Maternal health surveillance panel: a tool for expanding epidemiological surveillance of women's health and its determinants.

OBJECTIVE: To present the methodology used in the development of two products for maternal health surveillance and its determinants and discuss their possible uses. METHODS: Based on a theoretical model of the determinants of maternal death and databases of Brazilian health information systems, two free products were developed: an interactive panel "surveillance of maternal health" and an educational material "Aparecida: a story about the vulnerability of Brazilian women to maternal death", both available on the website of the Brazilian Obstetric Observatory. RESULTS: More than 30 indicators were calculated for the period 2012-2020, containing information on socioeconomic conditions and access to health services, reproductive planning, prenatal care, delivery care, conditions of birth and maternal mortality and morbidity. The indicators related to severe maternal morbidity in public hospitalizations stand out, calculated for the first time for the country. The panel allows analysis by municipality or aggregated by health region, state, macro-region and country; historical series analysis; and comparisons across locations and with benchmarks. Information quality data are presented and discussed in an integrated manner with the indicators. In the educational material, visualizations with national and international data are presented, aiming to help in the understanding of the determinants of maternal death and facilitate the interpretation of the indicators. CONCLUSION: It is expected that the two products have the potential to expand epidemiological surveillance of maternal health and its determinants, contributing to the formulation of health policies and actions that promote women's health and reduce maternal mortality.

Maternal health surveillance panel: a tool for expanding epidemiological surveillance of women's health and its determinants / Painel de vigilância da saúde materna: uma ferramenta para ampliação da vigilância epidemiológica da saúde das mulheres e seus determinantes

ABSTRACT Objective: To present the methodology used in the development of two products for maternal health surveillance and its determinants and discuss their possible uses. Methods: Based on a theoretical model of the determinants of maternal death and databases of Brazilian health information systems, two free products were developed: an interactive panel "surveillance of maternal health" and an educational material "Aparecida: a story about the vulnerability of Brazilian women to maternal death", both available on the website of the Brazilian Obstetric Observatory. Results: More than 30 indicators were calculated for the period 2012-2020, containing information on socioeconomic conditions and access to health services, reproductive planning, prenatal care, delivery care, conditions of birth and maternal mortality and morbidity. The indicators related to severe maternal morbidity in public hospitalizations stand out, calculated for the first time for the country. The panel allows analysis by municipality or aggregated by health region, state, macro-region and country; historical series analysis; and comparisons across locations and with benchmarks. Information quality data are presented and discussed in an integrated manner with the indicators. In the educational material, visualizations with national and international data are presented, aiming to help in the understanding of the determinants of maternal death and facilitate the interpretation of the indicators. Conclusion: It is expected that the two products have the potential to expand epidemiological surveillance of maternal health and its determinants, contributing to the formulation of health policies and actions that promote women's health and reduce maternal mortality.

RESUME Objetivo: Apresentar a metodologia utilizada no desenvolvimento de dois produtos para a vigilância da saúde materna e seus determinantes e discutir as suas possíveis utilizações. Métodos: A partir de modelo teórico dos determinantes do óbito materno e bases de dados dos sistemas de informação em saúde brasileiros, foram desenvolvidos dois produtos gratuitos: um painel interativo denominado "Vigilância da saúde materna" e um material educativo chamado "Aparecida: uma história sobre a vulnerabilidade da mulher brasileira à morte materna", ambos disponíveis no site do Observatório Obstétrico Brasileiro. Resultados: Foram calculados mais de 30 indicadores para o período 2012-2020, contendo informações sobre condições socioeconômicas e de acesso a serviços de saúde, planejamento reprodutivo, assistência pré-natal, assistência ao parto, condições de nascimento e mortalidade e morbidade materna. Destacam-se os indicadores relacionados à morbidade materna grave em internações públicas, calculados pela primeira vez para o país. O painel permite análises por município ou agregadas por região de saúde, unidade da federação, macrorregião e país; análises de série histórica; e comparações entre localidades e com padrões de referência. Dados de qualidade da informação são apresentados e discutidos de forma integrada aos indicadores. No material educativo, visualizações com dados nacionais e internacionais são apresentadas, visando auxiliar na compreensão dos determinantes do óbito materno e facilitar a interpretação dos indicadores. Conclusão: Espera-se que os produtos tenham o potencial de ampliar a vigilância epidemiológica da saúde materna e seus determinantes, contribuindo para a formulação de políticas e ações de saúde que promovam a saúde das mulheres e reduzam a mortalidade materna.

