Dissatisfaction with the dental services and associated factors among adults

This study aimed to identify factors associated with dissatisfaction with dental services among adults. It analyzed 830 adult participants of an epidemiological survey of oral health. The dependent variable was dissatisfaction with the dental service, and the independent ones were selected according to the theoretical model set forth by Andersen and Davidson (1997). Estimates were corrected by the sample design effect, and Binary Logistic Regression was carried out. It was found that about 11% of adults were dissatisfied with the dental service. In the final model, dissatisfaction with dental services was lower among older adults (OR = 0.559) and among smokers (OR = 0.332). On the other hand, it was higher among adults who self-perceived their chewing as negative (OR = 2,804), who self-perceived some discomfort in the mouth and head and neck region (OR = 2.065), and among those who did not have access to information on how to avoid oral problems (OR = 3.020). Therefore, the services need to access the perceptions and expectations expressed by users, and provide information in appropriate quantity and quality, in the context of “health literacy” in order to achieve greater satisfaction among its users.


Introduction
Health services' evaluation by users provides essential information for the definition of the quality standards of care provided.It has been highly recognized 1 and is an expected care outcome 2 .User satisfaction is a fundamental component of health services' evaluation and is even considered by some authors as the ultimate goal of such services 2,3 .Donabedian's 2 proposal is classically used when evaluating the quality of health services.It argues that evaluation should consider three basic components: structure (resources used by the service), process (procedures used to manage patients' problems) and outcome (whether or not the patient benefits from the health service provided).Patient satisfaction is an important outcome targeted by services 4 .
User's incorporation into the evaluation has been highly valued, especially within the scope of the Unified Health System (SUS), which seeks to encourage community participation in planning and evaluation processes to strengthen social control.Thus, the user's perspective reveals essential aspects to complete and balance the quality of services 5 .The perception of users' dissatisfaction can direct health services to meet more specifically the demands of the population, as it allows complementing care quality technical assessments with a shared view of the individuals receiving treatment 6,7 .
In addition, satisfaction assessment has been identified as a key element for successful treatment.Unsatisfied patients adhere less to the proposed treatment 8 and are less likely to continue using health services 1,9 .
Allied to the importance of understanding what leads individuals to express dissatisfaction with services used is the collection of epidemiological data on adults' oral health, since the interest in these data has greatly increased 10 .Adults are the majority of the population; they demand dental services, decisively influence the behavior of their dependents, have specific oral health problems and epidemiological peculiarities 11 .Workers, in general, may have difficulty accessing such services during working hours, causing aggravation of the problems and reasons for work absenteism 11 .
Users' dissatisfaction with dental services is still a little explored field.It certainly has many aspects to be unveiled, since it can carry particularities of each context and setting 12 .Thus, this study evaluated from a recognized multidimen-sional theoretical model the factors associated with adult dissatisfaction with dental services.

