Melhora na qualidade e completitude da base de dados do Registro de Câncer de Base Populacional do município de São Paulo: uso das técnicas de linkage

Smoking, mortality, access to diagnosis, and treatment of lung cancer in Brazil

INTRODUCTION

Lung cancer (LC) is a relevant public health problem in Brazil and worldwide, given its high incidence and mortality. Thus, the objective of this study is to analyze the distribution of smoking and smoking status according to sociodemographic characteristics and disparities in access, treatment, and mortality due to LC in Brazil in 2013 and 2019.

METHOD

Retrospective study of triangulation of national data sources: a) analysis of the distribution of smoking, based on the National Survey of Health (PNS); b) investigation of LC records via Hospital-based Cancer Registry (HCR); and c) distribution of mortality due to LC in the Mortality Information System (SIM).

RESULTS

There was a decrease in the percentage of people who had never smoked from 2013 (68.5%) to 2019 (60.2%) and in smoking history (pack-years). This was observed to be greater in men, people of older age groups, and those with less education. Concerning patients registered in the HCR, entry into the healthcare service occurs at the age of 50, and only 19% have never smoked. While smokers in the population are mainly Mixed-race, patients in the HCR are primarily White. As for the initial stage (I and II), it is more common in White people and people who have never smoked. The mortality rate varied from 1.00 for people with higher education to 3.36 for people without education. Furthermore, White people have a mortality rate three times higher than that of Black and mixed-race people.

CONCLUSION

This article highlighted relevant sociodemographic disparities in access to LC diagnosis, treatment, and mortality. Therefore, the recommendation is to strengthen the Population-Based Cancer Registry and develop and implement a nationwide LC screening strategy in Brazil since combined prevention and early diagnosis strategies work better in controlling mortality from the disease and continued investment in tobacco prevention and control policies.

Deterministic linkage for improving follow-up time in a Brazilian population-based cancer registry

Population-based cancer registries (PBCR) are the primary source of cancer incidence and survival statistics. The loss to follow-up of these patients is concerning since it reduces the reliability of any statistical analysis. The linkage techniques have been increasingly used to improve data quality in various information systems. The linkage was performed between the databases of the PBCR-Barretos and the mortality database of the state of São Paulo. To evaluate the improvement in the follow-up time of patients, the comparability of the two databases, pre- and post linkage, was made. Three analyses were performed: a comparative analysis of the absolute number of deaths, a comparative analysis of the follow-up time of patients and the survival analysis. After linkage, there was an increase of 813 deaths. The follow-up time of patients was extended and observed in most types of tumours. The comparability of the survival analyses at both time points also showed a decrease in survival probabilities for all tumour types. Deterministic linkage is effective in updating the vital status of registered patients, improving patient follow-up time, and maintaining good quality data from PBCRs, consequently producing more reliable rates, as seen for the survival analyses.

Stark differences in cancer epidemiological data between GLOBOCAN and GBD: Emphasis on oral cancer and wider implications

BACKGROUND

GLOBOCAN 2020 and Global Burden of Disease (GBD) 2019 are the two most established global online cancer databases. It is important to examine the differences between the two platforms, to attempt to explain these differences, and to appraise the quality of the data. There are stark differences for lip and oral cancers (LOC) and we attempt to explain these by detailed analysis of ten countries at the extremes of differences.

METHODS

Age-standardised incidence rates (ASIR) of LOC were obtained from GLOBOCAN 2020 and GBD 2019. Five countries with the greatest and smallest fold differences were selected. A systematic search of PubMed and Embase electronic databases was then performed to identify publications reporting the incidence of LOC in the selected countries between 2015 and 2022. Specifically, data sources of the articles were examined and evaluated.

FINDINGS

For LOC, greatest differences were found in Papua New Guinea, Vietnam, China, Pakistan, and Indonesia (group A). In contrast, the United States of America (USA), Brazil, France, Germany, and India (group B) had the least differences between the two databases.

