Listening to science in policy design

View more insights in Social Sciences - Political Science

View more figures from University of Montreal

View the page of Antoine Lemor

Read the Study

Next in Political Science

A Modern American Conservative

Russ Freure | University of Waterloo

To Be a Jewish State

Yaacov Yadgar | University of Oxford

Antoine Lemor discusses the role of information selection and processing in policy design by studying school closure decisions during the early days of the COVID-19 pandemic.

Image courtesy of interviewee. September 11, 2024

Generating Workspace...

Export as PDF

This video discusses the contrasting public health approaches taken by Sweden and Quebec during the COVID-19 pandemic, particularly focusing on their distinct public health advisory systems and decision-making processes. Sweden's public health agency operated autonomously, allowing scientists to lead and influence policy decisions, resulting in the choice to keep schools open during the pandemic based on evidence suggesting lower risks for children. In contrast, Quebec's public health agency was politically centralized and subordinate to the government's authority, leading to more conservative measures such as closing all educational institutions due to perceived risks.

The comparison highlights how differences in the organization of public health systems influenced responses to uncertainty and the framing of the pandemic as a health threat. In Sweden, uncertainty was interpreted in a way that prioritized children's development and minimized unnecessary measures, while in Quebec, uncertainty was viewed as a broader threat, justifying more stringent measures. The discussion emphasizes that political decision-makers, when equipped with less evidence, tended to enforce stricter policies, whereas scientific perspectives cautiously balanced health risks with societal impacts.

Moreover, the text explores how public health framing affected policy decisions, with Quebec’s framing emphasizing imminent threats leading to swift actions, compared to Sweden's moderate approach that considered various societal factors. Ultimately, the effectiveness of pandemic responses in Sweden and Quebec can be traced back to the power dynamics between political authorities and public health scientists in their respective systems.

Language:

Export as PDF

Export as PDF
Here are five key points from the video discussing the comparison between public health decision-making in Sweden and Quebec during the COVID-19 pandemic:

1. **Institutional Differences in Public Health Advisory Systems**: The video emphasizes that Sweden and Quebec have fundamentally different public health advisory systems. Sweden's public health agency operates with significant autonomy and influences governmental decisions, especially during the pandemic. In contrast, Quebec's public health agency remains subordinate to the provincial government, particularly the premier, leading to a more centralized decision-making process.

2. **Impact of Decision-Making Processes on Policy Outcomes**: The contrasting institutional structures in Sweden and Quebec resulted in different policy outcomes during the pandemic. For example, Quebec chose to close all primary schools at the onset of the pandemic, while Sweden kept primary schools open, recommending remote learning only for secondary schools. This divergence highlights how the organization of public health systems affects the selection and application of evidence in policy-making.

3. **Handling of Uncertainty**: The video discusses how uncertainty regarding COVID-19 influenced decision-making. Swedish public health officials viewed uncertainty as a risk to children’s development if they implemented unnecessary measures, such as closing schools. In contrast, Quebec authorities perceived uncertainty surrounding potential transmission risks as a threat to the broader population, leading to more precautionary measures, such as school closures.

4. **Framing of the Pandemic**: The framing of the pandemic significantly affected public health responses. Quebec officials framed the pandemic as an imminent and severe threat, justifying stringent and immediate actions like lockdowns. Conversely, Sweden adopted a more moderate framing, balancing health risks with societal impacts, which allowed for more flexible measures regarding school closures.

5. **Influence of Political Convideo on Policy Decisions**: The analysis illustrates how the political convideo and the nature of political leadership impact public health decisions. In Quebec, the political leadership's interpretation of uncertainty led to conservative, risk-averse decisions. In Sweden, the scientific perspective guided the response, where more evidence was required to justify strict measures, reflecting a reliance on scientific guidance rather than political pressure alone.

Export as PDF
Here are some key quotes from the video, highlighting the main ideas and themes discussed:

1. **Comparison of Regions**:
- "It makes them comparable... [because they have] similar population and trust towards governments."

2. **Public Health Advisory Systems**:
- "Quebec and Sweden have, uh, fundamentally different public health advisory systems."
- "Only Sweden's public health agency operates with, uh, true autonomy."

3. **Decision-Making Process**:
- "This comparison allows us to observe the outcomes when public health scientists lead the decision making process as in Sweden versus when political leaders in Quebec are in charge."

4. **School Closure Decisions**:
- "Quebec chose to close down all primary schools... Sweden chose to maintain the opening of schools while recommending remote learning for secondary schools."

5. **Handling Uncertainty**:
- "The way political decision makers and scientists handle and interpret uncertainty can actually differ significantly and then have an effect on policy."
- "Scientists that tend to be more accustomed to uncertainty... but political decision makers... can perceive uncertainty as a threat for the public."

6. **Role of Evidence in Decision-Making**:
- "In Sweden, the public health agency... considered that the lack of solid knowledge... could not lead to the implementation of measures that could actually harm children's development, such as school closures."
- "The dangerous frame created a sense of urgency and justified the implementation of precautionary measures... even though the level of evidence can be low."