Methodology
This is a cross-sectional study conducted among adults participating in the epidemiological survey on oral health carried out in the city of Montes Claros, Minas Gerais, Brazil, in 2009.Montes Claros is the main urban center in the north of the State of Minas Gerais, and, for that reason, shows characteristics of regional capital, with radius of influence that covers all the north of Minas Gerais and part of the south of Bahia.Currently, the municipality has 71 Oral Health Teams (ESB) linked to the Family Health Strategy (ESF), of which 58 are modality I and 13 modality II, and a Dentistry Specialty Center (CEO) type II 13 .
Sampling: Complex probabilistic samples by conglomerates in two stages (census tracts and blocks), ensuring proportionality by gender.For the population aged 35-44 years, standard group for the evaluation of oral health conditions in adults 14 , the calculations evidenced the need to evaluate 762 individuals, considering the occurrence of events or diseases in 50%, 95% confidence level, sample error of 5.5%, deff (design effect) equal to 2.0 and non-response rate of 20%.The primary sample units were randomly selected, with 53 of the 276 urban census tracts and two of the 11 rural areas selected by draw.In a second stage, approximately seven blocks were drawn in each included urban sector.In the rural area, all households located at a distance of up to 500 meters from a reference institution were selected 15 .All households in the selected areas were sequentially visited and adults (35-44 years old) were invited to participate.
Calibration of examiners: Twenty-four trained and calibrated dental surgeons participated in data collection and achieved satisfactory Kappa agreement (Kappa inter / intra-examiner and intraclass correlation coefficient ≥ 0.60); they were accompanied by trained note takers / typists.The diagnostic criteria of the fourth edition of the Oral Health Surveys: Basic Methods, the World Health Organization (WHO) were implemented 14 .
Data collection: At the households, after signing the informed consent form, interviews and intrabuccal examinations were performed.Examinations were conducted under natural light, using a mirror and a previously sterilized probe indicated for evaluating the Community Periodontal Index (CPI), using all codes/criteria proposed by the WHO 14 .Data were recorded on handheld computers using a program specifically created for this purpose.More details about the methodology adopted are found in a previous study 16 .

Analyses performed
Analyses were performed using the PASW (Predictive Analytics Software -SPSS ® ) version 18.0 and were conducted respecting the need for correction for the design effect since they derive from samples by conglomerates.Such correction refers to the calculation using different weights for sample elements in order to offset their unequal selection probabilities 17 .
The dependent variable -satisfaction with the dental service -was obtained by answering the following question: "Were you satisfied with the dental service you used last time?"(Extremely, very, fairly, little, not satisfied, never went to the dentist).Individuals who reported never having gone to the dentist were excluded from the analyses.Answers were aggregated into two categories: satisfied (extremely and very) and dissatisfied (fairly, little and not satisfied).This aggregation allowed investigating dissatisfaction with the use of dental services.
The independent variables were grouped using the multidimensional theoretical model of Andersen and Davidson 18 .This model was the most widely used for the analysis of factors independently associated with the use of dental services 19 , and it was also used to evaluate factors related to satisfaction with the use of these services, since satisfaction appears as an outcome related to the use of services.Authors 18 argue that independent variables can be gathered into three groups: primary determinants of oral health (external environment, oral health care system and personal characteristics), oral health behaviors (personal practice, formal use of dental services) and oral health outcomes (normative and subjective oral health conditions).
The exogenous variables of the theoretical model by Andersen and Davidson 18 , which refer to the report of belonging to an ethnic group (self-declared ethnic group) and to an age group, were considered as personal characteristics in the group of primary determinants of oral health.The following variables were also included as personal characteristics: gender, marital status, schooling, per capita income, current employ-ment situation and reason for the last dental visit.
The context of the external environment considered the reported general health (chronic diseases and use of drugs) and the influence of general health on quality of life, using a Brazilian version of the 12-Item Short-Form Health Survey (SF12), with weighted scores for the Physical and Mental Realms.The lower limit of the CI-95% of the estimated mean was used as the cutoff point in each realm 20 , and individuals who obtained scores below the cutoff for each realm separately were considered as having a poor quality of life for the realm concerned.Satisfaction with life, while collected as a Likert scale, was dichotomized.Regarding the oral health care system, we analyzed the type of dental service used and the report of the insertion of the household in the ESF.
In the group of oral health behaviors, personal practices included information about oral hygiene, oral self-examination, and current and past smoking and alcohol habits.The formal use of dental services included access to information on preventing oral problems, oral hygiene, diet, oral cancer and how to perform oral self-examination.In addition, we analyzed the use of the dental service in the previous year.
In relation to the oral health outcomes, we evaluated normative conditions, including oral mucosal alterations, tooth count, use of dental prostheses, DMFT (decayed, missing and filled teeth) index, periodontal disease and normative need for dental treatment 14 .Periodontal patients were those with a periodontal pocket ≥ 4 mm and loss of insertion ≥ 4 mm 21 in the same sextant.
Oral health subjective conditions were represented by self-perception of oral health, chewing, the appearance of teeth and gums, speech due to teeth and gums, relationship as a result of oral condition, some discomfort in the mouth, head and neck, the need for dental treatment, the report of tooth and gum pain in the last six months and the evaluation of the impact of oral health on its physical and psychosocial realms, through the Brazilian validated version of the Oral Health Impact Profile (OHIP-14).The responses to each OHIP-14 question were dichotomized in no impact (sometimes, rarely, never) and with impact (always, often), and the individual who reported an impact on at least one item was considered to have been impacted 22 .
All variables were categorically worked out.The absolute (n) and relative (% corrected for sample design) frequencies were obtained through descriptive analysis.In addition, mean values and standard error were estimated for income and schooling variables.Bivariate analyses were performed using the Pearson chi-square test.Variables with "p value" of less than or equal to 0.2 were selected and included in the multiple model.The multiple models were adjusted through binary logistic regression, adopting the stepwise backward procedure, estimating the odds ratio and 95% confidence interval.In the final model, only the variables that showed significance level less than or equal to 0.05 (p≤0.05) were maintained.We also estimated the pseudo R-squares (R 2 ) in order to measure the capacity of the adjusted final model to explain the variation of the dependent variable.