INTERPRETATION

It is not surprising that when GLOBOCAN and GBD could not obtain high-quality or accessible LOC data from national or local cancer registries, as in group A, discrepancies would be seen between the two online databases. In contrast, where only minor differences were seen between GLOBOCAN and GBD, as in group B, presumptively due to those countries having well-established cancer registries and healthcare administrative systems, the literature is more consistent. Moreover, many studies have grouped lip and oral cavity with pharynx and categorised outputs as "oral and oropharyngeal cancer" or "oral cavity and pharynx cancer". Those categorisations lacked subsite accuracy and failed to realise that oral cancer and oropharyngeal cancer have completely different etiological factors, pathogeneses, prognosis, and treatment outcomes.

FUNDING

This research received no specific grant or funding from any funding agency in the public, commercial, or not-for-profit sectors, and the authors received no financial support for the research, authorship, and/or publication of this article.

Procedures for health data linkage: applications in health surveillance

OBJECTIVE

To present a standardized methodology for linking different public health databases.

METHODS

This was a methodological review article specifically describing data processing procedures for deterministic linkage between structured databases. It instructs on how to: treat data, select linkage keys, and link databases using two databases simulated in R software.

RESULTS

The commands used for the deterministic linkage of the inner_join type were presented. The linkage process resulted in a database with 40,108 pairs using only the "Name" key. Adding the second key, "Name of mother", the resulted dropped to 112 pairs. By adding the third key, "Date of birth", only two pairs were identified.

CONCLUSION

Database linkage and its analysis are valid and valuable tools for health services in supporting health surveillance actions.

Analysis of Factors Associated with the Risk of Suicide in a Brazilian Capital: Cross-Sectional Study

Suicide and suicide attempts are considered global health problems. With regard to the main causes of unnatural deaths, Brazil ranks as the eighth country with the highest absolute number of suicides. The aim of this study was to analyze the factors associated with the risk of suicide in a metropolitan city in Brazil. This was a cross-sectional study carried out in Recife, in the northeast region of Brazil. Epidemiological surveillance was conducted on the information systems regarding attempted suicide, suicide, and undetermined deaths in the period from 2007 to 2017. A record linkage between the information systems was performed. Descriptive statistics, bivariate analysis, and logistic regression were performed with an adopted a significance level of 5%. In the city of Recife, there were 4495 suicide attempts in the period, and the most frequent suicide attempts were by single females, aged between 20 and 39 years old, and who used either medication or poison to die by suicide. A total of 141 individuals died by suicide, and exogenous intoxication was the most common method. Knowing the common characteristics and associated factors of those who attempt and die by suicide is key for the development of prevention policies and intervention strategies for suicide.

How to Estimate Food Prices and Diet Costs in Population-Based Studies?

Health and nutrition surveys usually comprise detailed information on health characteristics and food consumption of certain population groups; however, the lack of data collection on the food prices may pose challenges for the estimation of the diet costs for the comprehensive analysis of food demand. The Household Budget Survey (HBS) represents an opportunity to obtain the data on the food prices for the nutrition surveys in the diverse countries worldwide. Although the HBS and the health and nutrition surveys may correspond to different periods, the application of the appropriate Consumer Price Index (CPI) allows to address the changes in the relative prices to perform the linkage between the data of food consumption with information on the food prices. Therefore, the aim of this study was to present the methods for the extraction and linkage of the food prices data from the Brazilian HBS (2002–2003 and 2008–2009) by using the pairing features related to the household characteristics to match the Health Survey of São Paulo [Inquérito de Saúde de São Paulo (ISA-Capital)] conducted in 2003, 2008, and 2015. Data referring to the household characteristics and food prices acquired by the household members living in São Paulo municipality were selected from the HBS datasets for integration with the ISA-Capital dataset. Specific deflators referring to the food items surveyed in São Paulo were obtained from the datasets of the Brazilian Broad Consumer Price Index (BCPI). Therefore, the pairing criteria referring to time, location, and household characteristics were adopted to allow linking foods consumed by the individuals in the ISA-Capital with the prices from the foods acquired by household members interviewed in the HBS. Matching data on the key pairing criteria (location/year/household income per capita/number of residents/family profile) resulted in the linkage of 94.4% (2003), 92.6% (2008), and 81.2% of the cases (2015). Following the data linkage, it was possible to estimate diet costs per gram and per calorie including application of cooking and conversion factors. Data were presented in the International Monetary Unit under the purchasing power parity (PPP) to allow the comparison at the international level. The mean diet costs identified in the population of São Paulo municipality were $8.45 (dp = 0.38) per capita per day in 2003, $8.72 (dp = 0.24) per capita per day in 2008, and $9.62 (dp = 0.23) per capita per day in 2015. Thus, it was possible to estimate the diet costs based on the prices of food items through pairing linkage of information from the household surveys, such as the Brazilian HBS, with the health and nutrition surveys lacking information on the expenditures or prices such as the ISA-Capital. Similar procedures may be used in the diverse countries with availability of the datasets of the household expenditures and health and nutrition surveys, allowing the researchers worldwide to associate the diet quality with food demand.