7. **Framing the Pandemic**:
- "Framing can then highlight certain aspects of a problem while downplaying others... and thereby shape the policies that are considered then appropriate."

8. **Impact of Political Control**:
- "When political decision makers are central in advisory systems, they tend to rely less on evidence."
- "When scientists are more essential, more evidence is needed to implement stringent policies."

These quotes encapsulate the significant themes of the video, focusing on the differences in public health strategies between Sweden and Quebec, the influence of political autonomy on decision-making, and the role of evidence and uncertainty in shaping health policies during the pandemic.

X-TIMESTAMP-MAP:LOCAL=00:00:00.000,MPEGTS=900000

00:00.000 --> 00:02.885
Important to, interesting to compare Sweden and,

00:03.225 --> 00:04.245
and, and Quebec.

00:04.245 --> 00:05.485
Well, it's two regions

00:05.485 --> 00:07.205
with distinct institutional structures

00:07.205 --> 00:10.685
and decision making processes, but similar population and,

00:10.985 --> 00:12.325
and trust towards governments.

00:12.325 --> 00:14.725
So, so it makes it, it makes them comparable.

00:15.025 --> 00:17.085
But more precisely, why is it interesting

00:17.085 --> 00:18.565
to compare these two regions?

00:19.195 --> 00:20.925
Well, Quebec and Sweden have,

00:20.925 --> 00:23.805
fundamentally different public health advisory systems.

00:24.195 --> 00:27.845
Basically, avise systems are structures through,

00:27.845 --> 00:31.925
through which governments, organize the use

00:31.925 --> 00:33.125
and the productions of evidence.

00:33.945 --> 00:37.445
So in public health, what you can do when you create a a,

00:37.445 --> 00:39.845
when, well, in your organiz of your address system, is

00:39.845 --> 00:41.845
to create a public health agencies

00:41.845 --> 00:44.725
that would produce the evidence in, in, in, in public case,

00:44.825 --> 00:47.485
and give it, give to the agency powers

00:47.485 --> 00:49.685
regarding health policies to design them.

00:49.715 --> 00:52.605
Only Sweden's public health agency operates with,

00:52.605 --> 00:55.765
true autonomy, meaning it get designed public, policies

00:56.385 --> 00:59.165
and through public re recommendations they did during,

00:59.165 --> 01:00.925
during the pandemic,

01:01.105 --> 01:02.765
and influence governmental decisions.

01:02.765 --> 01:06.345
But in Quebec, the public health agency, the NS Pku

01:06.345 --> 01:09.285
and the NS were during the pandemic

01:09.305 --> 01:11.805
and are still subordinate to the government,

01:12.065 --> 01:16.125
and particularly the premier, who was the main, well,

01:16.125 --> 01:18.525
the primary decision maker during the, during the pandemic.

01:18.945 --> 01:21.725
So one which is autonomous in Sweden, in in Sweden,

01:21.865 --> 01:23.045
in, in Sweden, sorry.

01:23.385 --> 01:26.165
And the other one who is government controlling Quebec.

01:26.585 --> 01:28.485
And this comparison actually allows us

01:28.485 --> 01:31.805
to observe the outcomes when public health scientists,

01:31.835 --> 01:33.125
lead the decision making process

01:33.505 --> 01:37.205
as in Sweden versus when political leaders in Quebec,

01:37.475 --> 01:40.485
lead, lead, lead, lead well are in charge.

01:40.705 --> 01:44.245
So in this context, one particular decision, the, the, the,

01:44.305 --> 01:46.205
the decision to close down school at the beginning

01:46.205 --> 01:49.565
of the pandemic can serve as a very interesting case,

01:50.345 --> 01:52.845
of, of the, of the effect of this differences in,

01:52.845 --> 01:55.325
in the organization of, of public health advisor systems.

01:55.895 --> 01:58.525
While Quebec chose to close down all primary schools

01:58.625 --> 02:01.965
and well all actually educational school, institutions

02:01.965 --> 02:04.965
during the pandemic, at the beginning of the pandemic,

02:05.145 --> 02:09.405
Sweden chose to close down to, to maintain,

02:09.405 --> 02:12.445
the opening of, of, of schools,

02:12.815 --> 02:15.565
while recommending remote learning for secondary schools

02:15.665 --> 02:17.245
and also, universities.

02:17.905 --> 02:20.845
Uh, and these differences can actually be linked

02:20.985 --> 02:23.445
to the way the public health visionary systems are,

02:23.465 --> 02:26.125
are organized in each country, which, as we said,

02:26.125 --> 02:29.285
influence the way the the evidence was selected, used,

02:29.425 --> 02:31.525
and even their, their uncertainty.

02:31.665 --> 02:34.205
We all remember during the, the pandemic policy makers that,

02:34.205 --> 02:35.445
that were stating,

02:35.555 --> 02:37.645
that they were following the science, right.

02:38.225 --> 02:40.885
Uh, but in reality, especially at the beginning

02:40.885 --> 02:43.325
of the pandemic in March, 2020 when they were implementing

02:43.565 --> 02:45.725
lockdowns and, and, and very stringent measures

02:45.725 --> 02:47.845
that we all know now, we actually

02:48.045 --> 02:49.445
didn't know much about the virus.

