RESEARCH ARTICLES
Completeness and consistency of data on notified HIV-positive pregnant women
Diego Jorge Maia LimaI; Ana Carolina Maria Araújo ChagasII; Igor Cordeiro MendesIII; Mônica Oliveira Batista OriáIV; Priscila de Souza AquinoV; Ana Karina Bezerra PinheiroVI
INurse. PhD in the Postgraduate Program in Nursing of the Nursing Department of the Federal University of Ceará. Fortaleza, Ceará, Brazil. E-mail: diegojorge19@hotmail.comABSTRACT: This descriptive epidemiological study examined the completeness of information in records of HIV-positive pregnant women notified in Ceará. Data on pregnant women who tested positive for Human Immunodeficiency Virus (HIV) between 2000 and 2009 were collected in January 2011 at the Ceará Department of Health. The main variables bearing on the dynamics of preventing vertical HIV transmission were examined for completeness. The scale used ratings ranging from ‘excellent’ to ‘very bad’. Flaws were found in completion of fields on the notification forms, mainly in socio-demographic items. Levels of completeness of the variables ‘prenatal care’ and ‘delivery type’ was considered good. There is a need for discussion on the theme of information system databases during health professionals’ training process, so as to contribute to reducing incomplete and inconsistent data.
Keywords: health information systems; database; vertical transmission of infectious disease; HIV.
INTRODUCTION
Aids is considered one of the main problems of public health of the world history, a pandemic of difficult control and high morbidity and mortality. It is also multiple epidemic suffering through time, significant transformations in it epidemiologic profile, currently being marked by the heterosexuality, internalization, impoverishment and feminization1.
The feminization of the Human Immunodeficiency Virus (HIV) brings among its problems, the high number of women in reproductive age infected by the virus. In Brazil, from 2000 to 2009, the percentage of women between 20 and 39 years old infected with HIV was about 66% of cases, being a risk increased of vertical transmission (VT) of the virus2.
In the State of Ceará, the epidemic of aids also follows the national tendency of feminization. Since the year of the first case in 1983 until April 2010, 9.249 cases of aids were registered in Ceará, being 30% of them in women. There was a reduction of the male/female ratio in 1986 was 11 men for each woman and from 2005 was 1.9 men for each woman3.
There is a prevalence of 0.41% of HIV infection in pregnant women in Brazil, estimating that 12.456 newborns were exposed to the virus for one year4. The VT is the main way of HIV infection in child population, being responsible for 90% of cases in minors of 13 years in the country5.
It is very important, then, the evaluation of quality of information given by the information systems, an evaluation of this question, which can be measured indirectly by analyzing the validity of the information being displayed, as well as for assessing the completeness of filling the field notification.
The term completeness is the level of fulfillment of the analyzed field, measured by the proportion of field notifications with different category fron the indicators of data absence6. In this way, the health professionals should strengthen the information systems through works searching a qualified data feeding, through complete and reliable information, reducing the ignored fields.
From all the above, the objective of this study is to analyze the completeness of the information of the HIV pregnant record notified in the Information System of Notifiable Disease (ISND) in the State of Ceará, between 2000 and 2009.
LITERATURE REVIEW
The ISND was created by the Brazilian government as a strategy to make feasible the epidemiologic surveillance, allowing the follow up of children with compulsory notifications and other diseases of national interest. Thus, it allows the health professionals having information access available for the community. Therefore, it becomes a relevant instrument to help the health planning, defining intervention priorities besides, assess its impact7.
Soon, the notifications and its incorporation to the system allow monitoring epidemic spatio-temporal of aids in the country, as well as the follow up of the disease dissemination by exhibition category, subsidizing the prevention and control actions. Thus, it is important the epidemiologic surveillance as ones of the control ways of vertical transmission of HIV. However, for an efficiency surveillance, it is necessary that information be of good quality8.
Currently, the quality of all the available data are increasing coming from different information sources. However, the developed studies with the purpose of evaluating the data quality are poor9. Therefore, it is relevant the development of studies aiming to verify the consistency and completeness of this compulsory notification data.
METHODOLOGY
It is an epidemiological study, descriptive and retrospective, performed in the database of the ISND, with data of the positive HIV pregnant women from the State of Ceará, characterizing side data. For this study, It was considered the period between 2000 and 2009. This decade was chosen to be studied because of the cases surveillance of infected pregnant women by HIV and exposed children become compulsory in Brazil from 2000 and for being a recent period, enabling to evaluate the quality of information of the variables of this period.
The data collection was in January 2011, from the state database available in the Information Core and Health Analysis (ICHA), of the Health Secretary of Ceará (SESA-CE) having compiled information of the ISND records.