Demographic and Clinical Outcomes of Brazilian Patients With Stage III or IV Non-Small-Cell Lung Cancer: Real-World Evidence Study on the Basis of Deterministic Linkage Approach

Non–small-cell lung cancer (NSCLC) is the most common type of lung cancer and accounts for 80%-90% of the cases. In Brazil, between 2018 and 2019, lung cancer was ranked as the second most frequent cancer among men and the fourth among women. The primary objectives were to describe the journey and survival rates of patients with advanced NSCLC treated in the Brazilian private health care system (HCS).

Agreement and validity of asbestos-related cancers in the hospital information system of the Brazilian Unified Health System

OBJECTIVE

To estimate the degree of agreement and validity of diagnoses of asbestos-related malignant neoplasms registered in the Hospital Information System of the Brazilian Unified Health System (SIH/SUS), in comparison to the Hospital Cancer Registries of the State of São Paulo (HCR/SP).

METHODS

Deaths with records of malignant neoplasms associated with asbestos were identified and extracted from SIH/SUS between 2007 and 2014. Deaths in cases registered in the HCR/SP were extracted for the same period. The databases were linked using software Link Plus. A single ICD-10-coded diagnosis selected from each system was analyzed. The proportion of agreement, and the sensitivity, specificity and predictive values were estimated.

RESULTS

19,458 pairs were found with records in both bases. The proportion of agreement was high, ranging from 92.4% for the unknown primary site, to 99.7% for cancer of the pleura. The Kappa Index ranged from 0.05 (95%CI 0.04 - 0.07) for cancer of the pleura to 0.85 (95%CI 0.84 - 0.87) for lung cancer. Sensitivity varied from 0.08 (95%CI 0.01 - 0.25) for cancer of the pleura, to 0.90 (95%CI 0.90 - 0.91) for lung cancer.

CONCLUSION

Diagnosis of asbestos-related malignancies reached higher levels of agreement and validity when common. Rare diagnoses showed low accuracy in SIH/SUS.

Administrative Data Linkage in Brazil: Potentials for Health Technology Assessment

Health technology assessment (HTA) is the systematic evaluation of the properties and impacts of health technologies and interventions. In this article, we presented a discussion of HTA and its evolution in Brazil, as well as a description of secondary data sources available in Brazil with potential applications to generate evidence for HTA and policy decisions. Furthermore, we highlighted record linkage, ongoing record linkage initiatives in Brazil, and the main linkage tools developed and/or used in Brazilian data. Finally, we discussed the challenges and opportunities of using secondary data for research in the Brazilian context. In conclusion, we emphasized the availability of high quality data and an open, modern attitude toward the use of data for research and policy. This is supported by a rigorous but enabling legal framework that will allow the conduct of large-scale observational studies to evaluate clinical, economical, and social impacts of health technologies and social policies.

