2 -> The Humanity Projects Team
Review Date: February - 2023
1 - Overall rates of reports by registrants
The overall rates of severity of events experienced by participants is shown in the figure below. The data is comprised of all mobile app registants in the V-Safe database from Decemeber 2020 to the end of July 2022.
We evaluated all participants, numbering about 10 million, and categorised as:
#1 - Visit to outpatient clinic, urgent care clinic, emergency room, or hospitalisation. if not, then:
#2 - Missed work or school. If not then:
#3 - Had no reported effects. If not then:
#4 - All other outcomes.
The most severe outcome was the first category, where participants sought medical care, followed by missed work or shcool. Participants seeking medical help may have also missed work or school. All other outcomes (4) include all participants who did not seek medical help and did not miss work or school, but reported symptoms or were limited in their daily abilities. The mildest outcome was #3, where no adverse symptoms were found.
Approximately 5% of participants needed urgent medical care or were hospitalised. Additionally, 30% of participants were not able to work or attend school. Only about 22% of registrants reported no symptoms.
2 - Range of symptoms for different health impacts
The granularity of the overall statistics is improved by analysing the distribution of symptoms for those participants who fell in category 1 (went to a medical facility or were hospitalised) or 2 (missed work or school). The symptoms were ordered in 4 sub-categories: no effect, mild, moderate and severe.
The classification of severe was attributed to a participant that reported any symptom as "severe". If the max symptom was moderate, the symptoms would fall into the moderate category, and the same for the "mild" classification. "No effect" does not preclude registrants choosing not to report a symptom.
Cohort #1 - Emergency room or hospitalisation (5.0%)
Cohort #2 - Lost work or school (29.5%)
This section shows the number of mobile app registrations over time.
The plot below, shows the time series of the number of mobile app registrations and the number of reported events of "unable to work". The majority of registrations occurred from January to May of 2021. The number of new registrations for the app declined significantly from June 2021 and went almost to zero in 2022. Consequently, the data obtained from the start of 2022 onwards is unreliable, as the app fell into disuse among participants.
This section investigates how the rate of the two most severe outcomes evolved over time (following either dose 1 or dose 2 of the vaccine). The 30-day moving average is calculated by first totaling categories 1 and 2 of the number of health reports and then dividing by the number of inoculations during the 30-day window. The results allow us to visualise any temporal patterns in the rate of adverse reactions associated with the covid-19 vaccinations.
The charts below refer to reports made by participants after receiving either doses of the vaccine.
1 - Hospitalisation events
The plot below shows the evolution of the 30-day moving average rate of participants that required medical care (emergency room visit or hospitalisation). The different time series refer to different lengths of the health impact, ranging from <=1 (1 day of lost work) that likely corresponds to patients that sought emergency care but were quickly discharged, to 9+ days of lost work that are likely to be the most severe cases. The time series is obtained by combining the “hospitalised yes/no” variable with “lost work/school for x number of days” variable.
It should be noted that the denominator for computing the rate of participants that required medical care is the number of inoculations during the 30-day window. As the is a delay between the inoculation and the health event, the computed rate is an approximation. Furthermore, some health events, namely those with more serious impact, might take a long time to manifest, which could be up to a year or more, as we'll show in Part 3.
The most frequent medical care was associated with two to four days of lost work (black curve), followed by the cases with a health impact of five to eight days of lost work. The registrants that had a health impact of one day or less were not highly represented indicating that most hospitalisations likely referred to moderate to severe cases.
2 - Missed work (health impact)
The plot shows the evolution of the 30-day moving average rate of inoculations that led to the resgistrant missing work or school. This implies that a mild to moderate symptomatic clinical picture was present.
Most cases of lost work corresponded to one day or less or those who missed two to four days. Most of these cases were likely mild to moderate, as they excluded hospitalisations. Cases of lost work for five or more days were much more rare, and were observed at comparable rates to hospitalisations with five or more days of lost work. Therefore, these cases are likely to include some moderate to severe cases that did not seek medical care but suffered a severe reaction.
Histogram of health visits
Different individuals went through several health visits (emergency room or hospitalisation). We investigate the distribution of the number of health visits for category 1.
The chart below refers to reports of participants after receiving any vaccine dose (at any point). It shows that the most common situation is one health visit, amounting to 53% of the participants. The 92% of participants who sought emergency care had three health visits or less.
Histogram of health impact length
Health impacts occurred for category 2 (Missed work/school). The health impact length refers to the number of days for which the event lasted. Investgating the distribution of the length of health impacts gives us an idea of the distribution of the severity of these events.
The chart below refers to reports of participants after receiving any vaccine dose.
It shows that most common situation is a health impact of two days amounting to 33.2% of the participants. For 90% of participants who had a health impact, it lasted five days or less.