gums and the need for dental treatment, reporting having felt pain in the teeth and gums in the previous six months and the perceived impact of oral health on the physical and psychosocial dimensions, using the Oral Health Impact Profi le (OHIP-14) validated for use in Brazil, categorised by absence or not of impact. Those individuals who responded in the affi rmative (rarely / sometimes / often / always) to one or more of the 14 questions were classed as experiencing physical or psychosocial impact of oral health.
We used PASW ® Statistics 18.0 software to analyse the data. As the study involved complex cluster sampling we adjusted it through the design effect by weighting. The descriptive analysis included absolute (n) and relative (%) frequency and relative frequency adjusted for design effect (% * ) for categorical variables, mean and standard error (SE), adjusted for design effect for the quantitative variables. We used Poisson regression and adjustment for the study's design effect to estimate the adjusted and unadjusted ratio of prevalence in the bivariate and multivariate analysis of the correlation of the dependent variable and the 95% confi dence intervals (95%CI) adjusted for design effect.
The study adhered to the National Health Council (NHC) ethical principles resolution (n° 196/96) and was approved by the Research Ethics Committee of the Universidade Estadual de Montes Claros (n° 318/06).

RESULTS
The invitation to participate in this study was made to 800 elderly people living in randomly selected clusters, of which 92% accepted and were evaluated. Of these, 735 responded to the question on using dental services and 1.6% were not included as they reported never having made use of this service. The cognitive capacity questionnaire was completed by 632 eldery people and 20.7% were not included as they demonstrated cognitive problems. The study included 67.25% of those elderly people invited to take part who responded and met the two criteria for inclusion.
Taking into account adjustment for design effect, 91.4% of the elderly people questioned stated they were satisfi ed with dental care: 0.7% judged it to be very bad; 1.2% to be bad, 6.8% as regular, 67.1% as good and 29.7 as excellent. The majority of the elderly people did not use services offered by the SUS and reported living in a household included in the FHS. Mean age was 68.4 years old (SE = 0.15), the greater part were female, self-reported black, mixed, indigenous or yellow, stating per capita income of minimum wage or below (67.2%) and with access to information on avoiding oral health problems. Mean DMFT was 28.4 (SE = 0.41) (Tables 1 and 2).
In the bivariate analysis (p ≤ 0.05), satisfaction with dental care was statistically associated with the following variables: chronic illness; being on medication; type of service and access to information on avoiding oral health problems. In the multivariate analysis, the prevalence of satisfaction with the dental care, independent of the other variables, was greater among those elderly people who used the SUS services, who had access to information on avoiding oral health problems, who reported they were not on medication and who classifi ed the appearance of their teeth and gums as excellent or good (Tables 3 and 4).

DISCUSSION
The analysis of the elderly people's satisfaction with their dental care is a sensitive indicator of the quality of service provided, able to support the planning of programs and policies for health and in greater use of services, because it is a measure established by the user. The effectiveness of the health care is determined, up to a point, by the service user's satisfaction with the service provided, and a satisfi ed consumer is more likely to make use of the health service, follow the recommended treatment and stay in contact with the care provider. 20,23 Consumers' satisfaction with the care received is considered to be a parameter for analysing health service outcomes. 5  The level of satisfaction with dental services among the elderly people we evaluated was high, and higher still with those who accesses services offered by the SUS. A high level of satisfaction has previously been observed in both national and international studies, 9,14,16,20 independently of the form in which it was conceived and measured, whether from a quantitative or qualitative approach. 5 In this study, of those assisted by SUS, 95.4% were satisfi ed, similar to the value recorded among service users (89%) of the SUS in Belo Horizonte, MG, in the study by Lima-Costa & Loyola Filho. 9 However, in this study, the variable 'satisfaction with the health service used' was constructed based on the respondent stating they would recommend this service to others.