Racial inequalities in mental healthcare use and mortality: a cross-sectional analysis of 1.2 million low-income individuals in Rio de Janeiro, Brazil 2010-2016.

INTRODUCTION: Mental health inequalities across racial and ethnic groups are large and unjust in many countries, yet these inequalities remain under-researched, particularly in low-income and middle-income countries such as Brazil. This study investigates racial and socioeconomic inequalities in primary healthcare usage, hospitalisation and mortality for mental health disorders in Rio de Janeiro, Brazil. METHODS: A cohort of 1.2 million low-income adults from Rio de Janeiro, Brazil with linked socioeconomic, demographic, healthcare use and mortality records was cross-sectionally analysed. Poisson regression models were used to investigate associations between self-defined race/colour and primary healthcare (PHC) usage, hospitalisation and mortality due to mental disorders, adjusting for socioeconomic factors. Interactions between race/colour and socioeconomic characteristics (sex, education level, income) explored if black and pardo (mixed race) individuals faced compounded risk of adverse mental health outcomes. RESULTS: There were 272 532 PHC consultations, 10 970 hospitalisations and 259 deaths due to mental disorders between 2010 and 2016. After adjusting for a wide range of socioeconomic factors, the lowest PHC usage rates were observed in black (adjusted rate ratio (ARR): 0.64; 95% CI 0.60 to 0.68; compared with white) and pardo individuals (ARR: 0.87; 95% CI 0.83 to 0.92). Black individuals were more likely to die from mental disorders (ARR: 1.68; 95% CI 1.19 to 2.37; compared with white), as were those with lower educational attainment and household income. In interaction models, being black or pardo conferred additional disadvantage across mental health outcomes. The highest educated black (ARR: 0.56; 95% CI 0.47 to 0.66) and pardo (ARR: 0.75; 95% CI 0.66 to 0.87) individuals had lower rates of PHC usage for mental disorders compared with the least educated white individuals. Black individuals were 3.7 times (ARR: 3.67; 95% CI 1.29 to 10.42) more likely to die from mental disorders compared with white individuals with the same education level. CONCLUSION: In low-income individuals in Rio de Janeiro, racial/colour inequalities in mental health outcomes were large and not fully explainable by socioeconomic status. Black and pardo Brazilians were consistently negatively affected, with lower PHC usage and worse mental health outcomes.

Associations between primary healthcare and infant health outcomes: a cohort analysis of low-income mothers in Rio de Janeiro, Brazil.

Background: Expanding primary healthcare to urban poor populations is a priority in many low-and middle-income countries and is essential to achieve universal health coverage (UHC). Between 2008 and 2016 the city of Rio de Janeiro undertook an ambitious programme to rapidly expand primary care to low-income areas through the family health strategy (FHS). Infant health impacts of this roll out are unknown. This study examines associations between maternal FHS utilisation and birth outcomes, neonatal and infant mortality. Methods: A cohort of 75,339 live births (January 2009-December 2014) to low-income mothers in Rio de Janeiro was linked to primary care, birth, hospital and death records. The relationship between maternal FHS use and infant health outcomes was assessed through logistic regression with inverse probability treatment weighting and regression adjustment. Socioeconomic inequalities in the associations between FHS use and outcomes were explored through interactions. Primary outcomes were neonatal and infant death. Thirteen secondary outcomes were also examined to explore other important health outcomes and potential mechanisms. Results: A total of 9002 (12.0%) infants were born to mothers in the cohort who used FHS services either before pregnancy or in the first two trimesters. There was a total of 527 neonatal and 893 infant deaths. Maternal FHS usage during the first two trimesters was associated with substantial reductions in neonatal [adjusted odds ratio (aOR): 0.527, 95% confidence interval (95% CI): 0.345; 0.806] and infant mortality (aOR: 0.672, 95% CI: 0.48; 0.924). Infants born to lower-income mothers and those without formal employment had larger reductions in neonatal and infant mortality associated with FHS use. Maternal FHS in the first two trimesters use was also associated with more antenatal care consultations and a lower risk of low birth weight and preterm birth. Interpretation: Expanding primary care to low-income populations in Rio de Janeiro was associated with improved infant health and health equity benefits. Funding: DFID/MRC/Wellcome Trust/ESRC.

