Factors for the Use of Dental Services Associated with Untreated Caries in Children of a Cohort and the Care Offered in the Primary Health Care of Recife , Brazil

Objective: To analyze factors of use of dental services associated with untreated caries in children from a cohort in the Primary Health Care of Recife and the integral care offered to oral health of children in the areas studied. Material and Methods: 1st step: Analytic study with secondary data from the dental caries inquiry conducted with the cohort in Health District II and IV in 2010. 425 records of children aged 5-7 year were included. The Dependent variable was carious component of the dmft (d≥1) index and independent variables were sociodemographic and use of dental services. Data were analyzed using descriptive and inferential statistics. Variables associated with p≤0.20 outcomes were included in the logistic regression multivariate model with permanency criteria of p≤0.05. 2nd step: Descriptive study with 15 key individuals: (03) managers, (01) central level, (02) district level, (12) members of 2 health family teams from the selected districts. The validation of the data collection instrument was performed (semi structured questionnaire). Data were processed by a descriptive analysis in Excel spreadsheet. Results: The prevalence of untreated caries was high (58.1%). The factors associated with use of dental service was: age: 7 years old (OR: 1.62; p<0.001), toothache in the last 6 months (OR: 9.63; p<0.001), unsatisfactory care (OR: 3.57; p<0.014). There was valuation of epidemiological knowledge to plan integral and integrated oral health actions for children, but deficiencies in inter-professional actions were identified to perform these actions, which are recommended by family health teams. Conclusion: The factors associated with the outcome evidenced fragilities in the dental care directed to children and the structuring of the integral care to the oral health offered, although it indicates inter-professional action, it has not been effective to positively impact the d≥1 levels observed.


Introduction
Among dental health problems that affect children, dental caries is considered a matter of sanitary importance, both due to its high prevalence and due to its severity in populations of socioeconomic and nutritional disadvantages, in addition to being a cause of pain, discomfort and feeding difficulties, which compromise the quality of life and the biopsychosocial development of children [1].
Despite the overall improvement in the oral health situation, children with deciduous dentition have not been benefited by the decline of the disease in the same proportions as 12-year-old students in both developed and developing countries [2,3].In Brazil, based on data from the last two national oral health surveys conducted in 2003 and 2010, a situation of maintenance of unfavorable caries levels in the deciduous dentition is observed [4].
Despite the advances achieved with the inclusion of children under five years of age in the recent oral health policies [5,6], deficiencies in planning and fragility in the implementation of effective oral health care actions are identified, as well as difficulties of access to dental treatment [7].These issues have historical roots that are related to the exclusionary way in which the oral health care was conceived in the country, with prioritization of the offer of dental care to schoolchildren aged 6-12 years and to permanent dentition [5].In addition, professional resistances to the new organizational logic required by APS for integral care in oral health are discussed [5][6][7][8].
National studies have demonstrated problems related to the use of dental services among low-income children [7][8][9].In the context of expanding oral health actions in the public health network of Recife, a study identified 77.9% of children aged 5-12 years who had visited the dentist, a higher prevalence of use of public services [10].However, in the context of Basic Health Care (ABS) in Recife, there are still few analyses on aspects related to the oral health situation of children, use of dental services and organization of health care offered [11].To this end, in 2006, a cohort of children was constructed and a dental caries survey was carried out on 2,020 preschool children aged 18-36 months and five years, assisted by the ABS network of Recife, when high prevalence of dental caries and risk factors of social behavior were identified [12].In 2010, a new evaluation of these children examined in 2006 (18 and 36 months), now aged five to seven years, was performed, collecting the same data -in addition to new information related to the use of oral health servicesto perform the follow-up over time of this population (cohort) [13].
This study presents an analysis of factors of use of dental services associated with untreated caries of children of this cohort when they were evaluated in 2010 and on the structuring of integral oral health care offered in the study areas.