The higher levels of satisfaction among users of the SUS contrasted with possible infrastructure defi ciencies still present in the health system which could, according to Donabedian 3 (1988), negatively infl uence service users' evaluation of the services. 22 On the other hand, possible improvements in access to services and in their quality, due to the oral health policies implemented over the last few years, may have generated feelings of satisfaction. The apparent improvement in access may trigger satisfaction with the dental care itself. In the studies by Martins et al 10 and Paim et al, 15 5.8% of the elderly people reported never having used such services; in this investigation, 1.6% said this.
It should be highlighted that this work took place in the home, not in the environment in which care is delivered by the dental care provider, which minimises possible feelings of recognition and consequent satisfaction. Some health and professional development policies in the municipality such as investments in professional training and qualifi cations, primary health care and giving a more human face to care, as well as wide social participation and greater awareness of the right to health care, may have infl uenced the evaluation of dental services provided to the elderly by the SUS. 7,15,22 The elderly relate higher levels of satisfaction than those of younger service users. 9,16,23 A population based national survey, incorporating fi ve thousand Brazilian households, with adults over the age of 18, chosen by random sampling, showed that 27% of the individuals were satisfi ed with the functioning of the country's health care, without distinguishing the type of service used. 19 Marked socioeconomic inequalities, as well as the infl uence of social circumstances on the health of Brazilians, might be directly linked to divergent results, mainly due to regional differences in health care policies. 6 There were higher levels of satisfaction with dental care among those who reported they were not on medication.
These results are consistent with those of other studies which reported lower levels of satisfaction with the organisational quality of care among consumers with health problems. 9,16,23 According to the adopted theoretical model from Andersen & Davidson 1 (1997), and that proposed by Donabedian 3 (1988), we expected correlations between satisfaction with dental care services and level of education, income and skin colour according to user's self-reported race. However, in this study and in another carried out in Taiwan, the service users' levels of education did not correlate with satisfaction with care received. 20 In South Africa, both socioeconomic status and race were found to be associated with satisfaction. 14 The expected correlations were not confi rmed in this research, possibly due to the homogeneity of the population's income (82.0% earning minimum wage or less per month per capita).
Access to information on avoiding oral health problems has been independently associated with satisfaction   18 This could be explained in future research. It is possible that the differentiated training of these professionals has brought about more humanised health care and affected the way users are received, benefi tting the patient-professional-service interaction.
We expected lower levels of satisfaction with dental care among those who experienced the negative impact of oral health problems on their quality of life, as this satisfaction could be infl uenced by health outcomes. 3 However, the normative situation of oral health we found may refl ect factors related to the conditions of self-care and lifelong dental care. Consequently, it may not imply satisfaction with recently received dental care.
The higher level of satisfaction found among the elderly who had a positive perception of the appearance of their teeth and gums indicated the importance of this parameter for users' perception of their own health in relation to services received. 1,11 There was independent correlation with access to information on avoiding oral health care problems. SUS service users and those who have a positive perception of their own appearance showed higher levels of satisfaction. In the model proposed by Andersen & Davidson 1 correlation was found between variables from the four sub groups. This theoretical model 1 and that proposed by Donabedian 3 (1988) were both confi rmed in this research.
The elderly who used SUS dental services were more satisfi ed with the care, as were those who were not on medication, those who perceived their own teeth and gums in a positive way and those who reported having more access to information on avoiding oral health problems. We concluded, therefore, that the SUS health policies implemented in the municipality have started to yield the hoped for positive results.
The limitations of this study lie in the construction of the dependent variable and the lack of assessment of the resolution and the structure of health services. We would also like to point out that the process of evaluating dental services and the variables investigated is dynamic. Therefore, cause and effect will certainly vary with the passage of time and, as this is a cross-sectional study, it is not possible to establish a temporal relationship between the associations observed. On the other hand, it is noteworthy that the results are reliable; being a result of random sampling of clusters representative of the elderly of the municipality, and the data analysis was carried out with adjustments for design effect.