[Evolution of the completeness and consistency of the gastroschisis registry in the Brazilian Live Birth Information System, from 2005 to 2020]. / Evolução da completude e consistência do registro de gastrosquise no Sistema de Informações sobre Nascidos Vivos no Brasil, de 2005 a 2020.

This study aimed to evaluate the evolution of the completeness and consistency of the gastroschisis registry in the Brazilian Live Birth Information System (SINASC). It is a time-series study on the completeness of the variable "occurrence of congenital anomaly" and the consistency of gastroschisis diagnosis in SINASC, in biennia from 2005 to 2020, for federative units, region, and Brazil. The consistency was estimated by the ratio between deaths from gastroschisis registered in the Brazilian Mortality Information System (SIM) and the total number of cases recorded in SINASC. Temporal trend was analyzed by joinpoint regression. In the period, 46,574,995 live births and 10,024 cases of gastroschisis were recorded. A total of 5,632 infant deaths due to gastroschisis were identified. The percentage of incompleteness decreased from 6.52% to 1.87%, with an annual percentage variation (APV) of -14.5%, and completeness reached excellence (≤ 5% of incompleteness), except in the Central-West Region. Case/death ratios above 1 were found in the North and Northeast regions and in some federative units in the Central-West, but there was a decrease, approaching the mortality found in studies in the South and Southeast regions. Its reduction was more pronounced until 2009-2010 (APV = -10.7%) and smaller later (APV = -4.4%). The quality of the gastroschisis registry reflects regional differences in the overall quality of SINASC, constituting as a marker for malformations that require complex neonatal care.

O objetivo deste estudo foi avaliar a evolução da completude e da consistência do registro de gastrosquise no Sistema de Informações sobre Nascidos Vivos (SINASC) no Brasil. Trata-se de estudo de série temporal sobre a completude da variável "ocorrência de anomalia congênita" e a consistência do diagnóstico de gastrosquise no SINASC, nos biênios entre 2005 e 2020, para Unidades da Federação, região e Brasil. A consistência foi estimada pela razão entre óbitos por gastrosquise registrados no Sistema de Informações sobre Mortalidade (SIM) e o total de casos registrados no SINASC. A tendência temporal foi analisada por regressão joinpoint. No período, havia registro de 46.574.995 nascidos vivos e 10.024 casos de gastrosquise entre eles. Identificamos 5.632 óbitos infantis por gastrosquise. O percentual de incompletude diminuiu de 6,52% para 1,87%, com variação percentual anual (VPA) de -14,5%, e a completude atingiu a excelência (≤ 5% de incompletude), exceto no Centro-oeste do país. Razão óbito/caso acima de 1 foi encontrada nas regiões Norte e Nordeste e em alguns estados do Centro-oeste, mas houve diminuição, aproximando-se da mortalidade encontrada em estudos no Sul e Sudeste. Sua redução foi mais acentuada até 2009-2010 (VPA = -10,7%) e menor posteriormente (VPA = -4,4%). A qualidade do registro de gastrosquise reflete as diferenças regionais da qualidade geral do SINASC, configurando-se uma condição marcadora para malformações que demandam atenção neonatal complexa.

El objetivo de este estudio fue evaluar la evolución de la completitud y consistencia del registro de la gastrosquisis en el Sistema de Información de los Nacidos Vivos (SINASC) en Brasil. Se trata de un estudio de serie temporal sobre la completitud de la variable "aparición de anomalía congénita" y la consistencia del diagnóstico de gastrosquisis en el SINASC, en los bienios entre 2005 y 2020, en estados, regiones y Brasil. La consistencia se estimó por la razón entre las defunciones por gastrosquisis registradas en el Sistema de Información de Mortalidad (SIM) y el total de casos inscritos en el SINASC. La tendencia temporal se analizó mediante regresión joinpoint. Durante el período, hubo 46.574.995 nacidos vivos y 10.024 casos de gastrosquisis entre ellos. Identificamos 5.632 muertes infantiles por gastrosquisis. El porcentaje de incompletitud disminuyó de 6,52% a 1,87%, con una variación porcentual anual (VPA) de -14,5%, y la completitud alcanzó la excelencia (≤ 5% de incompletitud), excepto en la región Centro-oeste del país. Se encontró una relación defunciones/caso superior a 1 en las regiones Norte y Nordeste y en algunos estados del Centro-oeste, pero hubo una disminución, acercándose a la mortalidad encontrada en estudios en el Sur y Sudeste. Su reducción fue más pronunciada hasta 2009-2010 (VPA = -10,7%) y después menor (VPA = -4,4%). La calidad del registro de gastrosquisis refleja diferencias regionales relacionadas con la calidad general del SINASC, configurando una condición manifiesta de malformaciones que demandan cuidados neonatales complejos.