02:49.905 --> 02:51.125
We, didn't know.

02:51.235 --> 02:52.965
Well, it wasn't recognized at that time,

02:52.965 --> 02:54.405
even though there were studies about it,

02:54.425 --> 02:58.685
but the, about the prey, pre presymptomatic transmission or,

02:59.345 --> 03:01.845
Um, or even the, the, the characteristic, the,

03:01.845 --> 03:04.285
the exact characteristic of how the virus was spreading.

03:05.145 --> 03:08.645
So in some ways, policymakers were, were

03:08.705 --> 03:10.285
as much following the uncertainty

03:10.345 --> 03:11.845
as they were following the actual science.

03:12.505 --> 03:14.965
Uh, now what's interesting in that is that the way

03:15.075 --> 03:19.405
that political decision maker, and scientists handle

03:19.585 --> 03:23.005
and interpret uncertainty can actually differ significantly

03:23.185 --> 03:25.925
and then have an effect on policy in that, in that context.

03:26.665 --> 03:29.125
So scientists that tend to be more accustomed

03:29.185 --> 03:30.285
to uncertainty,

03:30.285 --> 03:32.085
because they actually work with it, their job is

03:32.085 --> 03:35.325
to actually reduce the uncertainty through more research.

03:35.945 --> 03:38.045
But political decision makers, on the other hand,

03:38.045 --> 03:41.445
they can perceive uncertainty as a threat for the public

03:41.475 --> 03:44.365
that they are, responsible for,

03:44.465 --> 03:48.245
and especially when it comes to, their health,

03:48.245 --> 03:50.845
leading them to approach it with more, coaching than,

03:50.845 --> 03:52.605
than, than than scientists.

03:53.065 --> 03:55.125
And this differences in the relationship

03:55.125 --> 03:57.405
with uncertainty then becomes very crucial

03:57.405 --> 03:58.405
during your health crisis.

03:59.065 --> 04:00.685
And when actually poli,

04:00.905 --> 04:03.325
public health visionary system grant more influence

04:03.345 --> 04:07.045
to scientists and their interpretation of uncertainty than

04:07.105 --> 04:08.445
to, politician

04:08.905 --> 04:12.005
who have different interpretation of uncertainty.

04:12.625 --> 04:13.845
So regarding school,

04:13.845 --> 04:15.645
school closure is a very good example of this.

04:15.705 --> 04:17.645
So in, in Sweden, where, as we said,

04:17.645 --> 04:18.885
the public health agency operated

04:18.995 --> 04:23.045
with significant o autonomy, well, the agency considered

04:23.045 --> 04:27.125
that the lack of solid knowledge, regarding,

04:27.715 --> 04:30.085
regarding the ity of the virus for,

04:30.145 --> 04:34.525
for children could not lead to the implementation of, of,

04:34.705 --> 04:36.285
of, of measures

04:36.285 --> 04:39.325
that could actually harm children's development, such as,

04:39.545 --> 04:43.005
so school closures, in other words, uncertainty was not seen

04:43.025 --> 04:44.485
as a health risk in itself,

04:44.905 --> 04:47.685
but as a risk if it led to a severe,

04:47.745 --> 04:50.365
but unnecessary measures for, for the children.

04:51.425 --> 04:55.395
Um, actually when, when they, when they, when they,

04:55.785 --> 04:59.715
when they recommended, to keep, schools open,

05:00.145 --> 05:03.875
they justified their, measure with a report,

05:03.875 --> 05:06.235
narrowly report of the WHO that was published at the end

05:06.235 --> 05:09.835
of February 20th, 2020, which was based on,

05:09.895 --> 05:13.315
on data from from China, which, showed that,

05:13.895 --> 05:15.555
the children were at lower risk

05:15.735 --> 05:18.315
of developing severe outcomes from COVID-19 than,

05:18.495 --> 05:19.715
for example, elderly people.

05:20.135 --> 05:23.115
And that they were also compared to other,

05:23.365 --> 05:28.155
other pandemics such as influenza, if, if influenza,

05:28.855 --> 05:31.635
not great, not big, vectors of transmission.

05:32.925 --> 05:35.625
So Swedish scientists, what they did is that they used this,

05:35.655 --> 05:38.625
this evidence to justify more targeted approach, a kind

05:38.625 --> 05:40.745
of unique approach compared to its neighbors and,

05:40.925 --> 05:43.945
and to Quebec, where they kept primary schools opens

05:43.945 --> 05:45.625
because they were not a, a high risk

05:46.155 --> 05:48.305
while recommending remote learning for,

05:48.445 --> 05:49.545
for a secondary school.

05:50.365 --> 05:54.145
Um, because actually, the report was showing that,

05:54.285 --> 05:58.605
individual plus 18 were, bigger vectors

05:58.605 --> 06:00.085
of transmission than the, than the children.

