The article analyzes a database of causes of death compiled from individual parish registers for the town of Detva located in the present-day territory of Slovakia. The database covers the period from 1861 to 1885 and contains 9,600 entries. The town and its environs reflect the population of the seventh-largest settlement in Slovakia, which ranged from 10,000 to 12,000 inhabitants during the period under study. The population was 97% Catholic and predominantly Slovak-speaking. The article presents the motives for creating the database, the methods used to construct it, and the specific features of the parish registers from which the data were drawn. After construction, a basic classification of causes of death was performed according to the HistCat methodology. Across the entire database, the most common reasons for death came from infections, accounting for 25% of the total. The second most common category was diagnoses classified as ill-defined, with a 12.3% share, and the third most common category was tuberculosis (11.8%). These results confirm that the diagnosis of diseases displayed evident shortcomings, especially since the cause of death was not determined by a doctor, thus presenting new interpretative challenges.
This article examines cause-of-death statistics in Sweden during the Finnish War of 1808–1809 through an in-depth analysis of the so-called Wegelius Journal from the military hospital in Röbäck, near Umeå. Kept by the physician Jakob Esaias Wegelius, the journal constitutes a rare and detailed individual-level source documenting admissions, diagnoses, medical outcomes, and deaths among soldiers treated during the final phase of the war. The study situates the journal within the broader Swedish tradition of population registration and mortality statistics, particularly the work of Tabellverket, while emphasising its complementary character as a source based on contemporaneous military hospital records. The historical context is the large-scale evacuation of sick and wounded soldiers from Finland to northern Sweden in the autumn of 1808, which overwhelmed local infrastructure and resulted in extremely high mortality. The journal contains more than 1,100 admission records from soldiers belonging to over 30 Swedish military units, many from the eastern parts of the realm that after the war became Russian-ruled Finland. Analysis of the material shows that epidemic diseases, especially typhus and dysentery, accounted for the majority of illness and death, far exceeding battle-related injuries, although war wounds were more common among those early admitted. A central contribution of the article is its detailed description of the digitisation process. The handwritten journal was transcribed, structured, and standardised into a research database while preserving the original content of the source. Dates, regimental affiliations, and diagnoses were harmonised, and medical terminology was mapped both to the historical disease categories used by Tabellverket, the Swedish agency responsible for population statistics, and to the ICD10h classification system for historical causes of death. An additional category was introduced to better identify combat-related injuries to the Tabellverket categories. The digitised database enables fine-grained analyses of military morbidity and mortality during the crisis. As such, it makes the Wegelius Journal readily usable for research in fields such as historical demography, military medical history, and historical epidemiology.
Tracing vital events in the 19th century across Albanian provinces is a complex research endeavour related to the organisation of Albanian population into three different communities (Muslim, Catholic and Orthodox) under the Millet system applied in the Ottoman Empire, which grouped population by religion rather than ethnicity. The registration of births, deaths, and marriages was the responsibility of the respective religious communities, and as a result, the methods, regularity, language and quality of these records vary. This study draws on archival sources from the Central State Archives in Tirana, mainly from the Shkodra Archdiocese collection, to analyse how deaths among the Catholic population were reported and registered. We analyse the accuracy and reliability of the registers, and the main causes of death as recorded in the registers, combining the data with other official sources, especially sources regarding epidemics of the 19th century in Albanian territories. These documents provide evidence not only of how the state classified deaths and responded to mortality crises, but also of how shortages of medical personnel affected the implementation of legislation.
The pronounced male disadvantage in life expectancy in Eastern Europe since the 1970s is a well-established finding in mortality research. However, a longer-term perspective has been largely absent, despite extensive documentation of changes in the sex gap in life expectancy across several Western European countries. The first objective of this paper is to examine the evolution of the sex gap in life expectancy in Estonia from the 19th century onward and compare it to other European nations. The second aim is to analyse the contributions of age- and cause-specific mortality and their interaction using Arriaga decomposition. The results show that since the 19th century, life expectancy at birth for Estonian men has consistently been at least 10% lower than that of women, albeit with some temporal variation. This means that the sex gap in Estonia has remained wider than in Western European countries throughout this time and not merely since the 1970s. The age and cause of death contributions have shifted in line with the health transition. Notable divergences from Western European patterns reflect Estonia’s distinct socio-political trajectory.
Historical Irish cause-of-death data have been digitized to varying standards, and while digital copies can facilitate greater access they are not research ready. Having suffered enormous destruction over the centuries through fire and inadequate storage, Irish records are very fragmented in nature, and under-registration occurs within extant data because of poor state/subject relations and confessional differences. To reconcile absences and provide a fuller impression of historical cause-of-death, we contend that it is important to create life courses from all data types. This paper traces the development of Irish mortality data and discusses our efforts to render machine readable and interoperable data from death registration records and two decennial census datasets using innovative Low-Code/No-Code (LC/NC) solutions. Our methods and approaches will be particularly interesting to countries struggling to create Free, Accessible, Interoperable and Reusable (FAIR) data from legacy datasets that adopted a partial index linked to a PDF or Tiff of original handwritten sources.