In 2006, there was a change in the ISND records about the seropositive HIV pregnant women, so the notification/investigation record of the positive HIV pregnant woman and exposed children, in use until 2006, was divided into two new records. One of notification/investigation of HIV infected pregnant women and other one of the HIV exposed child, using until today. However, the HIV exposed child is used only for notification, not being launched in the ISND, not having information data collection.
In this way, the present variables were selected in notification/investigation record of positive HIV pregnant woman and exposed child, as well as the notification/investigation record of the HIV infected pregnant woman.
The Ministry of Health preconized a set of measures to be done in prenatal, delivery and puerperium of seropositive HIV pregnant women reducing the vertical transmission rate of HIV from 25% to 1/2%. Some factors help to the VT reduction of HIV. Among them, there are viral, maternal, behavioral, obstetrics factors inherent to the newborn (NB) and related to brestfeeding7.
Thus, the completeness of essential variable records for the dynamic comprehension of the TV and HIV prevention were analyzed, whatever are the socio-demographic profiles of the seropositive pregnant women (age, education, race, color), prenatal performance, delivery type, as well as the use of antiretroviral by the pregnant woman and the exposed child.
The data collection were organized in tables, expressed in absolute and relative frequencies. The relative frequency was calculated from the total of the notification cases of each one.
For evaluation of data consistency, the analysis tool use was the field completeness. In this way, the field in the database filled with the category ignored or number zero was used as analysis tool, besides ignored data or terms indicating data absense9. The score used was adapted from previous studies, having the following evaluation levels: excellent, when the variable is less than 5% of incomplete fulfillment; good (5% to 10%); regular (10% to 20%); bad (20% to less than 50%); terrible (50% or more)10. There were analysis of the absolute number and the percentage of the not information completeness of the records of HIV pregnant woman, being the base of ISND information. After that, the results were discussed according to the relevant literature.
The ethic and legal precepts involving research with human beings were protected, following the Resolution number 196/96 of the National Health Council. The study was approved by the Ethic Committee in Research of the Federal University of Ceará, under protocol 302/10.
RESULTS AND DISCUSSION
In the period of 2000 and 2009, 1364 cases of seropositive pregnant women were notified in Ceará State. They showed a big increasing during the study period. Verifying separately the years of 2000 and 2009, it was observed a percentage increase of 198.5% in notification numbers, not necessarily representing the increase of cases in HIV infected pregnant women, but it was an improvement in the notifications.
For the age variable, in the notification record there are not ignored option, however, there was inconsistency in analysis of the presented information, making the data not viable. Therefore, the variability of these data was between 2.2% and 27.2%, showing a quality pattern varying from bad to excellent.
The education variable presented parameters between 17.7% and 72.3% of not completeness, being in the information quality between regular and terrible. The race/color variable showed a variability of 6.2% to 71% of not completeness with a score varying from terrible to good, as shown in Table 1.
It is noteworthy to show a significant improvement of completeness of all the variables from 2007, period occurring the change of the notification records, implementing the separate record for the HIV pregnant women. The completeness of race/color field, for example, having a terrible level of completeness, started showing a good completeness level after the record changing.
The pregnant age variable, being an obligatory fulfillment field, had the quality from bad to excellent during the years, different from other study performed in Espíritu Santa State – Brazil, finding in the same type of investigation an excellent level7. This fact was only in the study of 2008, when using the new notification record. This finding allows for questioning the quality of the notification records fulfillment, especially those items considered obligatory.
Similarly, researchers found total completeness of data for age variable in the period of 1998 to 2007 in evaluation of epidemiological variables of the Mortality Information System (MIS) in breast cancer of dead women in the Southeast region11.
The completeness of education and race/color variables of Ceará vary from terrible to good, different from findings in other references. Masters evaluating these similar relative variables to the Control Program of Tuberculosis in the state of Espíritu Santo – Brazil, found regular completeness12. Whereas, race/color and education fields in analysis of HIV pregnant of Espíritu Santo showed excellent completeness8.
In a study evaluating the MIS completeness by suicide, in the Brazilian Southeast, is shown that education, race/color variables, Espíritu Santo presented the worst fulfillment indexes13. So, it is observed a different in the findings of completeness about the study development place and the evaluated database.
Study about the MIS completeness of women with breast cancer indicated that the education was 42% to 79% of not completeness, being terrible in information quality. In the race/color variable, the variability of ignored data was 18% to 35%, showing a bad quality.
The education completeness was also a variable from a study performed in Vitória-Espírito Santo (Brazil) with data coming from the Information System of Cervix Cancer (ISCC), in the period of January to December 2007, revealing 99.8% of reports did not have information, characterizing it as bad fulfillment14.