06:00.745 --> 06:03.605
And the strategy was part of what was coin as by literature

06:03.605 --> 06:06.285
as personalized, personalized lockdown approach

06:06.295 --> 06:07.965
where recommendations were tailored

06:07.985 --> 06:11.285
to specific groups based on their level of risk,

06:11.725 --> 06:12.925
evidence based level of risk,

06:13.785 --> 06:16.165
partly focusing on elderly people who were advised

06:16.165 --> 06:17.685
to stay at home at, at the beginning

06:17.685 --> 06:21.525
of the pandemic in which, was, criticized then,

06:21.625 --> 06:22.645
throughout the pandemic.

06:23.145 --> 06:27.085
But on the other hand, in Quebec, where,

06:27.145 --> 06:29.565
the public health adversary system was more politically

06:29.565 --> 06:34.205
centralized, it led actually led to a different outcome.

06:35.025 --> 06:37.645
The public health advisory system in Quebec, as we said, so,

06:37.665 --> 06:38.965
was more politically driven

06:38.995 --> 06:42.445
with a premier holding full control over the policies

06:42.445 --> 06:43.805
that were implemented.

06:44.425 --> 06:47.005
And the centralized, I, centralized, sorry,

06:47.285 --> 06:49.965
structures actually con contributed to a different,

06:50.255 --> 06:53.565
conception of, of the risk, a unified conception of risk

06:53.855 --> 06:57.045
where uncertainty this time was viewed as a danger

06:57.225 --> 06:59.485
to the en entire population, rather than

06:59.785 --> 07:02.525
as a danger if it led to unnecessary measures

07:03.025 --> 07:04.165
during press conferences.

07:04.225 --> 07:06.365
For instance, the premier and the head of the public,

07:06.425 --> 07:07.965
he health administration here, Dr.

07:08.265 --> 07:12.405
aa, who is a public official directly, subordinate,

07:12.685 --> 07:14.245
subordinated to the government.

07:14.245 --> 07:15.965
And the health ministry stated

07:15.995 --> 07:18.325
that while children were not at high risk themselves,

07:18.325 --> 07:19.565
they knew that,

07:19.565 --> 07:21.885
the government did not know, sorry.

07:21.885 --> 07:23.285
They could transmit the disease.

07:23.285 --> 07:25.325
They could actually transmit the disease to,

07:25.545 --> 07:27.005
to elderly, people.

07:27.945 --> 07:31.845
Um, this was said, even though at that time,

07:31.865 --> 07:33.485
the government could not know the prevalence

07:33.485 --> 07:35.805
of the virus throughout the population, so that uncertainty

07:35.805 --> 07:37.805
around the possibility of the children

07:38.155 --> 07:40.405
that could transmit the disease, even though they,

07:40.405 --> 07:42.725
they could not know about that transmission at that time,

07:43.185 --> 07:45.925
was seen as a threat to the entire population, leading

07:46.065 --> 07:49.165
to the closure of, of all educational institutions.

07:49.185 --> 07:50.565
So compared to to Sweden,

07:51.725 --> 07:55.495
this decision showed a more ous approach where the fear

07:55.495 --> 07:58.775
of the political consequences of underestimating the risks

07:59.275 --> 08:01.615
for everyone, and especially the elderly people,

08:01.835 --> 08:03.335
was considered too high leader

08:03.355 --> 08:06.535
to more conservative measures than in, in, in Sweden.

08:07.555 --> 08:10.415
But also, what I, what I wanna add is that responding

08:10.415 --> 08:14.175
to uncertainty, it also involve selecting certain pieces

08:14.195 --> 08:15.855
of information to fill the gaps

08:15.925 --> 08:17.695
that the uncertainty creates.

08:17.795 --> 08:20.695
So, in other words, when we, don't know about something,

08:20.695 --> 08:22.775
we tend, to rely on other sources

08:22.795 --> 08:24.775
and of information to inform us about,

08:24.865 --> 08:26.655
about the thing that we wanna know.

08:27.925 --> 08:30.535
Well, in an up upcoming study, that we've contacted

08:30.535 --> 08:32.415
with, with colleagues, we show that in Quebec,

08:32.415 --> 08:34.775
during the pandemic, actually public opinion data,

08:34.915 --> 08:38.415
so political information significantly influenced the way

08:38.445 --> 08:39.695
that the government

08:39.695 --> 08:42.855
and the decision makers used the hospitalization projection

08:42.855 --> 08:44.375
during, during the pandemic.

08:44.445 --> 08:47.735
What we show is that the more the government was supported

08:47.735 --> 08:51.415
by the public, and the more, they tended to follow,

08:51.775 --> 08:54.175
hospitalization, projections, the projects of

08:54.195 --> 08:55.685
of their own experts,

08:56.345 --> 08:59.005
and this is actually one way to manage uncertainty.

08:59.605 --> 09:01.165
'cause since it was impossible to know

09:01.165 --> 09:03.205
with certainty the evolution of the pandemic,

09:03.205 --> 09:05.685
because the projections were still uncertain, we,

09:05.745 --> 09:09.005
we could not know how the people were reacting to measures.