This article reassesses mortality in the First World War East Africa Campaign by foregrounding newly recovered cause of death evidence for African soldiers and labourers, many of whom were historically omitted from named commemoration. It situates these findings within the commemorative and administrative frameworks of the Imperial/Commonwealth War Graves Commission and the recent CWGC Non-Commemoration Programme. Drawing on dispersed death lists, theatre casualty returns, and — most significantly — service records and casualty cards of the King's African Rifles (KAR), the study outlines how digitisation, transcription, and the collation of data enable systematic analysis of wartime mortality. The resulting datasets confirm that disease overwhelmingly outweighed combat as a cause of death and identify leading killers including pneumonia, smallpox, dysentery, malaria, and cerebro spinal meningitis, with temporally concentrated peaks linked to operational movement and epidemic dynamics (including late 1918 influenza). Beyond correcting the historical record through improved commemoration, the article argues that these sources offer rare demographic and public health insights for colonial East Africa, opening new avenues for military, epidemiological, and social historical research.
The distribution of healthcare infrastructure has historically influenced health inequalities. Consequently, geographic accessibility — the alignment between facility locations and population distribution — is crucial to mitigating these disparities. In Italy, territorial inequalities have persisted since unification (1861), and have been exacerbated by delayed public health investments. While regional disparities in healthcare infrastructure have often been linked to mortality differences, accessibility remains a neglected issue of study due to data limitations. This study aims to address this gap by using data from the Inchiesta delle Condizioni Igienico Sanitarie del Regno d'Italia (1884–1885) to study healthcare access. The study uses a two-dimensional framework, encompassing availability, defined as the number of facilities, and spatial accessibility, which is measured by travel distance and geographic barriers. Adopting an ego-network approach incorporating altitude and distance provides an updated and more sophisticated study of inequalities in access to primary and institutionalized healthcare services, identifying potential origins of persistent inequalities.
This article describes the creation of a new dataset on historical morbidity. The data derive from 26,500 United Kingdom Post Office pension records covering the period 1860–1908. Each of these forms contained information on the health of a postal worker applying for retirement. This information took the form of a cause of retirement, often medical in nature, and a table providing the number of sick days taken in each of the ten years prior to their retirement. The article describes the source in detail before covering the means of digitisation from imaging through transcription, checking and coding. It details some of the recent publications using this dataset before concluding with some suggestions for future research and comparative work.
Aragonese Socioeconomic Inequality at Death Database (ASIDD) is an ongoing project. Once finalized (expected in early 2029), it will contain mortality data for approximately 65% of all individuals who died in the Aragón region (Spain) between 1874 and 1994, primarily sourced from burial records. The project consists of two subprojects: one covering the period 1874–1939 and the other covering the period 1940–1994 (which is expected to be completed by the end of 2026). Aragón comprises 731 municipalities — ranging from small settlements to the city of Zaragoza (ca. 700,000 inhabitants) — with diverse geography, climate, and levels of development, spread over nearly 50,000 square kilometers. The database will include anonymized information on nearly one million individuals, including date and place of death (and often origin), age, sex, marital status, and presence of descendants at death. For about 75%, socioeconomic status is available, approximated through occupation and literacy, or that of close relatives. Additionally, for a representative sample of 30 urban, semi-urban, rural, and almost-depopulated localities, cause of death is known for nearly all individuals, based on civil registers and parish records. For each locality, information on access to health and emergency services is also available. ASIDD aims to enable the study of social inequality in mortality during the modern era, providing valuable insights into the determinants of health disparities.
The administrative regulations introduced by the partitioning powers were enforced in the Polish lands under Prussian, Russian, and Austrian rule. These regulations included the methods of vital statistics registration, the language in which entries were recorded, and the types of data collected. Differences in registration pose significant challenges for contemporary researchers attempting to unify these records. Additionally, questions arise regarding the credibility and reliability of the registrations, which varied across the territories of the three partitions. Prussia had by far the most advanced statistical system. Strict adherence to deadlines for reporting marriages, births, and deaths, by both officials and ordinary citizens, undoubtedly contributed to the high quality of records maintained in Prussian-controlled territories. This paper aims to characterize the sources relating to causes of death in Poznań, then the capital of the Poznań Province in the 19th century, and to examine the diversity and validity of these sources, the quality and reliability of medical statistics, the development of the Poznań Historical Population Database, and, finally, the research opportunities this database enables. The study highlights the importance of a critical methodological approach to historical cause-of-death data. It demonstrates that, although imperfect, such sources can provide valuable insights into past health patterns and administrative practices when interpreted with linguistic, medical, and contextual sensitivity. The resulting database offers a foundation for further interdisciplinary research on historical epidemiology and urban demography.