The effect modification of occupational social class in the association between sex and type 2 diabetes: results from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil).

We evaluated data from 14,156 baseline participants of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) collected from 2008 to 2010, to analyze the effect modification of occupational social class on the association between sex and prevalence of type 2 diabetes. The crude and age-adjusted prevalence, according to sex and occupational social class, were estimated using generalized linear models with binomial distribution and logarithmic link function. This model was also used to estimate prevalence ratios (PR), adjusting for age group, race/skin color, and maternal education. The effect modification was measured in the multiplicative and additive scales. Males had higher crude and age-adjusted prevalence in all occupational social class strata. As occupational social class increases, the prevalence among males and females decreases. The PR of males to females decreased according to occupational class: 66% (PR = 1.66; 95%CI: 1.44; 1.90), 39% (PR = 1.39; 95%CI: 1.02; 1.89), and 28% (PR = 1.28; 95%CI: 0.94; 1.75) in the high, middle, and low occupational social classes, respectively. We found an inverse effect of the occupational social class on the association between sex and type 2 diabetes on the multiplicative scale, suggesting that it acts as an effect modifier.

Trends in prevalence of gastroschisis in Brazil from 2007 to 2020: A national population-based cross-sectional study.

BACKGROUND: Gastroschisis is a congenital abnormality of the abdominal wall which worldwide prevalence is increasing with time. Up to now, young maternal age is the main factor associated with gastroschisis. Brazil has a great proportion of adolescent mothers. OBJECTIVES: To estimate prevalence of gastroschisis in Brazil, explore the effect of very young maternal age, and analyze regional distribution and time trends. METHODS: Cross-sectional population-based study using data from official national information systems from 2007 to 2020, covering 9,831 gastroschisis cases. Low maternal age was assessed as risk factor through odds ratio. Prevalence time trends by maternal age and geographic area were analyzed using joinpoint regression. RESULTS: Overall prevalence of gastroschisis in Brazil was 2.47 per 10,000 births. From first to last biennium, there was a 23% relative increase. Last biennium prevalence was higher in the Central-west region and lower in the Northeast. Compared to mothers of 20-34 years old (YO), the odds-ratio for prevalence and 95% confidence intervals for gastroschisis were, as follows: (a) 10-14 YO-4.9 (4.3,5.5); (b) 15-19 YO-3.6 (3.5,3.8); and (c) 35-49 YO-0.3 (95% CI 0.3, 0.4). CONCLUSIONS: Prevalence of gastroschisis in Brazil relatively increased 23% from 2007 to 2020. This happened in all maternal age groups. Differences between regions may be associated to data quality. Young mothers had a higher prevalence of gastroschisis and still higher for those under 15 YO. Mother age over 34 YO was a protective factor.

The effect modification of occupational social class in the association between sex and type 2 diabetes: results from the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) / O efeito da modificação da classe social ocupacional na associação entre sexo e diabetes tipo 2: resultados do Estudo Longitudinal de Saúde do Adulto no Brasil (ELSA-Brasil) / El efecto del cambio de clase social ocupacional en la asociación entre sexo y diabetes tipo 2: resultados del Estudio Longitudinal de Salud del Adulto en Brasil (ELSA-Brasil)

We evaluated data from 14,156 baseline participants of the Brazilian Longitudinal Study of Adult Health (ELSA-Brasil) collected from 2008 to 2010, to analyze the effect modification of occupational social class on the association between sex and prevalence of type 2 diabetes. The crude and age-adjusted prevalence, according to sex and occupational social class, were estimated using generalized linear models with binomial distribution and logarithmic link function. This model was also used to estimate prevalence ratios (PR), adjusting for age group, race/skin color, and maternal education. The effect modification was measured in the multiplicative and additive scales. Males had higher crude and age-adjusted prevalence in all occupational social class strata. As occupational social class increases, the prevalence among males and females decreases. The PR of males to females decreased according to occupational class: 66% (PR = 1.66; 95%CI: 1.44; 1.90), 39% (PR = 1.39; 95%CI: 1.02; 1.89), and 28% (PR = 1.28; 95%CI: 0.94; 1.75) in the high, middle, and low occupational social classes, respectively. We found an inverse effect of the occupational social class on the association between sex and type 2 diabetes on the multiplicative scale, suggesting that it acts as an effect modifier.