09:10.155 --> 09:12.245
Well, one way to manage that uncertainty was

09:12.245 --> 09:15.285
to rely on other, so sources of information here,

09:15.315 --> 09:17.405
political information, public opinion data,

09:18.105 --> 09:20.845
and the fact that Quebec's Public Health advisory system was

09:20.845 --> 09:21.885
politically centralized,

09:21.885 --> 09:23.685
it actually played a key role in allowing,

09:24.365 --> 09:25.525
allowing such influence.

09:25.625 --> 09:27.765
The beginning of the pandemic. The, the government asked,

09:27.785 --> 09:31.645
asked a firm to produce this, these data,

09:32.305 --> 09:34.925
and we tried, actually tried to conduct a comparative study

09:34.955 --> 09:36.365
with, with Swedish Sweden.

09:36.545 --> 09:39.565
So we asked the public health agency to give us,

09:39.945 --> 09:41.725
the data, public opinion data

09:41.725 --> 09:44.845
that they could have produced, but they just didn't.

09:45.505 --> 09:49.285
Um, so what it shows is that the way that public health,

09:50.105 --> 09:53.045
system are organized, have an influence on

09:53.045 --> 09:55.965
how uncertainty is, is, is interpreted,

09:56.185 --> 09:57.485
how evidence is used,

09:57.705 --> 09:59.685
but also how other information,

09:59.795 --> 10:02.125
what informations are produced to,

10:02.505 --> 10:04.365
as a response to that uncertainty.

10:04.785 --> 10:06.125
And in Sweden, in Sweden,

10:06.125 --> 10:07.685
because it was public health agency,

10:07.685 --> 10:10.765
which was leading the response, well,

10:10.765 --> 10:13.085
the focus was more directly on health related studies.

10:13.855 --> 10:16.625
They actually made studies, but it was opinion studies,

10:16.805 --> 10:19.265
but it was based on the relation, the, the,

10:19.495 --> 10:22.505
with the relationship of, of people with their health,

10:23.125 --> 10:25.545
not the, the support for the government

10:25.565 --> 10:26.585
or the measures themselves.

10:26.635 --> 10:29.465
These differences in handling uncertainty were not just

10:29.475 --> 10:32.425
about cous versus a more evidence driven approach.

10:32.525 --> 10:34.825
It was, it was also deeply rooted in

10:34.825 --> 10:37.305
how the COVID-19 was, was defined and framed.

10:37.305 --> 10:40.625
So in social science framing refers to the way that,

10:40.625 --> 10:43.265
problems are, defined, presented

10:43.265 --> 10:44.305
and inter interpreted

10:44.305 --> 10:46.505
because, problems are multidimensional.

10:46.505 --> 10:48.385
So we tend to focus on one dimension

10:48.385 --> 10:50.825
of the problem rather than on, on another.

10:51.165 --> 10:53.385
But focusing on one dimension of the problem tend

10:53.385 --> 10:55.785
to then influence the way that we interpret the problem.

10:55.885 --> 11:00.425
And so the solution that we, want to use to, to,

11:00.485 --> 11:02.105
put an end or to, to fix the problem.

11:03.225 --> 11:06.645
So framing can then highlight certain aspects of,

11:06.785 --> 11:08.765
of a problem while downplaying others,

11:09.265 --> 11:11.205
and thereby shape the, the policies

11:11.205 --> 11:13.125
that are considered then appropriate.

11:13.225 --> 11:16.845
So, for example, we can frame the pandemic as an imminent

11:16.845 --> 11:19.485
and survey threat to threat, sorry, to human health,

11:20.165 --> 11:22.445
focusing on, on the dangers of the disease in terms

11:22.445 --> 11:27.325
of death, hospitalization,

11:27.825 --> 11:30.525
yeah, death or, or hospitalization.

11:30.625 --> 11:33.405
But the hospitals are just overwhelmed, overwhelmed of,

11:33.405 --> 11:35.085
of overwhelming of hospitals,

11:35.235 --> 11:37.365
what we might call here a dangerous frame.

11:37.365 --> 11:39.125
So a dangerous way to define the, the,

11:39.185 --> 11:40.325
the, the, the pandemic.

11:40.745 --> 11:44.685
But alternatively, you can also define the, the problem,

11:44.875 --> 11:48.445
such as to, moderate the, per that perceived threat to,

11:48.985 --> 11:52.565
human health by also emphasizing other dimensions

11:52.565 --> 11:55.205
of the problems such as the ne ne negative impacts

11:55.205 --> 11:57.445
of drastic measures on mental health, on,

11:57.445 --> 11:58.525
on economic downturns.

11:58.585 --> 12:01.485
And as the, the, the case of, the case of,

12:01.585 --> 12:03.965
of schools show on the development of children,

12:04.435 --> 12:07.485
what we might call so moderate, moderate frame,

12:07.765 --> 12:09.565
moderate way to define the, the problem.

12:10.625 --> 12:13.765
So in an upcoming research with Con, with Eric Monti,

12:14.065 --> 12:16.645
we used a, a data set of all the press conferences

12:16.645 --> 12:19.885
that were held in Quebec and, and,

12:19.905 --> 12:22.765
and Sweden by public health, representative,

12:22.905 --> 12:24.205
but also policy makers.