Ethical Issues
This study was conducted in accordance with the ethical principles of the Resolution Nº 196/96 of the National Health Council (CNS) and was approved by the Research Ethics Committee of the State University of Montes Claros.Participants were duly informed about the research and agreed to participate by signing the informed consent form.

Results
Overall, 841 adults were evaluated as they resided in the selected census tracts.Among the adults evaluated, 11 (1.3%) were excluded from the study because they had never been to the dentist.Thus, 830 subjects were included in the analyses.Of these, 91 (10.8%) were dissatisfied with their last dental service.In the descriptive analysis, we noticed that most were female and self-declared indigenous, black or brown.The average per capita income was US$ 637.56 (EP=21.50).The investigated adults studied a mean of 9.01 years (EP=0.34).Most adults used dental services not provided by the SUS, reported living in a household inserted in ESF and had access to information on how to prevent oral problems (Table 1).
In the bivariate analysis, we observed that dissatisfaction with dental services showed a relationship (p ≤ 0.20) with variables age group, self-declared ethnic group, gender, per capita income, current employment situation, type of dental service used, current or past smoking habit, normative need for dental treatment and with most of the variables of access to information.In addition, all variables that reported subjective oral health conditions were associated (Table 1).
In the adjusted multiple analysis, the odds of dissatisfaction with dental care were lower among older adults (borderline p-value) and among those who smoked.On the other hand, the likelihood of dissatisfaction was greater among individuals who did not have access to information on how to prevent oral problems, which negatively self-perceived their chewing (fair, poor or bad) and self-perceived some discomfort in the mouth, head and neck (Table 2).Adults who did not receive information about how to avoid oral problems were three times more likely to be dissatisfied with the dental services used (OR = 3.020) than those who received such information.The adjusted final model accounted for 16.1% of the variability of the dependent variable.