Nós avaliamos dados de 14.156 participantes do Estudo Longitudinal de Saúde do Adulto no Brasil (ELSA-Brasil) coletados entre 2008 e 2010 para analisar o efeito de modificação da classe social ocupacional na associação entre sexo e prevalência de diabetes tipo 2. A prevalência bruta e ajustada por idade, de acordo com sexo e classe social ocupacional, foram estimadas usando modelos lineares generalizados com distribuição binomial e função de ligação de logaritmo. Esse modelo também foi utilizado para estimar razões de prevalência (RP), ajustando para faixa etária, raça e escolaridade materna. Medimos a modificação do efeito nas escalas multiplicativa e aditiva. Os homens apresentaram prevalência bruta e ajustada por idade mais alta em todos os estratos de classe social ocupacional. À medida que a classe social ocupacional aumenta, há uma redução na prevalência entre homens e mulheres. A RP de homens para mulheres diminuiu de acordo com a classe ocupacional: foi de 66% (RP = 1,66; IC95%: 1,44; 1,90), 39% (RP = 1,39; IC95%: 1,02; 1,89) e 28% (RP = 1,28; IC95%: 0,94; 1,75) nas classes sociais ocupacionais alta, média e baixa, respectivamente. Houve um efeito inverso da classe social ocupacional na associação entre sexo e diabetes tipo 2 na escala multiplicativa, sugerindo que ela atua como um modificador de efeito.

Evaluamos datos de 14.156 participantes del Estudio Longitudinal de Salud de Adultos en Brasil (ELSA-Brasil) recopilados entre 2008 y 2010 para analizar el efecto del cambio de clase social ocupacional en la asociación entre género y prevalencia de diabetes tipo 2. La prevalencia bruta y ajustada por edad según el sexo y la clase social ocupacional se estimaron utilizando modelos lineales generalizados con distribución binomial y función de enlace logarítmico. Este modelo también se utilizó para estimar las razones de prevalencia (RP) ajustando por grupo de edad, raza y educación materna. Medimos la modificación del efecto en las escalas multiplicativa y aditiva. Los hombres tuvieron mayor prevalencia bruta y ajustada por edad en todos los estratos de clase social ocupacional. A medida que aumenta la clase social ocupacional, se reduce la prevalencia entre hombres y mujeres. La RP de hombres a mujeres disminuyó de acuerdo con la clase ocupacional: fue del 66% (RP = 1,66; IC95%: 1,44; 1,90), 39% (RP = 1,39; IC95%: 1,02; 1,89) y 28% (RP = 1,28; IC95%: 0,94; 1,75) en las clases sociales ocupacionales alta, media y baja, respectivamente. Hubo un efecto inverso de la clase social ocupacional en la asociación entre el sexo y la diabetes tipo 2 en la escala multiplicativa, lo que sugiere que actúa como un modificador del efecto.

Expansion of primary healthcare and emergency hospital admissions among the urban poor in Rio de Janeiro Brazil: A cohort analysis.

Background: Robust evidence on the relationship between primary care and emergency admissions is lacking in low- and middle-income countries. This study evaluates how the phased roll out of the family health strategy (FHS) to the urban poor in Rio de Janeiro Brazil affected emergency hospital admissions and readmissions from ambulatory-care sensitives conditions (ACSCs). Methods: A cohort of 1.2 million adults in Rio de Janeiro city were followed for five years (Jan 2012 to Dec 2016). The association between FHS use and the likelihood of emergency hospital admissions and 30-day readmissions were evaluated using multi-level Poisson regression models with inverse probability treatment weighting and regression adjustment (IPTW-RA) for socioeconomic and household characteristics. Inequalities in associations were examined across groups of causes and by key socioeconomic groups. Results: Records from 2,551,934 primary care consultations and 15,627 admissions were analysed. In IPTW-RA analyses, each additional FHS consultation was associated with a 3% lower rate of ACSC admission (RR: 0.97; 95%CI: 0.95, 0.98), a 63% lower rate of 30-day readmissions from any non-birth cause (RR: 0.37; 95%CI: 0.30, 0.46), and an 57% lower rate of 30-day readmissions from ACSCs (RR: 0.43; 95%CI: 0.33, 0.55). Individuals who were older, had the lowest educational attainment, were unemployed, and had higher incomes had larger reductions in ACSC admissions associated with FHS use. Interpretation: Investment in primary care is important for reducing emergency hospital admissions and their associated costs in LMICs. Funding: DFID/MRC/Wellcome Trust/ESRC.