12:24.585 --> 12:27.045
And by applying up applying natural language,

12:27.325 --> 12:28.965
language processing techniques, we were able

12:28.965 --> 12:32.845
to measure the prevalence in each, press conferences

12:33.065 --> 12:36.045
of the, the, the actual frames that we just talked about.

12:36.225 --> 12:38.045
So the dangerous frame, the moderate frame,

12:38.065 --> 12:40.605
but also the number of evidence that were seated,

12:41.065 --> 12:42.365
in each sentence.

12:42.825 --> 12:45.165
And what we found is that in Quebec,

12:45.425 --> 12:47.285
at least at the beginning of the pandemic of,

12:47.405 --> 12:49.165
'cause of course, it fluctuated over time,

12:49.825 --> 12:53.445
but at the beginning of the pandemic, the problem was,

12:53.465 --> 12:55.685
initially framed as an imminent

12:55.685 --> 12:57.525
and severe threat to, to, to public health.

12:57.525 --> 12:59.045
So the dangerous frame that we talked about.

12:59.705 --> 13:03.685
So they emphasized the risks associated with the virus,

13:03.685 --> 13:06.845
the po potential death, the overwhelmed hospitals and, and,

13:06.845 --> 13:08.325
and the widespread transmission.

13:08.385 --> 13:10.445
But more importantly, what we found is

13:10.445 --> 13:15.045
that this framing was associated with, the enactment of,

13:15.065 --> 13:16.125
of suppression policy.

13:16.305 --> 13:18.405
So lockdowns closure, closure

13:18.465 --> 13:19.925
of schools and, and, and so on.

13:20.665 --> 13:22.775
But especially in times

13:22.915 --> 13:25.215
and even when the level of evidence was very low,

13:25.485 --> 13:28.975
less the evidence, the more the, the stringency

13:28.975 --> 13:30.055
of, of, of measures.

13:30.755 --> 13:33.495
And what, what it, what it shows is that, is

13:33.495 --> 13:36.575
that the dangerous frame, created a sense of, of urgency

13:36.635 --> 13:38.775
and justified the inflammation of,

13:38.775 --> 13:43.215
precautionary measures, but without

13:43.215 --> 13:45.975
however, relying

13:46.115 --> 13:48.895
or using as a justification, evidence.

13:48.995 --> 13:52.375
So it excuses that when political decision makers are at the

13:52.455 --> 13:56.735
helm, a a perceived threat is sufficient to, act,

13:57.125 --> 14:00.455
even though the level of evidence can be, can be low.

14:00.955 --> 14:05.015
But, in contrasting in, in Sweden at the beginning

14:05.015 --> 14:07.895
of the pandemic, as we just saw with the, the, the,

14:08.035 --> 14:09.855
the closure of school, well, they let,

14:09.855 --> 14:10.935
they let schools open,

14:11.035 --> 14:15.175
but the, the case of school, they initially adopted a,

14:15.195 --> 14:16.695
a more moderate frame.

14:16.695 --> 14:18.935
So the, the other definition that we just talked about

14:18.965 --> 14:22.335
that balance the health risk spur by COVID-19 with,

14:22.365 --> 14:25.135
with the societal impacts of, of stringent measures.

14:25.875 --> 14:30.735
But in interestingly, well, Sweden also implemented,

14:30.875 --> 14:34.175
stringent measures, for instance, they implemented,

14:34.175 --> 14:36.455
restrictions in, in, in restaurants.

14:37.115 --> 14:40.615
Um, and what we, what we show in that study is that

14:41.195 --> 14:43.615
the dangerous frame is also influential on

14:43.615 --> 14:44.815
stringent me measures.

14:45.275 --> 14:46.655
But compared to Quebec,

14:47.005 --> 14:49.455
only when the level evidence is high, the use

14:49.455 --> 14:50.855
of these frames have evolved over

14:50.855 --> 14:51.855
Time. Uh, what, what

14:51.855 --> 14:53.725
I detailed more, more

14:54.245 --> 14:55.445
specifically in my thesis, but

14:56.165 --> 14:58.375
what these findings clearly demonstrate is that

14:59.035 --> 15:00.775
the different relationship with evidence,

15:00.835 --> 15:03.415
but also with uncertainty in the two jurisdiction,

15:03.475 --> 15:06.335
and how the organization of the public health,

15:06.535 --> 15:11.175
advisory system, were organized, affected the use

15:11.235 --> 15:13.175
of scientific evidence and, and,

15:13.355 --> 15:15.055
and the interpretation of uncertainty

15:15.235 --> 15:17.655
and the policy that were implemented.

15:17.955 --> 15:20.575
So when political decision maker are central in

15:20.575 --> 15:21.775
that ary systems, they tend

15:21.775 --> 15:24.295
to rely less on evidence if a problem is

15:24.675 --> 15:29.055
or justify less, scientifically their, their policies.