Material and Methods
It is a mixed observational study, consisting of two steps.The first analytical step of quantitative approach was performed with secondary data extracted from the second cross-sectional study of caries and associated factors, carried out in 2010 in the children's cohort of the ABS of Recife in Health Districts (DS) II and IV, when they aged 5-7 years [13].In the second step, a descriptive study on the structuring of the oral health care offered at the Family Health Units (USF) of the DS studied was conducted.
The City of Recife is considered totally urban, and has as characteristic the heterogeneity in the living conditions of its inhabitants.There are highly valued areas and others with urban infrastructure problems.Recife has not yet a public fluoridated water supply network and has irregular supply of treated water [14].In 2010, in relation to the public health network, Recife had 243 family health teams and care coverage around 50%.With regard to Dental Care, it had 123 oral health teams, distributed in 117 USF and the presence of 06 Centers for Dental Specialties (CEO) one per Sanitary District (DS).Despite this expansion, coverage remains low.In February 2014, the reported number of oral health teams was 144, corresponding to 32% oral health coverage [13].
The population of the first step of the study consisted of the total sample of children (n = 469) from the 2006 survey, when they aged 18-36 months, and were recruited to be reexamined in 2010 (5 to 7 years).References to sample size and sample selection from the 2006 caries survey (age: 18-36 months) can be analyzed in previous publications [13,14].The inclusion criterion considered the quantitative sample that had visited the dentist (n = 425) and whose data on the independent and dependent variables (component d of the dmft index) were stored and validated for analyses in the survey database of 2010.
Data for association analyses were obtained through structured interviews (questionnaires) and clinical exams for the measurement of the dmft index applied in the 2010 survey.Data were collected by 34 examiners and 33 auxiliaries belonging to the oral health teams of USF of DS II and IV.They were trained and calibrated for 18 hours.The codes and criteria adopted for caries examination followed international standards [15] and compliance with ethical requirements, at that time, of National Health Council Resolution 196/96.The calibration results were observed using the General Percentage of Concordance (PGC) and the Kappa statistical test.The general values obtained reached recommended reliability scores for epidemiological surveys of dental caries [15].
The dependent variable was the "d" component of the dmft index greater than or equal to one (d≥1).The independent variables used in the analysis (modeling) of the "d" component were extracted from the blocks of variables explored in the 2010 study.A new regrouping of variables related to the use of health services and on sociodemographic conditions of the sample was performed.
Data were descriptively analyzed through absolute and percentage frequencies and inferentially through Pearson's Chi-square test, in order to evaluate the significant association between occurrence of untreated caries and independent variables.To assess the strength of the association, the Odds-Ratio (OR) value was obtained.To determine the prevalence of untreated dental caries (d≥1), a multivariate logistic regression model was adjusted with variables that showed significant association of p ≤0.20.To obtain the final regression model, the backward stepwise selection method was used, based on the logarithm of the likelihood ratio, taking as a criterion of permanence of the variable in the model a 5% significance level (p≤0.05).The margin of error used in the decisions of statistical tests was 5% and confidence intervals were obtained with 95% reliability.
The software used to enter statistical data and for the elaboration of statistical calculations was SPSS version 21.

Second Step
The second step was reserved to descriptively describe, in the light of official recommendations [5,16], aspects of the structuring of integral child care focused on oral health implemented in the study areas.The study population consisted of 15 key actors involved in the management of the health policies of Recife and in the provision of integral health care to users enrolled in two USFs of Sanitary Districts II and IV, distributed as follows: (01) manager responsible for the Primary Heath Care policy (01) Sanitary District II manager, (01) Sanitary District IV manager and (02) family health teams, one from each participant Sanitary District, who were randomly selected.Six members of each health team (physician, nurse, nursing assistant, dentist, oral health auxiliary and community health agent) were included.The inclusion criterion of participants of the second step was to freely accept participating in the study as questionnaire respondent.
The data collection instrument used was a semi-structured interview form, with specific questions for managers (municipal and district) and health team members, which was validated by the instrument [17].Its elaboration was preceded by bibliographical review and to guarantee the quality control of data, three specialists were selected, with experience related to the theme of the study to evaluate the instrument.To this end, the interview form was evaluated to verify its content and clarity regarding the formulation of sentences.After external evaluations, corrections / adjustments were performed based on observations of each reviewer; the interviewer received theoretical training and tested the instrument on three professionals (a central level manager, a district level manager and a dentist) appointed by the Health Department of Recife.After these procedures, the final revision and necessary adjustments of the instrument were performed.