Racial and socioeconomic disparities in multimorbidity and associated healthcare utilisation and outcomes in Brazil: a cross-sectional analysis of three million individuals.

BACKGROUND: Evidence is limited on racial/ethnic group disparities in multimorbidity and associated health outcomes in low- and middle-income countries hampering effective policies and clinical interventions to address health inequalities. METHODS: This study assessed race/ethnic and socioeconomic disparities in the prevalence of multimorbidity and associated healthcare utilisation, costs and death in Rio de Janeiro, Brazil. A cross-sectional analysis was carried out of 3,027,335 individuals registered with primary healthcare (PHC) services. Records included linked data to hospitalisation, mortality, and welfare-claimant (Bolsa Família) records between 1 Jan 2012 and 31 Dec 2016. Logistic and Poisson regression models were carried out to assess the likelihood of multimorbidity (two or more diagnoses out of 53 chronic conditions), PHC use, hospital admissions and mortality from any cause. Interactions were used to assess disparities. RESULTS: In total 13,509,633 healthcare visits were analysed identifying 389,829 multimorbid individuals (13%). In adjusted regression models, multimorbidity was associated with lower education (Adjusted Odds Ratio (AOR): 1.26; 95%CI: 1.23,1.29; compared to higher education), Bolsa Família receipt (AOR: 1.14; 95%CI: 1.13,1.15; compared to non-recipients); and black race/ethnicity (AOR: 1.05; 95%CI: 1.03,1.06; compared to white). Multimorbidity was associated with more hospitalisations (Adjusted Rate Ratio (ARR): 2.75; 95%CI: 2.69,2.81), more PHC visits (ARR: 3.46; 95%CI: 3.44,3.47), and higher likelihood of death (AOR: 1.33; 95%CI: 1.29,1.36). These associations were greater for multimorbid individuals with lower educational attainment (five year probability of death 1.67% (95%CI: 1.61,1.74%) compared to 1.13% (95%CI: 1.02,1.23%) for higher education), individuals of black race/ethnicity (1.48% (95%CI: 1.41,1.55%) compared to 1.35% (95%CI: 1.31,1.40%) for white) and individuals in receipt of welfare (1.89% (95%CI: 1.77,2.00%) compared to 1.35% (95%CI: 1.31,1.38%) for non-recipients). CONCLUSIONS: The prevalence of multimorbidity and associated hospital admissions and mortality are greater in individuals with black race/ethnicity and other deprived socioeconomic groups in Rio de Janeiro. Interventions to better prevent and manage multimorbidity and underlying disparities in low- and middle-income country settings are needed.

Record linkage under suboptimal conditions for data-intensive evaluation of primary care in Rio de Janeiro, Brazil.

BACKGROUND: Linking Brazilian databases demands the development of algorithms and processes to deal with various challenges including the large size of the databases, the low number and poor quality of personal identifiers available to be compared (national security number not mandatory), and some characteristics of Brazilian names that make the linkage process prone to errors. This study aims to describe and evaluate the quality of the processes used to create an individual-linked database for data-intensive research on the impacts on health indicators of the expansion of primary care in Rio de Janeiro City, Brazil. METHODS: We created an individual-level dataset linking social benefits recipients, primary health care, hospital admission and mortality data. The databases were pre-processed, and we adopted a multiple approach strategy combining deterministic and probabilistic record linkage techniques, and an extensive clerical review of the potential matches. Relying on manual review as the gold standard, we estimated the false match (false-positive) proportion of each approach (deterministic, probabilistic, clerical review) and the missed match proportion (false-negative) of the clerical review approach. To assess the sensitivity (recall) to identifying social benefits recipients' deaths, we used their vital status registered on the primary care database as the gold standard. RESULTS: In all linkage processes, the deterministic approach identified most of the matches. However, the proportion of matches identified in each approach varied. The false match proportion was around 1% or less in almost all approaches. The missed match proportion in the clerical review approach of all linkage processes were under 3%. We estimated a recall of 93.6% (95% CI 92.8-94.3) for the linkage between social benefits recipients and mortality data. CONCLUSION: The adoption of a linkage strategy combining pre-processing routines, deterministic, and probabilistic strategies, as well as an extensive clerical review approach minimized linkage errors in the context of suboptimal data quality.