15:29.055 --> 15:32.055
If, if a problem is framed as being a potential threat

15:32.055 --> 15:35.215
to human health, but when scientists are more essential,

15:35.215 --> 15:39.255
such as in Sweden, more evidence is needed to implement,

15:39.255 --> 15:41.415
stringent policies such as school closure,

15:41.525 --> 15:43.415
even if a problem is framed as being a threat

15:43.415 --> 15:47.215
to human health, which, which is, very interesting

15:47.755 --> 15:48.855
in, in that light.

"Comparative Analysis of Public Health Advisory Systems: Sweden vs. Quebec During the Pandemic"

00:00:00 - Important to, interesting to compare Sweden and, and, and Quebec. Well, it's two regions with distinct institutional structures and decision making processes, but similar population and, and trust towards governments. So, so it makes it, it makes them comparable. But more precisely, why is it interesting to compare these two regions? Well, Quebec and Sweden have, fundamentally different public health advisory systems. Basically, avise systems are structures through, through which governments, organize the use and the productions of evidence. So in public health, what you can do when you create a a, when, well, in your organiz of your address system, is to create a public health agencies that would produce the evidence in, in, in, in public case,
00:00:47 - regarding health policies to design them. Only Sweden's public health agency operates with, true autonomy, meaning it get designed public, policies and through public re recommendations they did during, during the pandemic, and influence governmental decisions. But in Quebec, the public health agency, the NS Pku and the NS were during the pandemic and are still subordinate to the government, and particularly the premier, who was the main, well, the primary decision maker during the, during the pandemic. So one which is autonomous in Sweden, in in Sweden, in, in Sweden, sorry. And the other one who is government controlling Quebec.
00:01:28 - to observe the outcomes when public health scientists, lead the decision making process as in Sweden versus when political leaders in Quebec, lead, lead, lead, lead well are in charge. So in this context, one particular decision, the, the, the, the decision to close down school at the beginning of the pandemic can serve as a very interesting case, of, of the, of the effect of this differences in, in the organization of, of public health advisor systems. While Quebec chose to close down all primary schools and well all actually educational school, institutions during the pandemic, at the beginning of the pandemic, Sweden chose to close down to, to maintain,
00:02:12 - while recommending remote learning for secondary schools and also, universities. and these differences can actually be linked to the way the public health visionary systems are, are organized in each country, which, as we said, influence the way the the evidence was selected, used, and even their, their uncertainty. We all remember during the, the pandemic policy makers that, that were stating, that they were following the science, right. but in reality, especially at the beginning of the pandemic in March, 2020 when they were implementing lockdowns and, and, and very stringent measures that we all know now, we actually didn't know much about the virus.
00:02:51 - Well, it wasn't recognized at that time, even though there were studies about it, but the, about the prey, pre presymptomatic transmission or, or even the, the, the characteristic, the, the exact characteristic of how the virus was spreading. So in some ways, policymakers were, were as much following the uncertainty as they were following the actual science. now what's interesting in that is that the way that political decision maker, and scientists handle and interpret uncertainty can actually differ significantly and then have an effect on policy in that, in that context. So scientists that tend to be more accustomed to uncertainty,
00:03:32 - to actually reduce the uncertainty through more research. But political decision makers, on the other hand, they can perceive uncertainty as a threat for the public that they are, responsible for, and especially when it comes to, their health, leading them to approach it with more, coaching than, than, than than scientists. And this differences in the relationship with uncertainty then becomes very crucial during your health crisis. And when actually poli, public health visionary system grant more influence to scientists and their interpretation of uncertainty than to, politician who have different interpretation of uncertainty.

"Contrasting Approaches to School Closures During COVID-19: The Cases of Sweden and Quebec"

00:04:13 - school closure is a very good example of this. So in, in Sweden, where, as we said, the public health agency operated with significant o autonomy, well, the agency considered that the lack of solid knowledge, regarding, regarding the ity of the virus for, for children could not lead to the implementation of, of, of, of measures that could actually harm children's development, such as, so school closures, in other words, uncertainty was not seen as a health risk in itself, but as a risk if it led to a severe, but unnecessary measures for, for the children. actually when, when they, when they, when they,
00:05:00 - they justified their, measure with a report, narrowly report of the WHO that was published at the end of February 20th, 2020, which was based on, on data from from China, which, showed that, the children were at lower risk of developing severe outcomes from COVID-19 than, for example, elderly people. And that they were also compared to other, other pandemics such as influenza, if, if influenza, not great, not big, vectors of transmission. So Swedish scientists, what they did is that they used this, this evidence to justify more targeted approach, a kind
00:05:40 - and to Quebec, where they kept primary schools opens because they were not a, a high risk while recommending remote learning for, for a secondary school. because actually, the report was showing that, individual plus 18 were, bigger vectors of transmission than the, than the children. And the strategy was part of what was coin as by literature as personalized, personalized lockdown approach where recommendations were tailored to specific groups based on their level of risk, evidence based level of risk, partly focusing on elderly people who were advised to stay at home at, at the beginning of the pandemic in which, was, criticized then,
00:06:23 - But on the other hand, in Quebec, where, the public health adversary system was more politically centralized, it led actually led to a different outcome. The public health advisory system in Quebec, as we said, so, was more politically driven with a premier holding full control over the policies that were implemented. And the centralized, I, centralized, sorry, structures actually con contributed to a different, conception of, of the risk, a unified conception of risk where uncertainty this time was viewed as a danger to the en entire population, rather than as a danger if it led to unnecessary measures
00:07:04 - For instance, the premier and the head of the public, he health administration here, Dr. aa, who is a public official directly, subordinate, subordinated to the government. And the health ministry stated that while children were not at high risk themselves, they knew that, the government did not know, sorry. They could transmit the disease. They could actually transmit the disease to, to elderly, people. this was said, even though at that time, the government could not know the prevalence of the virus throughout the population, so that uncertainty around the possibility of the children that could transmit the disease, even though they, they could not know about that transmission at that time,
00:07:46 - to the closure of, of all educational institutions. So compared to to Sweden, this decision showed a more ous approach where the fear of the political consequences of underestimating the risks for everyone, and especially the elderly people, was considered too high leader to more conservative measures than in, in, in Sweden. But also, what I, what I wanna add is that responding to uncertainty, it also involve selecting certain pieces of information to fill the gaps that the uncertainty creates. So, in other words, when we, don't know about something, we tend, to rely on other sources and of information to inform us about,