Discussion
This study revealed a lower likelihood of dissatisfaction with dental services used among older adults with a history of current or past smoking.On the other hand, dissatisfaction was greater among individuals who reported not having been informed about how to prevent oral problems, which negatively self-perceived their chewing and self-perceived discomfort in the mouth, head and neck.In general, there was a low prevalence of dissatisfaction (10.8%) with these services in the adult population.Previous studies also found a low prevalence of dissatisfaction similar to the findings of this study, considering South African families (11%) 23 and adult users in public services in Belo Horizonte (MG) (11%) 24 .On the other hand, the prevalence of dissatisfaction was lower than that observed among respondents of all ages to a telephone interview in Taiwan, China (19%) 25 and among Brazilian adults participating in the National Oral Health Survey (SB Brazil 2010 ) (14.7%) 26 .While it has already been shown that users feel more comfortable to report dissatisfaction 27 outside the care setting, as is the case of this study (household collection), nevertheless, the prevalence of observed dissatisfaction may be considered low.Historically, adults have been almost systematically excluded from oral health agendas at the collective level 11 .The increasing access of adults to these services 26,28 could explain, in part, the high prevalence of satisfaction, since the mere fact of obtaining care could be sufficient to predispose to a better user evaluation vis-à-vis the service used.
Among the adults investigated, only 1.3% had never been to the dentist, a prevalence lower  than that among Brazilian adults (7.1%) 26 , suggesting greater access in Montes Claros (MG) and corroborating the hypothesis that greater access would possibly contribute with greater satisfaction.In addition, the use of the routine/ maintenance dental service, also known as use for review and/or prevention, was higher among adults surveyed (32.9%) when compared to Brazilian adults (21.4%) 26 .Routine use represents an important indicator of oral health 29 , and thus, better oral conditions are expected in patients who use preventive and regular dental services and could contribute to higher levels of satisfaction.
While the theoretical multidimensional model of Andersen and Davidson 18 predicts a possible relationship between satisfaction with dental services and sociodemographic factors, this relationship was not found in this investigation, except for age, which was associated with dissatisfaction with the use of dental services with a probability of borderline significance (p = 0.056).Other studies also found no such association 25,30 .However, a study carried out among patients serviced at medical centers in Sweden aged 20 years or older revealed that individuals with lower educational level were more satisfied with the care given 31 .Ethnic group and socioeconomic status were significantly associated with satisfaction with health services in research conducted in South Africa with families in their respective households 23 .In this investigation, while the income variable was shown to be associated with dissatisfaction with the use of services in the bivariate analysis, this association did not remain in the multiple model.It is necessary to consider the highly homogeneous studied population in relation to socioeconomic factors, since most had per capita income equal to or less than US$828.00(78.7%) and 5 to 11 years of schooling (60.6%).
It should be noted that the satisfaction of individuals was not influenced by the nature of dental services used.Public services (SUS) are ensuring high prevalence of satisfaction to their users (87%), close to that found among users of non-public services (90%), and this difference is not statistically significant.A previous study, conducted among the elderly, found greater satisfaction with dental services among users of SUS 30 services.These findings reflect the positive effects of public oral health care strategies.
On the other hand, it was observed that only 35% of the adults used SUS dental services, although 53% lived in areas covered by the ESF.In a complementary analysis using the chi-square test (data not shown), we found that, among individuals who live in areas covered by the ESF, 41.4% used dental services from SUS, while among the adults residing in areas not covered by the ESF, only 27% used SUS dental services (p <0.001).Thus, it is evident that such coverage facilitates, but does not ensure the use of public services.Access problems may be difficulties that prevent higher prevalence of public service use, even in areas covered by the ESF.One should consider that younger adults have difficulty accessing health units during conventional working hours 11 , and the provision of care at alternative times could contribute to the higher prevalence of use of these services.
In the final model, we observed that younger adults were more dissatisfied with dental services when compared to older adults.Since such relation evidenced borderline p-value, we chose to keep such variable in the final model.In addition, a previous study conducted on elderly people in the same region showed lower prevalence of dissatisfaction among the elderly aged 65-74 years 30 .A study conducted in Sweden with people over 20 years of age revealed that older people were generally more satisfied than younger people.A lower proportion of dissatisfied patients (10%) was observed among the elderly, especially those who reported better health status 31 .This inverse relationship between age and dissatisfaction with the use of dental services was also verified among Brazilian adults 26 .
Current or past smoking habits were independently associated with dissatisfaction with dental care, with smokers being less dissatisfied than nonsmokers are.A study conducted among SUS users in the Metropolitan Region of Belo Horizonte (MG) found an inverse relationship, that is, greater dissatisfaction among current smokers 24 .However, in the quoted study, the recommendation of health care received to another person 24 was an indicator of satisfaction with the services used.In this context, the lack of standardized instruments implemented to evaluate satisfaction with the use of health services represents a difficulty that may even compromise the comparison of results of different studies conducted on the same topic 32 .
The independent factor most strongly associated with dissatisfaction with health services was the lack of access to information on how to avoid oral problems from services.The likelihood of dissatisfaction among adults who did not have access to this information was about three times that observed among adults who reported access to this information, which is the main finding in this study.Such a relationship has already been demonstrated in previous studies 25,30 .However, no studies have been identified that have evaluated this relationship among some Brazilian adult population.
It has already been verified that the provision of information is a key factor among health education strategies 33 .In dentistry, health education has been an important part of health care, and health services were the setting of this practice.The impact of these educational measures can be observed in patients' health status and in user satisfaction 34,35 .The approaches used to carry out health education vary from the simple provision of information to the use of complex programs that involve behavior change strategies 34 .Although this study did not investigate the methodologies used, it was still possible to observe that the simple report of having had access to information on how to prevent oral problems had an impact on satisfaction with the health services used.
