CORDIS - EU research results

Disability and Disease during the 1918 Influenza Pandemic: Implications for Preparedness Policies

Periodic Reporting for period 1 - DIS2 (Disability and Disease during the 1918 Influenza Pandemic: Implications for Preparedness Policies)

Reporting period: 2019-06-03 to 2021-06-02

People with disabilities are at increased risk during pandemics for biological and social reasons. The complex relationships between interacting health conditions and social factors fit the definition of a syndemic. For example, some disabilities might make individuals more susceptible to severe illness or death from infectious diseases. People with disabilities also typically have less access to education, employment, and health care, which enhances risks of exposure and illness. Despite these concerns, there is a lack of previous research on the outcomes for people with disabilities during past pandemics, including the 1918 influenza pandemic.

The aim of this project was to examine historical, epidemiological and anthropological data to enhance understanding of disability as a risk factor for infectious disease and improve pandemic preparedness, reducing vulnerability of people with disabilities in the future. Specific objectives were:
1. To identify social factors that contribute to inequalities in morbidity (illness) and mortality (death) among people with disabilities during influenza pandemics.
2. To develop a simulation tool for evaluation of interventions to reduce the spread of epidemics among institutionalised populations of people with disabilities, with which to inform public health policy.
3. To establish a foundation for future research on disability and infectious disease within the theoretical framework of syndemics, and enhance my professional capacity to be a leader in this field.
Research activities proceed in three main phases. First, I collected historical and archival data, particularly from records held at the National Archives of Norway in Oslo. I focused on institutions such as psychiatric hospitals and schools for children with disabilities. I looked for material on daily life and activities, physical environments, demographic characteristics, and 1918 flu experiences. These records provided context for quantitative analyses and for the construction of the simulation model described below. I also analyzed published data on the 1918 flu at psychiatric institutions in the Bergen area. In addition to details on the spread of disease through these institutions, the directors also reported the numbers of cases and deaths for both residents and staff. My analyses data showed that although illness was significantly lower for patients (only 24% became ill vs. 53% of the staff), there was significantly higher case fatality among them (10% vs. 1% for the staff).

Second, I worked with colleagues at Umeå University, Sweden, to extract records from the Demographic Data Base for individuals living in nine parishes from 1918-1920. Among other variables, these records noted disability and institutionalisation statuses, as well as causes of death. More than 22,000 records were complete enough for further analysis. Statistical comparisons suggest that mortality from influenza and related causes during the pandemic was significantly higher, relative to the non-disabled population, for people with disabilities who were also institutionalised, people with psychiatric or psychological but not other types of disabilities, and people with only one disability but not those with multiple recorded disabilities. This last finding may be due to small sample sizes and/or different rates of institutionalisation and needs further investigation.

Third, quantitative and qualitative data were used to construct a simulation model of a school for children with disabilities, based on annual reports for the Holmestrand School for the Deaf and other schools in early 20th century Norway. This model consists of a population of students and staff who engage in daily activities in social spaces such as bedrooms, classrooms and a dining hall. Simulations of the model show that, on average, epidemics have different sizes, timing and patterns of spread depending on whether a teacher or student is the first case and on how many students share each bedroom.

Dissemination of project results include a peer-reviewed paper based on the first phase of research and published in the Scandinavian Journal of Disability Research. A paper reporting model results is currently under review. Results from both the first and second phases have been presented at conferences, including the American Anthropological Association and the European Social Science History Conference. A final talk was given online, posted to the project website and shared via social media. I also recorded two podcasts with Oslo Metropolitan University and was interviewed about my project for an article in Horizon magazine discussing its relevance for disabled people during COVID-19.
Previous research has considered, for example, disparities for people with certain types of disabilities during seasonal flu and the 2009 flu pandemic, but in general, these are reported in epidemiological literature alongside other chronic health conditions and risk factors. To my knowledge, this project was the first to focus on disability as a risk factor for morbidity and mortality during historical pandemics and specifically the 1918 flu. Data on morbidity are rare for this pandemic compared to death records, so these analyses were especially important. Additionally, historical research on disability, particularly for larger sample sizes, is often limited to institutionalised populations because of available records. In the context of infectious disease spread, this limitation raises the concern of potentially conflating the influences of disability and living in confined settings. Records from the Demographic Data Base at Umeå enabled the analyses of individuals with recorded disabilities but no indication of institutionalisation.

Further, previous models have focused on day schools within communities, nursing homes and hospital wards. The simulation model I built links several important risk factors: disability, school environments, and institutionalisation. Model results therefore provide new insights into populations with these intersecting risk factors. The effectiveness of different potential interventions can also be tested, in order to develop strategies for responding to outbreaks within such institutions.

To my knowledge, this project is also among the first to look at disability and influenza pandemics using a syndemics perspective. Both pandemics and disability are strongly shaped by interacting biological and social factors at multiple scales. This project, therefore, advances disciplines such as anthropology, epidemiology, disability studies and history. Results also can be used to inform COVID-19 recovery and the development of inclusive and equitable preparedness plans for future pandemics.
Visualization of agent-based model