Embedded Files
NOTE: STILLBIRTH DATA IS SIMULATED
Data
Data
All birth-related data was downloaded from open access Government of Alberta datasets, and examined both in Excel and R. As the dataset contained data from both residents and non-residents, we chose to only examine data from residents, as they are more likely to have been affected by the current rates of infection in the province. As well, there were far fewer non-resident births, so even one or two stillbirths within that group could create a stillbirth rate far higher than for the resident population. As already mentioned, we used these provincial-wide statistics to simulate stillbirth rates for the ten municipalities whose wastewater is currently being monitored.
The SARS-CoV-2 Wastewater data was scraped from the publicly accessible Respiratory Virus Dashboard, formatted in Excel, and analyzed in R. The wastewater data provides two different values for each sampling date, one that tests for the N1 gene and one that tests for the N2 gene in the virus. As most publicly available wastewater trend dashboards average these two values (and as they were very close throughout our data) we opted to do the same.
Stillbirth Time Series
Stillbirth Time Series
Figure 1. Time series analysis showing the the average monthly stillbirths occuring in each year, from 2001-2024, with standard error. This is not the annual sum of stillbirths for each year, but instead the average number of stillbirths occurring each month in that year.
Figure 2. Time series analysis of provincial stillbirth data from 2020-2024, with the total monthly provincial stillbirths with standard error in pink, and the simulated monthly averages for municipal stillbirth data.
Figure 3. Close up of simulated municipal stillbirth data from 2020-2024. Includes simulated data for 2025.
Above are three exploratory graphics that show both the real stillbirth data, at the provincial level (Figures 1 and 2), and the data we have simulated at the municipal level (Figures 2 and 3). Figure 1 shows the data over the entire period we have available, 2001-2024, whereas Figure 2 shows the provincial data and the simulated municipal stillbirth data. Figure 3 is just a closeup of the simulated data, and is therefore a little chaotic. This is the data that we have used in our analytic model.
Spliced Time Series
Spliced Time Series
Figure 4. Spliced time series analysis with Estimated Marginal means. On the right is the City of Calgary, on the left is the other locations overlaid. This connects data collected by the U of C and U of A from 2020-2023, and data collected by AH/APL from 2023 to the present.
In Figure 4, a more complete dataset of COVID-19 wastewater data is shown which has been spliced together by location with Estimated Marginal Means (emmeans) in R. As the initial sampling period (approximately 2020-2023) is not directly comparable with the current sampling period (2023-now), it is necessary to create a model that splices the two datasets together. The limitations of this method and this plot, will be discussed in the restuls section.
Page updated
Report abuse