Use of the International Classifi Cation of Diseases in the Analysis of Dental Absenteeism

OBJECTIVE: To compare the use of disease and injury classifi cation codes in workplace absences requests due to dental causes. METHODS: The study analyzed 240 requests in a federal public agency between January 2008 and December 2009. The use of the International Classifi cation of Diseases-10th Revision (ICD-10) was compared to the Application of the International Classifi cation of Diseases to Dentistry and Stomatology (ICD-DA). The degree of specifi city was determined for the codifi cations on workplace justifi cations, as well as for codifi cations assigned by offi cial dental experts in indirect inspections and expert examinations. RESULTS: Of the total number of dental certifi cates, 22.9% did not present the ICD, 7.1% used ICD-9, 3.3% used ICD-DA and 66.7% used ICD-10. The majority of codifi cations were concordant (55.1%), and greater specifi city was found in codifi cations assigned after evaluation by offi cial dental experts. CONCLUSIONS: The results indicate the need to improve use of ICD-10 among dentists and offi cial dental experts .For analysis of work absenteeism, it is suggested the use of ICD-DA and the International Classifi cation of Functioning, Disability and Health, which provide relevant data for monitoring absenteeism due to dental reasons.


INTRODUCTION
Programs addressing the oral health of workers should be stimulated and developed based on epidemiologic knowledge, in order to promote disease reduction and improve quality of life among workers.
The value of epidemiologic information is refl ected in the quality of the information system, which are important tools for planning and decision-making. 8formation is central to the health care process.A complete registry is essential for adequate follow-up of workers.
Morbidity data are fundamental for epidemiologic studies and for management, planning, organization and evaluation of health services. 10,13,14According to the World Health Organization (WHO), 14 morbidity data should be interpreted with an understanding of data quality and diagnostic reliability.
The method of information collection, processing, evaluation and use should follow effective technical criteria, through methods that study the worker as a whole and their relationship with the environment 16 and that include the complexity and richness of health phenomenon. 1 WHO classifi cations are tools that help occupational health professionals in the clinical-epidemiologic approach to individual health in relation to occupation. 16The concept of a family of health classifi cations was advanced for more comprehensive analysis of the health-disease process and inclusion of data beyond the diagnosis. 9The International Classifi cation of Diseases -10th Edition (ICD-10), the International Classifi cation of Diseases to Dentisty and Stomatology (ICD-DA) and the International Classification of Functioning, Disability and Health (ICF) belong to this group of classifi cation.
Knowledge on the applicability and purpose of WHO classifi cations is indispensable for strengthening and structuring an occupational health information system.Inclusion of morbidity data on workplace absences is essential in this process.
The sick leave benefi t is provided in Law No. 8,112, which discusses the legal administration of public service employees in the federal government, autonomous federal agencies and public federal foundations.a Sick leave is allowed upon offi cial expert examination.Sick leave with lack of expert examination is conditioned on presentation of a medical or dental justifi cation that should contain the ICD code or diagnosis.
Indirect inspection involves evaluation of sick leave by offi cial experts based exclusively on the justifi cation and other complementary review.The review committee consists of at least three experts.
The veracity of dental justifi cations should be respected, unless there is a divergence in opinion of the offi cial dental expert in the institution that performs offi cial reviews and evaluates occupational incapacity.b Recent regulations that delimit the action of dentists in offi cial expert examinations and that standardize dental examinations reinforce the importance of dental professionals in evaluation of occupational incapacity and the strengthening of relevant and consistent morbidity a Brazil.Law No. 8,112, from 11 December 1990.Describes the legal situation of public civil servants of the country, autonomous federal agencies and public federal foundations.Diario Ofi cial Uniao.12  data.This contributes to the construction of a health information system that supports dental health actions and policies for workers.b,c,d Nonetheless, increased knowledge of WHO classifi cation tools is necessary to improve information from the evaluation of occupational absenteeism.
The objective of the present study was to compare the use of disease and injury classifi cations in requests for occupational sick leave due to dental reasons.