This article examines parish registers as the primary source of mortality registration in 19th-century Ukraine and evaluates their potential and limitations for historical-demographic database construction. Drawing on Orthodox and Protestant parish registers from Left-bank and Southern Ukrainian territories of the Russian Empire, the study demonstrates that until the late 1830s these records rarely contained systematic information on causes of death and were characterized by substantial inconsistencies in age reporting, gender registration, and coverage of neonatal mortality. Based on an empirical analysis of more than 5,600 individual death records incorporated into the Ukrainian Mortality Database 19 (UMB19), the article identifies a clear institutional turning point around 1838, after which the recording of causes of death became more regular and increasingly standardized, though still dependent on the diligence and practices of individual clergy. Quantitative indicators such as age heaping and distorted sex ratios reveal persistent problems of underregistration, particularly of women and infants, while also allowing the identification of parish registers that meet minimum quality thresholds for inclusion in historical databases. The article further shows that 19th-century mortality statistics produced by ecclesiastical and civil authorities were entirely derivative of parish registers and therefore reproduced their structural biases. By outlining concrete criteria for source selection and data quality assessment, and by documenting the practical challenges of coding and standardizing historical causes of death, this study contributes to the methodological integration of Ukrainian materials into comparative European research on mortality and population history.
This paper presents Hungarian sources that provide information on the causes of death at both individual and aggregate levels. It demonstrates how causes of death appeared in parish records from the early 19th century onwards, alongside denominational differences and legal prescriptions for recording them. The paper also outlines the process of increasing professionalism, from the mandatory recording of causes, and the involvement of coroners or physicians as death inspectors, to the introduction of civil registration, and the use of prescribed lists of illnesses and international classifications in the first half of the 20th century. The paper also discusses the completeness and reliability of parish records and civil registration, as well as the emergence of aggregate-level statistics, which influenced the content and accuracy of individual-level registration. It presents existing family reconstitution databases containing causes of death alongside digitised statistical publications, enabling a thorough analysis of the mortality transition.
This paper introduces a new source for the study of mortality and the health transition in Italy: the Burial Permits. In the years leading up to the Italian Unification, local authorities began requiring official documentation, compiled by medical officers, for the burial of individuals in local cemeteries. These documents, preserved in the form of single sheets or registers, contain a wealth of individual-level data on deceased people, including the indication of the cause(s) of death. This feature, a novelty in the Italian historical demographic research, allows addressing a longstanding gap in the availability of individual-level information on causes of death, a factor that has limited and hampered the research on the evolution of the mortality patterns and the health transition in Italy. The paper provides a detailed description of this source and the type of information it contains, reviews what has been done so far, and investigates its possible applications to address new directions in the study of health and mortality in Italy.
The Norwegian Historical Population Register (NHPR) reconstructs the entire Norwegian population from 1801 to the present and links it with the modern National Population Register (NPR). It provides a multigenerational research infrastructure spanning seven generations and functions as an authoritative identifier system for documenting individuals and connecting diverse archival materials. This article presents the structure and development of the NHPR's two components: the open online register (NHPR-O), which documents deceased individuals using publicly accessible sources, and the closed register (NHPR-C), which integrates historical and modern microdata under privacy-preserving conditions. We describe the core sources and the hybrid linkage system combing large-scale algorithmic matching with extensive crowdsourcing. The paper reports current linkage rates across censuses, discusses the completeness and quality of historical sources, and examines challenges such as duplicate registrations, and the difficulty of identifying individuals with sparse information. We outline how continuous crowdsourcing and iterative quality control steadily improve linkage accuracy. Finally, the article discusses representativity, the constraints of privacy legislation, and the potential for international cooperation and distributed population registers. Together, these developments establish the NHPR as a scalable and evolving resource for research, genealogy, and public use.
In this article, we present the Madrid database on causes of death, which currently covers the entire city for the period 1905–1927 and includes 366,542 individual notices. Such a resource is unique in the Mediterranean context. The genesis of our data lies in a political and intellectual context dominated by a complex mixture of fears, class contempt, and sincere concern for the most vulnerable, especially children, which explain the development of social and health data and the widespread use of quantitative methods in early 20th-century Spain, particularly in its capital. This resulted in the sources which, over the last 20 years, have been patiently compiled to construct the database of causes of death. The original data were enriched by linking individual data and coding operations, particularly for causes of death, for which two different grids were used. This article questions the quality and reliability of causes of death, that were extraordinarily diverse (n = 1,444). We also summarize the work already accomplished and outline some avenues for future research.