Results
The results of the 1 st step can be verified in Tables 1 and 2. Table 1 shows the absolute and percentage distributions and bivariate analyses performed for the set of independent variables studied in the total group and the occurrence of untreated caries (d≥1).The d≥1 prevalence was 58.1%.Regarding the block of sociodemographic variables, it is highlighted that the majority (56.9%) aged 7 years; more than half of the sample was female (51.3%) and 50.6% of mothers had 8 to 10 years of schooling; regarding occupation, the highest percentage corresponded to unoccupied mothers (46.6%) and non-specialized (39.8%) mothers; and (52.2%) of fathers had specialized / semispecialized occupations.For the fixed margin of error (5%), variables age and maternal schooling were those that showed significant association with d≥1.
Regarding the group of variables of the use of dental services block: in the total group, 33.2% had had toothache in the 6 months prior to the survey; the two highest percentages of the age groups of the 1 st visit to dentist were more than 1 to 3 years (45.9%) and 4 years or more (43.8%);for the majority (90.8%) of respondents, visit to dentist was in the public service; the majority (61.2%) evaluated the quality of care as good and only 32.9% of interviewees rated it as very good.
Except for the first visit to dentist, for the other variables, a significant association with d≥1 (p≤0.05) was found.Have had toothache in the 6 months prior to the research (p<0.001);(p=0.021) and the evaluation of the quality of care performed (p=0.002),showed strong and statistically significant associations with the outcome.It was observed that the percentage of children with pain episodes of dental origin was high (33.2%)and 87% had d≥1 and also a high prevalence of untreated caries (65.1%) was found among children aged 4 years or older in the 1 st visit to dentist.Similarly, district managers affirmed that the actions of integral health for children, with the inclusion of oral health, follow the current municipal planning, in addition to ensuring that both Sanitary Districts (II and IV) formulate the local oral health care plan, considering the epidemiological profile of each area.They also affirmed that, in addition to preventive and assistance actions for preschool children with dental needs, integrated actions in childcare are recommended.
Regarding indicators of monitoring oral health actions of children, only Sanitary District IV interviewee had this knowledge.These were strong limitations to both Sanitary Districts: lack of basic material for service and inadequate infrastructure.
The characterization of the performance of health team members focused on the integral care of children's oral health is presented in Table 3.In both Sanitary Districts, 4 oral health team member and 8 members of two family health teams were interviewed.The majority considered that the actions performed are in accordance with official recommendations.Regarding Q2 (priority collective actions for the control and prevention of dental caries), among the list of actions, the one that received the highest consideration was the systematized implementation of topical application of fluoride in community social facilities.Among sociodemographic indicators that may interfere with the prevalence of caries (Q3), all respondents (12) considered maternal schooling and feeding as the most important factors.Regarding Q4 (integral actions in children health offered by the USF with the incorporation of oral health), the results observed in increasing order regarding the number of responses were: promotional and preventive actions in childcare and home visit; home and pediatric curative actions, followed by PSE; and adult literacy.On the combined oral health actions carried out by health teams for the most vulnerable social groups, the health team of Sanitary District II responded not to plan specific actions; but in Sanitary District IV, the health team prioritizes and performs actions for these groups (Q5).
Table 3. Characterization on aspects of integral oral health care for children performed by the health teams of basic health units studied in Sanitary Districts II and IV.
Health Team SD II Health Team SD IV Total Group Questions DH Team Other Members DH Team