It should be noted that, although the question regarding access to information on how to prevent oral problems does not explicitly indicate the origin of such information, it was placed in the block of issues "Access to dental services", exactly as carried out in SB Brazil 2003/2004 project 36 .In this context, we asked respondents the following questions: Have you ever been to the dentist?How long?Where at?Why?How do you rate the service?Did you receive information on how to avoid oral problems?Thus, following the sequence of questions asked, we observed that the question was assessed within health services.The clinical environment is an important learning setting, giving users greater opportunities for clarification and seeking to ensure that all have access to the resources necessary to make dental care effectively a human right 37 .
Informed individuals seem to use dental services in a timely manner, in the face of a poor or urgent dental condition 38 .Thus, by not offering information, health services end up contributing and even reproducing health inequities 39 .In this regard, we should emphasize that these services must fulfill their role of providing information in adequate quantity and quality, contributing to the strengthening of equity.In addition, more informed and consequently more satisfied users tend to become more confident about services as well as themselves, increasing their autonomy 40 and strengthening the user/professional/health service link.
Faced with the need of health services to promote the education of its users, several method-ologies have been proposed."Health literacy" has recently emerged as an outcome to be pursued by such strategies, as it refers to the outcomes of health education and communication activities, with a view to enhancing individuals' ability to use it in a practical and efficient way such information (empowerment) 41 .The findings of this work should be interpreted in this realm.The health services increasingly need to offer quality information, aiming to improve and increase users' capacity to tackle health difficulties.Empowering individuals via the provision of health information contributes to the process of the subject's self-transformation, providing more autonomy to individuals involved 40 .
Among adults investigated, 60.4% had access to information on how to avoid oral problems.A study carried out in Taiwan to verify whether perceived clinical quality and patient education interfere with satisfaction with the health services used revealed that 76.3% had access to information on disease prevention and control 25 .This high prevalence of access to information in health services may contribute to explain the high percentage of satisfaction observed in both studies.While it is recognized that simple access to information is not decisive for inducing behavioral change, the provision of information is fundamental, as the first step, in achieving the best levels of health literacy 42 .
Communication between professional and patient is a primordial aspect when it comes to the provision of health services information.The health professional becomes an essential element since he/she serves as an intermediary to the communication process 40 .A study conducted by Donabedian 2 found that the evaluation of health services addresses two main realms: technical performance and personal relationship with the patient, and most studies attribute satisfaction to humanitarian aspects of the professional-patient relationship, among which quantity and quality of information received stands out 43,44 .Therefore, satisfaction is embedded in the relational component between users and professionals 2 .
In a qualitative study carried out with users of the oral health service in the municipality of Grão Mogol (MG) 45 , respondents emphasized the importance of the interpersonal professional-patient relationship, which could also influence the outcome of the treatment.The patient-dentist relationship has been addressed by several studies 6 , and some have revealed that aspects related to education, cordiality, gentleness and communicability play a more important role in relation to patient satisfaction than proper professional technical competence 6,45 .
The normative conditions of oral health were not shown to be associated to satisfaction with the use of dental services.While this association was expected in the model of Andersen and Davidson  (1997)  18 , a previous study conducted among the elderly also found no such association 30 .Characteristics related to self-care conditions and dental care throughout life may influence the normative situation of oral health, which may not cause dissatisfaction with the last dental care 30 .
On the other hand, subjective oral health conditions were associated with satisfaction with dental services.Adults who negatively self-perceived their chewing (fair, poor or bad) and those who self-perceived discomfort in the mouth, head and neck were more dissatisfied with the use of services.Subjective issues negatively perceived possibly reflect users' longings/expectations that were not accessed and/or met by the service and thus manifest as dissatisfaction.Other studies have already shown that individuals with worse perception of their own health are more dissatisfied with the health services provided 24,30,31 .The lower dissatisfaction among adults with a positive perception of their oral condition evidences the importance of the patients' view of their own health when they evaluate the quality of the health services used 18,19 , and this view seems to be more affected by subjectively perceived symptoms than by objectively observed signs.
It should be noted that among the factors that have been shown to be associated with dissatisfaction in this research, the provision of information and the evaluation of the subjective questions of self-perception and user expectation are among the list of competencies applicable to services, that is, they are subject to be altered by it.Thus, services need to be aware of such issues in order to achieve higher levels of satisfaction for their users.
Regarding the limitations of this study, it should initially be considered that this is a cross-sectional study, thus presenting the temporal limitations inherent to this type of design.In addition, when determining the factors associated with satisfaction with the use of dental services, it would also be necessary to consider characteristics of the service, such as the physical structure of the service facilities, access to services, among others.This paper, however, addressed individual factors, that is, characteristics of the individuals and not the services that were used.Thus, pseudo R-square (R 2 ), while modest (16.1%), can be considered relevant, since this investigation did not evaluate all the realms that may influence the outcome.In addition, only quantitative methods were used in this research, and a combination of quantitative and qualitative methods has been suggested as the best option to investigate satisfaction with the use of health services 46 .
On the other hand, the sampling plan, calibration of examiners, data collection registered in handheld computer and the conduction of the analyses considering the correction by the design effect were strategies that ensured data validity and reliability.