METHODS
The study was performed in a federal public service in São Paulo state, Southeastern Brazil, from January 2008 to December 2009.There were 3,518 employees active in December 2009: 1,753 in the capital and 1,765 in 37 other municipalities.
The 240 requests for workplace absence due to dental reasons were analyzed (105 in 2008 and 135 in 2009).
Justifi cations were classifi ed into four groups: without ICD (no code, with a description of diagnosis or procedure performed); with ICD-9; with ICD-DA; with ICD-10.
The following situations were identified through comparative analysis of codifi cations on justifi cations and codifi cations by offi cial experts: concordant (codifi cation without change or new inclusions); discordant (other codifi cation assigned by offi cial experts, with changes in the three-character categories, four-character subcategories, ICD chapter or inclusion of fourth digit in the three-character category); additional code (no changes and additions with one or more codifi cations).
The ICD chapters are subdivided into categorical groupings of three characters, which can be divided in up to ten subcategories by use of a fourth character. 14r certifi cates with ICD-DA, classifi cation of the fourth character was analyzed, and for certifi cates with ICD-9, equivalence with ICD-10 was determined.

RESULTS
The 240 requests for occupational sick leave for dental cause totaled 482 days of absenteeism.The types of evaluation undertaken are described in Table 1, which shows that the majority of sick leave requests were evaluated by indirect inspection (70.0%).
Of the justifi cations, 22.9% were not codifi ed according to the ICD; 7.1% were codifi ed by ICD-9 and 3.3% coded according to the ICD-DA.Classification according to ICD-10 was present on 66.7% of justifi cations.
In 55.1% of cases there was agreement between the classifi cations on dental justifi cations and the classifications made by experts.The assignment of different codes was more frequent in evaluations by offi cial dental experts, and the additional code was defi ned by dental committee (Table 2).
The additions to codes involved the inclusion of code Z54.0 -convalescence following surgery.
Of classifi cations with three-character categories, 69% were assigned a four-character classifi cation by offi cial dental experts (Table 3).The combination of classifi cations refers to the use of more than one ICD code on the same justifi cation.
Table 5 shows the relationship between classifi cations with a low degree of specifi city and type of evaluation performed.The residual category.9(without other specifi cation) was most frequent in all types of evaluation performed.
Considering total number of employees (168), the majority (76.2%) presented only one request for occupational sick leave in two years.Employees that were absent two or more times (23.8%)accounted for 46.7% of requests.
The distribution among employees with more than one request was as follows: 26 with two requests (65%), eight with three requests (20%), four with four requests (10%), one with six requests (2.5%) and one with 14 requests (2.5%).
Of the 112 cases of repeat requests, 15 were extensions immediately following expiration of a fi rst request and 26 presented another request for sick leave within 60 days from the end of the fi rst request.
Classifi cations were distributed in fi ve chapters of ICD-10: Chapter XI -Diseases of the digestive system (n = 209); Chapter XXI -Factors infl uencing health status and contact with health services (n = 27); Chapter XIX -Injury, poisoning and certain other consequences of external causes (n = 2); Chapter XIII -Diseases of the musculoskeletal system and connective tissue (n = 1); and Chapter XX -External causes of morbidity and mortality (n = 1).
The main reasons for absenteeism, by order of prevalence, were: exodontal; surgery to insert implants; gingivitis and peridontal disease; extraction of impacted teeth; disorders of gingiva and edentulous alveolar ridge; pulpitis; periapical abscess without sinus.
The average duration of sick leave was two days (standard deviation -SD: 2.6; mode: 1 day; median: 1 day).