Discussion
The prevalence of untreated caries in the sample analyzed and users of oral health services offered by the Family Health Units of Recife, in the studied DS, was 58.1%, and seven-year-old children, who were the majority, were 1.56 more likely of having caries than younger children.In 2006, when the sample aged 18-36 months, this prevalence was lower (28%).When comparing this result with those obtained in the 2010 national survey [18], it was found that it was greater than the prevalence of d≥1 recorded for Brazil (48.2%) and for Recife (48.55%).In addition to a study carried out with data from the 2010 national survey, opposite relationship between d≥1 levels and contextual and individual factors in Brazilian children aged 5 years was identified [4].
In this study, low maternal schooling, considered a proxi variable of unfavorable living conditions, was a factor associated with caries in bivariate analysis, which corroborates other previous studies [3,9,19].Among the variables related to the use of dental services, only the of the 1 st visit to dentist was associated with d≥1 in this bivariate analysis, although most of the children obtained this access in the Family Health Units of Sanitary Districts II and IV, which was verified by means of the analysis of data related to the 1 st visit to dentist.Other similar studies also did not find difficulties in accessing public dental services by children with deciduous dentition for caries treatment [7][8][9], which demonstrates ABS efforts in different localities of the country to increase children's access to oral health services [5].
Multivariate analysis included one sociodemographic (age) variable and two variables related to the use of oral health services (tooth pain and quality of care received) in its final model, which were the factors associated with untreated dental caries of the sample studied.
Being older was an associated factor for the outcome in the health districts where children live, as demonstrated in other similar studies [2,3,12,20].Corroborating other studies [21][22][23], having had toothache in the last six months was a determining factor for d ≥1 among children investigated, compared to those who did not have this discomfort.These results may be indicative of the influence of unfavorable social and environmental factors in areas where the study population lives and whose accumulation of sociobiological risks for the occurrence and progression of caries may be continuously increasing with age [20][21][22][23][24].In addition to revealing difficulties related to oral health surveillance actions planned for the study areas aimed at preventing and controlling early caries and monitoring groups at greater risk of having pain episodes of dental origin that require priority access to clinical treatment [25], as this condition has a negative impact on the quality of life of children and their families [21].
At the same time, these findings suggest that there are problems of structuring integral child care and timely access to dental care offered in the study areas.Other previous studies have observed ABS efforts that favor the effective use of dental services by children from families covered by family health strategies [7,8,26].However, a study carried out in Paraná found that the prevalence of use of dental services by preschoolers was not influenced by the presence of dentists in the neighborhood of children's homes [24].Another cultural aspect related to the devaluation of the deciduous dentition by parents / guardians of children must be considered in these results.This aspect needs to be addressed with health education actions for families in the study areas, as widely recommended [6,12,16].A study carried out in the backlands of Paraíba reported that most parents of preschool children believe that their children do not need dental care, although they stated that there were no difficulties in accessing public dental services for their children [7].
In addition, the influence of the dental care quality, performed at USF, also presented associations with the outcome.Considering the dental care provided as regular to poor was associated to higher probability for the child to present d≥1 in relation to other users, whose parents considered the care as very good or good.This may be evidencing the difficulties of oral health teams in the study areas in carrying out recommended approaches to children [5], with humanization and adherence.Another previous study of qualitative approach identified humanization in care and professional competence as factors that influence the decision of mothers to take their children under five years of age to dental treatment in public ABS services [27].Another research showed dissatisfaction of mothers with the care received by their child by ABS health professionals, a fact that negatively affects the formation of link and attention to the child [28].In this result, aspects related to investments aimed at improving working conditions and permanent education of professionals in family health teams should also be considered, factors that influence the quality of care provided [29].
The descriptive analyses carried out in the 2 nd step of the study aimed at characterizing the structuring of the integral oral health care for children in the researched areas and compare them with results obtained in the first step of the study.It should be emphasized that the National Policy of Basic Health Care recommends that the work process of oral health teams should also seek the exchange of knowledge among other health team members, aiming at the collective and integrated construction of health interventions [30,31].
It was observed that the study subjects had knowledge about issues related to the planning and execution of actions aimed at integral oral health care for children.As an example, interviewed managers reported the use of epidemiological data to subsidize the elaboration of municipal health plans and that the formulation of integral health actions -promotional, preventive and care -focused on the child life cycle of study areas, occurs in a decentralized manner.Concordances were found in studies that point out the importance of professionals in health surveillance actions, since this area is a structural component in the organization and management of SUS practices.The efforts of the oral health sub-sector to implement oral health surveillance, strategically integrated with the National Health Surveillance System are highlighted [32,33].