This article reassesses the presence and agency of Portuguese women in Tenerife during the Iberian Union (1580–1640) through a prosopographic reading of notarial, inquisitorial and ecclesiastical documentation. Drawing on life‑course approaches, the study examines the legal and documentary moments through which Portuguese women became visible, situating their agency within the transitions that structured early modern family and mobility trajectories. Rather than attempting to reconstruct the Portuguese population as a whole, the study focuses on a small but analytically rich set of women whose actions — recorded in powers of attorney, wills, debt claims and trans‑archipelagic property transactions — make it possible to observe gendered strategies of mobility, representation and patrimonial management. By integrating a life‑course perspective with insights from Atlantic history and nesology, the analysis identifies three recurrent patterns: the central role of women in the transmission of property across islands; the heightened legal visibility associated with widowhood and the absence of male proxies; and the participation of certain households, particularly those linked to the Azores, in macro‑Atlantic circuits of craft, labour and migration. Comparisons by origin (Madeira, the Azores, continental Portugal) and civil status further clarify how women adapted a shared repertoire of strategies to different legal and familial contexts. The findings show that Portuguese women were not marginal actors but key architects of archipelagic continuity, transforming the constraints of insular life into forms of resilience that shaped kinship, identity and mobility across the early modern Atlantic.
Empirical research in historical demography is usually time-consuming and labour-intensive. Recent developments in machine learning offer new possibilities for building very large databases with reduced time and costs, though these new methods raise new challenges as well. This article describes the process of constructing the POPP database, a data collection project based on the exploitation of the nominative lists of the Parisian population censuses of 1926, 1931, and 1936. This database provides a host of information for almost 9 million individuals: their name and surname, year and location of birth, nationality, relation to the household head, and occupation. The article discusses the digitisation of archival sources — several hundred thousand handwritten pages — their transformation into a database by computer scientists using machine learning techniques, and the work required on the part of social scientists to correct and adapt the resulting data for statistical purposes. Beyond its methodological contribution, this article also discusses the various ways in which the POPP database will improve our knowledge of the economic, social, and demographic evolution of an important European urban population.
In the last 65 years several major historical databases with reconstructed life courses of large populations have been launched. Around 1990, we could find two important types of databases with longitudinal micro-data. The first type were event databases aimed on family reconstructions and usually based on baptism, marriage and funeral registers or on civil certificates introduced after 1800. The second type were databases with life courses: persons are observed on a more permanent basis using church examination or population registers. After 1990 a third type, databases with census data, really took off. In first instance in the form of samples, in second instance by entering full count samples which makes it possible to link the several censuses into one system, creating semi-longitudinal databases. Another development was the growth of special purpose samples into semi-longitudinal ones by following sampled persons from one source during their life course through linking with all kind of other sources. The development of these databases is indicative of considerable investments that have greatly expanded the possibilities for new research within the fields of history, demography, sociology, as well as other disciplines. In this paper I will compare 84 of these databases on several key figures like included sources, year of foundation, period of observation, area of observation, sample fraction and number of included observations, families and unique persons. An overview of all databases with all key figures is presented in the Appendix.
This paper draws together the results from a set of complementary analyses of the causes of infant mortality in eight European port cities in the late 19th century. The deaths were all coded according to ICD10h, with granular codes and a bespoke categorisation for infant deaths. The paper assesses and improves the categorisation of the causes of infant death by considering age and seasonal patterns. We find that while there were similarities in the levels and trends of infant mortality there were also important differences which are ripe for investigation by further research. Cause of death patterns were dominated by a transfer from vague to more specific terms over time, but the vague terms used tended to differ by location, and in the speed of their disappearance. Seasonality analysis suggested that most commonly used vague terms, including convulsions and weakness, probably reflected a variety of different underlying causes and should not be combined with more distinct and coherent categories such as airborne disease or food and water-borne diseases. Although teething is commonly treated as a proxy for diarrhoea, this does not appear to have been universally the case. We illustrate how comparative exercises such as this can further understanding of particular historic terms and their use in different settings, and can produce improved cause of death categorisations. Even the improved categorisation produced here, however, does not free the researcher from the need to consider changes in medical provision, knowledge and terminology within a sensitive and historically informed interpretation.
Using population reconstructions from linked civil certificates for the province of Zeeland, the Netherlands, for the period 1812–1913, I study the social gradient in maternal mortality. Maternal mortality is defined as deaths in the first 42 days after the birth of a child. Among the women — mother to at least one child and followed between age 20 and 45 — maternal mortality constitutes about one third of the total number of observed deaths. Maternal mortality is higher for upper class women in early 19th century Zeeland than for unskilled laborers. By the early 20th century, maternal mortality had become an uncommon event and social differences in its likelihood negligible. A comparison of the social gradient in maternal mortality to the social gradient in all mortality in the reproductive ages (age 20-45) in this period shows that the reverse social gradient in mortality is limited to maternal mortality — it is not found for all women's deaths in this period of life.