DISCUSSION
Of all the justifi cations evaluated, 22.9% did not present an ICD.Although ICD-10 has been in effect since 1993, Dentistry professionals should consider the possibility of more than one classifi cation in order to clarify the health situation analyzed. 12The possibility of a combination of classifi cations avoids loss of useful information for expert decision.In cases of classifi cation for multiple diagnoses, the hierarchy of the information should be considered 13 and selection rules for morbidity classifi cation should be respected. 14In this study, following expert evaluation, a combination of classifi cations was used in over one quarter of sick leave requests (26.3%); although, the second classifi cation was always Z54.0 (convalescence following surgery), meaning a new diagnosis was not included.
The use of four-character subcategories from ICD-10 provides greater detail in the description of the health condition observed, which allows for detailed analysis of diagnoses established and, therefore, more effective decisions and action.Health professionals that engage in classifi cation should use the most specifi city possible when classifying the diagnosis in one of the ICD categories. 2,6,7,14The ICD-DA provided the highest degree of specifi city through a more inclusive and consistent classifi cation of oral diseases and oral manifestations of other diseases. 3,18ICD-DA was little utilized in the justifi cations reviewed, which suggests a need to promote the classifi cation among dental surgeons.
The morbidity data with classifi cations at the fourcharacter level spanned 34 subcategories of the ICD in Diseases of Oral Cavity, Salivary Glands, and Jaws (K00-K14) in Chapter XI Diseases of the Digestive System.For 19 subcategories (55.9%), the ICD-DA provides greater details in diagnosis and could be used for most conditions coded according to ICD-10, supporting decision-making through indirect inspections.
The changes and additional codes that occurred following clinical evaluation suggest that more adequate and complete information could be obtained to describe the worker's health condition.
The main reasons that indirect inspections could include ICD-10 codes, changes and additions were: presence of the diagnostic basis or description of procedures performed in dental justifi cations and the analysis of clinic records of professionals credentialed by the dental assistance program provided to the federal public service.The description of procedures on justifi cations was most useful.
For the most specifi c information possible, the.8 and.9 residual categories should be avoided for classifi cations. 6,7The fourth character.8 is used for other conditions that belong to the three character category, and.9 has the same meaning as the title of the three character category, without adding any information, or it is like adding "unspecifi ed" to the category title. 14aluation by expert dentists allowed for codifi cation with ICD-10 of justifi cations without an ICD or with ICD-9 codes and the inclusion of the fourth character in three-character codes.There were codifi cations of low detail categories in 25.4% of the cases, mostly due to indirect inspections, meaning improvement of information through review of justifi cations was impossible.
The isolated use of code Z54.0 (convalescence following surgery) on three justifi cations evaluated by indirect inspection does not provide information on the health condition that required surgery, 12 since it does not describe any condition of morbidity.For these cases, expert examination is recommended.If the information on morbidity is insuffi cient or deficient on the justifi cations presented by public service employees, there is an investigative responsibility to improve information quality.
To improve information quality, with assignment of a code and a specifi c diagnosis, it is important to understand the selection rules of a primary condition for tabulation of morbidity records. 14When possible, the record should separately describe other conditions or problems that were treated during the visit.It is recommended to perform codifi cations and analyses of multiple conditions to improve routine data. 5,11,14alth conditions codifi ed with residual categories (.9) had greater variance in the average duration of leave, suggesting that more specifi c codes allow for a more precise estimate of the number of days necessary for occupational absences.
Each individual experiences health conditions differently, even if they have the same pathology. 15Therefore, the same health condition can require different periods of sick leave.The isolated use of ICD-10 does not permit greater detail in the analysis of these differences, which could be recorded and investigated with the complementary use of ICD-DA and ICF.
Determination of the adequate period of sick leave for patient recuperation is fundamental to guarantee a return to work at the ideal time, without compromising the capacity to work and without promoting premature return to work in detriment to health and employee well-being. 12The degree of return to work can be an indicator of quality in expert investigations.
Adequate records of health conditions related to extended sick leave contribute to the study of occupational incapacity.For sick leave extensions, the adequate use of ICD-10 can provide additional information, by providing a record of complications from surgical procedures, for example.
A predictor for absence from work is previous work absenteeism. 17Of employees, 23.8% requested more than one leave from work due to dental causes.In these cases, the codifi cation for each request should be analyzed for more detailed analysis of causes of absenteeism.
The data generated by expert evaluations allows for the organization of a database to better understand the situation 4 and to construct relevant indicators for monitoring absenteeism.
Understanding the logic in selecting a code assignment for a specifi c diagnosis is fundamental to generate relevant information.It is important to adopt protocols and to establish standards and uniform selection criteria for codes.Therefore, the construction of expert technical protocols and capacity building are fundamental to improve health information.
Since this study was descriptive and retrospective, using secondary data, it was not possible to analyze concordance between and within examiners, which is a study limitation.
The study results show the need to improve use of ICD-10 among dental professionals.It also supports the ICF use in occupational oral health studies and the need to increase dissemination of WHO classifi cation tools among dentists.Further studies with these tools are necessary for comparison of diverse experiences.
Understanding all aspects of the human beings, including expert behavior, is important for defi nition of prevention strategies and to implement necessary health actions.
Currently, with the implementation of new legislation to standardize expert evaluations and delimit the functions of dentists it is critical to increase detailed analysis of aspects related to sick leave for dental reasons through the adequate use of classifi cation systems.
The research project was approved by the Ethics Research Committee of the Dental School of São Paulo University (protocol number 111/2009, on 10 August 2009).

Table 1 .
Requests for occupational sick leave for dental reasons, by type of evaluation.São Paulo, Southeastern Brazil, 2008-2009.

Table 2 .
International Classifi cation of Diseases codes after comparative analysis of codes on justifi cations and codes assigned by expert evaluation.São Paulo, Southeastern Brazil, 2008-2009.

Table 5 .
Relationship between less specifi c codes and type of evaluation.São Paulo, Southeastern Brazil, 2008-2009.