In addition, this finding may be related to the professional qualification observed by the majority of professionals interviewed, since a high percentage had specialization in public health and other related areas.This is in line with governmental initiatives that have been implemented since 2001 to qualify professionals inserted in different strategic points of the SUS network, especially within the scope of the Family Health Strategy [29,34].
Although there has been an expansion of the oral health network in the country with financial support to municipalities for the implementation of oral health teams within the scope of the ABS of the country [5], structural problems and budget limits related to the network of health services of Recife were pointed out by three managers, and this issue was repeatedly addressed in different studies as limiting factors for the good performance of professionals' actions and the fulfillment of goals established for oral health [5,6,8,27,30].Thus, the high prevalence of d≥1 found may be revealing structure problems in the dental care provided to children, which guarantees clinical actions of treatment and control of the disease in the study areas.
The analysis of results in Table 3, in a general way, demonstrated a search for the integration of oral health actions for children in the direction of the interdisciplinary team work, as supported in nationally recommended guidelines [5,8,16].However, although professionals are oriented to perform an approach based on the common and modifiable risk factors for most childhood illnesses and diseases [12,35], only half of the "other members" considered implementing a healthy food guide in schools / day care centers and permanent education processes of teams on the oral health promotion aimed at children, as priority measures for the control, prevention and timely treatment of caries.A similar study identified the lack of structuring of public health services, with a multidisciplinary approach for children, as a challenge for the control of childhood caries [35].
The analysis of answers about the interviewees' knowledge on the relationship between social indicators and outcomes in child caries (Q3) suggested that subjects did not fully value this relationship, as has already been demonstrated in several studies [3,4,14].This finding reveals the need for greater investment in the qualification of health teams, since it is essential that professionals working in ABS intervene on the multiple social dimensions of different health-disease processes [34].
The last questions formulated were aimed at knowing the stage of incorporation of child oral health measures in the daily life process of health teams interviewed.A trend towards integrated action of professionals was identified (Q4), but it was not unanimous, despite the recommendations of the Ministry of Health towards the operationalization of integrated actions planned according to the life cycle and condition [5,6,8,16].Among the eight other interviewed members, only three of them considered the insertion of oral health actions in the PSE, even though these actions were included in the activities recommended for PSE [36].
In addition, low team interaction to plan combined oral health activities for vulnerable groups was observed (Q5).Concordances were obtained with other studies that discuss the challenge faced by Basic Health Care, from the perspective of the Family Health Strategy work to establish interdisciplinary work as a condition to promote integrality in health care and, consequently, to achieve satisfactory health outcomes [37][38][39].A study carried out with family health teams in the city of Goiânia observed that professionals were not aware of how to perform health care beyond biomedical aspects.In addition, difficulties of interaction between dentists and the other members of health teams were observed due to the tendency of this professional to isolate himself in the office [38].However, in southern Brazil, within the scope of the Family Health Strategy, another study verified a perception of teamwork associated with multiprofessionality, with the exchange of knowledge among members aiming at the implementation of health actions [39].
Regarding the study limitations, methodological care was taken aiming at ensuring the validity of data.In the first step, with regard to the selection bias, methodological care was taken to ensure the internal validity of data, since the sample size recruited from the initial 2006 study was reevaluated in 2010 [13].One should consider the limitation of the multivariate analysis method to be able to understand the complexity of the process studied, since all explanatory variables were attributed to the level of the individual.It was also necessary to pay attention to the generalizations of results.The study population is specific and comes from poor urban areas of Recife covered by APS policies, and the comparison with similar populations is pertinent.In the second step, the steps followed for the construction of the data collection instrument is considered as a strong point, so that the results in fact express the opinion of participants, minimizing information bias.
Finally, in areas with precarious socioeconomic conditions covered by the Recife Basic Health Care strategies, different conditions were identified that contributed to the presence of untreated caries in children belonging to the sanitary district investigated.It is hoped that the results obtained in this study contributed to qualify the planning, organization and execution of health practices directed to the child life cycle, consistent with the principles established for Basic Health Care.For an intentional change in the oral health situation presented by children included in the study, broad approaches are indicated on the social determinants of health, with social and public health policies that impact on the quality of life of these children.In addition to greater investments aimed at the qualification of integral health actions and training and permanent education policies focused on the strengthening of interdisciplinary work, in the perspective that synergistically, these actions reduce the high prevalence of observed d≥1 and improve satisfaction with the dental services offered to the population covered by APS strategies.