It is important to show the discrepancy of the completeness found between socio-economic data, collected during anamnesis and the epidemiological data, revealing the importance given to epidemiology in the socio-demographic context14.
In similar analysis of this study, the information quality of cancer hospital records, in the whole country was observed that the age variable presented excellent level of completeness, while race/color presented regular level and education presented a bad level15.
The understanding of these data is very important, because the combination of race/color and education variables can show a social inequality. This information is widely used in epidemiological studies, guiding effective measures for VT prevention15.
Furthermore, the researchers claim that few studies are based in education variables provided by information system in Brazilian health, because, not completeness of this variable is very high in all Brazil11. Similarly, a study in São Francisco (USA) showed 75% of hospitals never collected information about the patients´education16.
The completeness of the records in prenatal and delivery type variable are in good levels, with a variation of 3.5% to 23.5% and 4.1% to 26.2%, respectively. From the analyzed variables, these two variables had excellent completeness in at least one of the studied years.
The variables related to the use of antiretroviral by the mother and by the exposed child showed incompleteness of 9% to 20.9% and 7.5% to 28%, respectively, according to Table 2.
In the main measures of HIV transmission prevention in mothers for the concept and enabling the reduction of risk factors, there are the counseling and serology testing in pregnant women as early as possible. Therefore, prenatal is an important factor, for the diagnosis of HIV infection be offered with pre and post-test to all pregnant women at the first prenatal consultation, regardless of their apparent risk for HIV7.
Therefore, gaps in coverage testing during prenatal represent an alarming fact, since it effectively impossible to early adoption of therapeutic measures. In this study, the completeness of prenatal field was regular, showing the need to achieve a pattern of excellence for the information used in the evaluation of measures to prevent the VT from HIV.
Another key preventative measure in relation to vertical transmission of HIV is the choice and proper management of the birth routes7. Thereby, this study found a regular fulfillment of this item on the notification/investigation record, configuring a real need in the evaluation completing these forms.
It is noteworthy that among the analyzed variables a few had an excellent level of completeness, showing that there is no line between the epidemiological importance of the analyzed variables and their level of fulfillment. Finally, the qualitative analysis of the variables showed different levels of fulfillment for each evaluated field. Findings revealed significant failures in the fulfillment of the fields of investigation/notification records of positive HIV pregnant women and exposed child in Ceará between 2000 and 2009.
It is noteworthy that the losses in processing the data imply a mistaken estimate of the magnitude and burden of the epidemic leading to a sub-allocation of actions and resources for solving them. Furthermore, also prevent the implementation of prevention actions in order to reduce the progress of HIV infection in pregnant women and the VT virus resulting17.
Several studies have been developed in order to assess the quality of information contained in used as sources of information for the planning of health systems. These have been completed and pointed to the need for training for professionals responsible for fulfillment of notification/investigation records that feed information systems.
In this scenario, it was found that is very high the satisfaction level of the professionals who fill out records of three Health Information Systems on the scope of the information records/forms, though, many professionals have not been trained and have difficulties to filling of documents18.
The records have instructions for fulfillment in the attached document. However, they were little used since the deficiency of information found in this study. It is important to reinforce to those responsible professionals for reporting cases, the importance of the data provided by them so that the planning and subsequent implementation of actions compatible with the reality of the disease in is possible17.19.
CONCLUSION
The classification of the levels of fulfillment of the notification records of ISND ranged from terrible to excellent, according to the studied variables and the year. Thus, it is necessary to strengthen strategies that enhance the construction of databases suitable and reliable data, since such information provides a basis for research and public policy.
From the findings of this study, discussions about the theme related to database information systems are necessary, during the training of health professionals, especially for those professionals who will be responsible for the completion of the data records, contributing thereby to the reduction of incomplete data.
It is suggested to perform educational basis activities aiming to sensitize professionals responsible for work notifications, focusing on the importance of information for the prevention and control of the VT in the state of Ceará. It is worth highlighting the importance of quality of fulfillment of data present in the notification/investigation records.
It is important, however, that studies like this, made from secondary sources of information, have some limitations such as not being able to check the credibility of the recorded data, which can compromise the reliability of the assessment. Despite such obstacles, it is relevant to this study showing the local conditions over a given period, providing an evaluation of macroscopic shape of the analyzed problem.
REFERENCES
1. Barbará A, Sachetti VAR, Crepaldi MA. Contribuições das representações sociais ao estudo da aids. Interação Psicol. 2005; 9: 331-9.
2. Ministério da Saúde (Br). Secretaria de Vigilância em Saúde. Programa Nacional de DST e AIDS. Boletim Epidemiológico Aids e DST. Brasília (DF): Ministério da Saúde; 2009.