Data Reliability and Coding Completeness of Cancer Registry Information Using Reabstracting Method in the National Cancer Institute: Thailand, 2012 to 2014

PURPOSE

Data quality is a core value of cancer registries, which bring about greater understanding of cancer distribution and determinants. Thailand established its cancer registry in 1986; however, studies focusing on data reliability have been limited. This study aimed to assess the coding completeness and reliability of the National Cancer Institute (NCI) hospital-based cancer registry, Thailand.

METHODS

This study was conducted using the reabstracting method. We focused on seven cancer sites-the colon, rectum, liver, lung, breast, cervix, and prostate-registered between 2012 and 2014 in the NCI hospital-based cancer registry. Missing data were identified for coding completeness calculation among important variables. The agreement rate and κ coefficient were computed to represent data reliability.

RESULTS

For reabstracting, we retrieved 957 medical records from a total of 5,462. These were selected using the probability proportional to size method, stratified by topology, sex, and registered year. The overall coding completeness of the registered and reabstracted data was 89.9% and 93.6%, respectively. In addition, the overall agreement rate among variables ranged from 84.7% to 99.6%, and κ coefficient ranged from 0.619 to 0.995. The misclassification among unilateral organs caused lower coding completeness and agreement rate of laterality coding. The completeness of current residency could be improved using the reabstracting method. The lowest agreement rate was found among various categories of diagnosis basis. Sex misclassification for male breast cancer was identified.

CONCLUSION

The coding completeness and data reliability of the NCI hospital-based cancer registry met the standard in most critical variables. However, some challenges remain to improve the data quality. The reabstracting method could identify the critical points affecting the quality of cancer registry data.

Simulação e comparação de técnicas de correção de dados incompletos de idade para o cálculo de taxas de incidência

O objetivo foi comparar duas técnicas para estimar idade em bancos de dados com registros incompletos e analisar sua aplicação no cálculo da incidência de câncer. Utilizou-se a base de dados do Registro de Câncer de Base Populacional do Município de São Paulo, Brasil, contendo casos diagnosticados por câncer do trato urinário, entre 1997 e 2013. Foram aplicadas duas técnicas para estimativa de idade: fator de correção e imputação múltipla. Foram simuladas, usando a distribuição binomial, seis bases de dados com diferentes proporções de dados incompletos para idade de 5% até 50%. A razão entre as incidências foi calculada tendo, como referência, a base completa, cuja incidência padronizada foi de 11,83/100 mil; as demais incidências nas bases com 5% ou mais de dados incompletos para idade apresentaram-se subestimadas. Ao aplicar o fator de correção, as taxas corrigidas não apresentaram diferenças em comparação com as padronizadas, entretanto, essa técnica não permite corrigir taxas específicas. A imputação múltipla foi útil na correção das taxas padronizadas e específicas em bancos com até 30% de dados incompletos, entretanto, as taxas específicas para indivíduos com menos de 50 anos apresentaram-se subestimadas. Bases com 5% ou mais de dados incompletos necessitam de aplicação de correção. A imputação múltipla, apesar de complexa em sua execução, mostrou-se superior ao fator de correção. Todavia, deve ser utilizada com parcimônia, pois taxas específicas por idade podem manter-se subestimadas.

Recovering missing mesothelioma deaths in death certificates using hospital records

BACKGROUND

In Brazil, underreporting of mesothelioma and cancer of the pleura (MCP) is suspected to be high. Records from death certificates (SIM) and hospital registers (SIH-SUS) can be combined to recover missing data but only anonymous databases are available. This study shows how common data can be used for linkage and as an assessment of accuracy.

METHODS

Mesothelioma (all sites, ICD-10 codes C45.0-C45.9) and cancer of the pleura (C38.4) were retrieved from both information systems and combined using a linkage algorithm. Accuracy was examined with non-anonymous databases, limited to the state of São Paulo.

RESULTS

We found 775 cases in death certificates and 283 in hospital registers. The linkage matched 57 cases, all accurately paired. Three cases, 0.4% in SIM and 1.3% in SIH-SUS, could not be matched because of data inconsistencies.

CONCLUSIONS

A computer linkage can recover MCP cases from hospital records not found in death certificates in Brazil.