Conclusion
The factors of use of services associated with the study outcome showed weaknesses in the timely dental care offered to the children investigated with dental caries treatment needs.In addition, the structuring of integral actions in oral health for children living in the study areas, although presenting indications of performance in a multiprofessional team, is not being effective to positively impact the d≥1 levels observed.
Data collection was performed by a student of the UFPE Dentistry course, under the supervision of the researcher at the interviewees' workplace, after signing the Informed Consent Form, allowing their participation, according to requirements of the Resolution 466/12 of the National Health Council on research involving human beings.Based on the interviews, information on the sample characterization was obtained, which, after the descriptive analysis of data, was organized into two blocks to present the results: Block I -professional qualification and position and Block II -knowledge and actions performed for the integral oral health care for children.For the analysis of data collected in interviews, frequency distribution was produced.The statistical software used was Microsoft Excel 12.0, version 2007.Ethical Aspects The research project was approved by the Research Ethics Committee of the Health Sciences Center of UFPE (Protocol No. 942.271).

*
Dental Health Team: Dentist and Oral Health Assistant; **Other Members: Physician, Nurse, Nursing Technician and Health Agent; SD= Sanitary District.

Table 1 . Analysis of the association between prevalence of untreated dental caries and the sociodemographic factors and use of oral health services. Occurrence of Caries (c≥1)
*Significant association at 5.0% level; Pearson's Chi-square test.

Table 2
The evaluation of the model adjustment showed that there was agreement of 69.9% between observed and predicted values, with sensitivity of 75.7% and specificity of 61.8%; (p<0.001) and good adjustment of data to the model, according to the Hosmer & Lemeshow test (p=0.903).
shows the results of the multivariate logistic regression model for d≥1 in relation to the variables studied.Sociodemographic variables and the use of oral health services were included in the model, which presented p≤0.20 (Table1) and variables that had p≤0.05 were maintained in the final model.The variables included in the model were: age, sex, maternal schooling, mother's occupation, toothache in the 6 months prior to survey, age of the child in the 1 st visit to dentist, place of the 1 st visit and quality of care.In the final model, the following variables were maintained: age group (OR: 1.62; 95% CI: 1.04-2.52);toothache in the 6 months prior to the study (OR: 9.63; 95% CI: 5.46-16.99)and poor quality of care provided (OR: 3.57; 95% CI: 1.30-9.81).This meant that the probability (OR) of a child in the sample to have d≥1 is higher if he/she is 7 years old than those who were 5 or 6 years old; if he/she had toothache in the 6 months prior to survey; and whether parents rated care as regular / poor / very poor or good compared to those rated as very good.

Table 2 . Final model of multivariate analysis.
results of the second descriptive step refer to the 15 subjects interviewed.Regarding the sample characterization (Block I) related to professional training, it was demonstrated that 10 professionals had higher education degree and the others had high school degree.The results of Block II, on the knowledge of managers about planning, programs and actions aimed at the integral oral health care for children implemented in the ABS of Recife and respective sanitary districts, are described below: the central level manager demonstrated knowledge on issues of planning and implementation of oral health actions, stating that specific actions for oral health were implemented in the last two municipal plans; and that these are formulated based on epidemiological analyses from national statistics, data from the Primary Care Information System and other references.This actor listed different oral health actions of children that are developed in Sanitary Districts II and IV, such as: child care in CEOs, integral care offered in primary care units (care, promotional and preventive), oral health actions in the School Health Program (PSE); in addition, it demonstrates knowledge about the indicators that monitor the oral health actions of children (brushing index, 1 st visit to dentist in the pediatric age range, oral health activities in primary health care groups and PSE and the tooth extraction index).As limits, this manager reported aspects related to insufficient federal resources; inadequate physical structure, absence of state counterpart in the cost of oral health Good 1.97 (1.30 -2.99) 1.81 (1.14 -2.88) 0.012* Regular/ Poor / Very Poor 2.97 (1.17 -7.55) 3.57 (1.30 -9.81) 0.014* *Significant at 5.0%; selection process with inclusion if p ≤ 0.20 and maintenance if p ≤ 0.05.The actions, old / inadequate physical structure in some units and legal process of bidding for materials related to their quality.