Conclusion
The present study showed a low prevalence of dissatisfaction with the use of dental services among adults.Smokers and older adults were less dissatisfied with the dental care received.On the other hand, there was greater dissatisfaction among the individuals who did not receive information on how to prevent oral problems, which negatively self-perceived their chewing and which self-perceived some discomfort in the mouth, head and neck.Services must access the perceptions and expectations expressed by users, going beyond the purely normative view of the need for treatment.In addition, they need to provide information in adequate quantity and quality to the users, fulfilling their role of improving people's access to health information, having knowledge as a background to achieve equity in oral health, increasing "health literacy" of the population and increasing user satisfaction.

Collaborations
LL Roberto carried out the analyses, data interpretation and writing of the paper.AMEBL Martins worked on the conception of the study and coordinated fieldwork and data processing.AMB Paula worked in critical review of the paper.EF Ferreira worked on the data interpretation and critical review of the paper.DS Haikal acted in the design of the study, contributed in the analysis, data interpretation and critical review of the paper.

Table 1 .
Descriptive and bivariate analysis of the adult population, by dependent variable and primary determinants of oral health and oral health behaviors and outcomes.Montes Claros (MG), 2009.(n = 830).

Table 1 .
continuation * Variation at n = 830.Due to loss of information.

Table 2 .
Multiple regression model of dissatisfaction with the use of dental services and determinants of oral health and oral health behaviors and outcomes.Montes Claros (MG), 2009.(n = 830)