3. Secretaria da Saúde do Estado do Ceará (CE). Núcleo de Vigilância Epidemiológica. Informe Epidemiológico AIDS. [citado em 11 fev 2014]Disponível em:http://www.saude.ce.gov.br/index.php/boletins%3Fdownload%3D40%253Ainforme-aids+&cd=1&hl=pt-BR&ct=clnk&gl=br.
4. Ministério da Saúde (Br). Secretaria de Vigilância em Saúde. Programa Nacional de DST e Aids. Protocolo para a prevenção de transmissão vertical de HIV e sífilis - Manual de Bolso. Brasília (DF): Ministério da Saúde; 2007.
5. Ministério da Saúde (Br). Programa Nacional de DST e AIDS. Projeto Nascer. Brasília (DF): Ministério da Saúde; 2003.
6. Fonseca MGP, Coeli CM, Lucena FFA, Veloso VG, Carvalho MS. Accuracy of a probabilistic record linkage strategy applied to identify deaths among cases reported to the Brazilian AIDS surveillance database. Cad Saúde Pública. 2010; 26: 1431-8.
7. Ministério da Saúde (Br). Secretaria de Vigilância em Saúde. Departamento de Vigilância Epidemiológica. Sistema de Informação de Agravos de Notificação - SINAN: normas e rotinas. Brasília (DF): Ministério da Saúde; 2007.
8. Cerqueira ACB, Sales CMM, Lima R, Silva MZ, Vieira RCA, Brioschi AP, etal. Completude do sistema de informação de agravos de notificação compulsória de gestante HIV positivo entre 2001 e 2006, no Espírito Santo, Brasil. UFES revodontol. 2008; 10(1): 33-7.
9. Romanelli RMC, KakehasiFM, TavaresMCT, MeloVH, GoulartLHF, AguiarRALP,et al. Perfil das gestantes infectadas pelo HIV atendidas em pré-natal de alto risco de referência em Belo Horizonte. RevBrasSaudeMater Infant. 2006; 6: 329-34.
10. Romero DE, Cunha AB. Avaliação da qualidade das variáveis epidemiológicas e demográficas do Sistema de Informações sobre Nascidos Vivos, 2002. CadSaúde Pública. 2007; 23: 701-14.
11. Félix JD, Zandonade E, Amorim MHC, Castro DS. Avaliação da completude das variáveis epidemiológicas do Sistema de Informação sobre Mortalidade em mulheres com óbitos por câncer de mama na Região Sudeste - Brasil (1998 a 2007). Ciênc saúde coletiva. 2012; 17: 945-53.
12. Moreira CMM, Maciel ELN. Completude dos dados do programa de controle da tuberculose no sistema de informação de agravos de notificação no estado do Espírito Santo, Brasil: uma análise de 2001 a 2005. J BrasPneumol. 2008; 34: 225-9.
13. Macente LB, Zandonade E. Avaliação da completude do Sistema de Informação sobre Mortalidade por suicídio na região Sudeste, Brasil, no período de 1996 a 2007. J Bras Psiquiatr. 2010; 59: 173-81.
14. Almeida MVS, Amorim HC, Thule LCS, Zandonade E. Avaliação da Qualidade dos Dados do Sistema de Informação do Câncer do Colo do Útero em Vitória - ES, Brasil.RevBrascancerol. 2012; 58: 427-33.
15. Pinto IV, Ramos DN, Costa MCE, Ferreira CBT, Rebelo MS. Completude e consistência dos dados dos registros hospitalares de câncer no Brasil. CadSaúde Colet. 2012; 20: 113-20.
16. Gomez SL, Satariano W, Le GM, Weeks P, Mcclure L, West DW. Variability among hospitals and staff in collection of race, ethnicity, birthplace, and socioeconomic information in the greater San Francisco Bay Area.J Registry Manag. 2009; 36: 105-10.
17. Cerqueira ACB, Miranda AE, Maciel ELN. Completude do banco de dados de gestante HIV positivo e de aids em menores de treze anos do Sistema de Informação de Agravos de Notificação: Vitória, 2000 a 2006. Cad Saúde Colet. 2010; 18: 191-4.
18. Mota FRL. Registro de informação sistema de informação em saúde: um estudo das bases SINASC, SIAB e SIM no estado de Alagoas. Perspectivas em Ciências da Informação. 2009; 14: 238-9.
19. Fonte VRF, Spindola T, Martins ERC, Francisco MTR, Clos AC, Pinto RC. Conhecimento de gestantes de um hospitaluniversitário relacionado à prevenção de DST/AIDS. Revenferm UERJ. 2012; 20:493-9.