"Comparative Analysis of Public Health Decision-Making: The Influence of Political Support and Framing During the COVID-19 Pandemic in Quebec and Sweden"

00:08:27 - Well, in an up upcoming study, that we've contacted with, with colleagues, we show that in Quebec, during the pandemic, actually public opinion data, so political information significantly influenced the way that the government and the decision makers used the hospitalization projection during, during the pandemic. What we show is that the more the government was supported by the public, and the more, they tended to follow, hospitalization, projections, the projects of of their own experts, and this is actually one way to manage uncertainty. 'cause since it was impossible to know with certainty the evolution of the pandemic, because the projections were still uncertain, we,
00:09:10 - Well, one way to manage that uncertainty was to rely on other, so sources of information here, political information, public opinion data, and the fact that Quebec's Public Health advisory system was politically centralized, it actually played a key role in allowing, allowing such influence. The beginning of the pandemic. The, the government asked, asked a firm to produce this, these data, and we tried, actually tried to conduct a comparative study with, with Swedish Sweden. So we asked the public health agency to give us, the data, public opinion data that they could have produced, but they just didn't. so what it shows is that the way that public health,
00:09:53 - how uncertainty is, is, is interpreted, how evidence is used, but also how other information, what informations are produced to, as a response to that uncertainty. And in Sweden, in Sweden, because it was public health agency, which was leading the response, well, the focus was more directly on health related studies. They actually made studies, but it was opinion studies, but it was based on the relation, the, the, with the relationship of, of people with their health, not the, the support for the government or the measures themselves. These differences in handling uncertainty were not just about cous versus a more evidence driven approach.
00:10:34 - how the COVID-19 was, was defined and framed. So in social science framing refers to the way that, problems are, defined, presented and inter interpreted because, problems are multidimensional. So we tend to focus on one dimension of the problem rather than on, on another. But focusing on one dimension of the problem tend to then influence the way that we interpret the problem. And so the solution that we, want to use to, to, put an end or to, to fix the problem. So framing can then highlight certain aspects of, of a problem while downplaying others, and thereby shape the, the policies that are considered then appropriate.
00:11:16 - and survey threat to threat, sorry, to human health, focusing on, on the dangers of the disease in terms of death, hospitalization, yeah, death or, or hospitalization. But the hospitals are just overwhelmed, overwhelmed of, of overwhelming of hospitals, what we might call here a dangerous frame. So a dangerous way to define the, the, the, the, the pandemic. But alternatively, you can also define the, the problem, such as to, moderate the, per that perceived threat to, human health by also emphasizing other dimensions of the problems such as the ne ne negative impacts
00:11:57 - on economic downturns. And as the, the, the case of, the case of, of schools show on the development of children, what we might call so moderate, moderate frame, moderate way to define the, the problem. So in an upcoming research with Con, with Eric Monti, we used a, a data set of all the press conferences that were held in Quebec and, and, and Sweden by public health, representative, but also policy makers. And by applying up applying natural language, language processing techniques, we were able to measure the prevalence in each, press conferences of the, the, the actual frames that we just talked about.

"Framing Public Health Responses: The Impact of Perceived Threats and Evidence on Decision-Making in Quebec and Sweden During the Pandemic"

00:12:38 - but also the number of evidence that were seated, in each sentence. And what we found is that in Quebec, at least at the beginning of the pandemic of, 'cause of course, it fluctuated over time, but at the beginning of the pandemic, the problem was, initially framed as an imminent and severe threat to, to, to public health. So the dangerous frame that we talked about. So they emphasized the risks associated with the virus, the po potential death, the overwhelmed hospitals and, and, and the widespread transmission. But more importantly, what we found is that this framing was associated with, the enactment of, of suppression policy.
00:13:18 - of schools and, and, and so on. But especially in times and even when the level of evidence was very low, less the evidence, the more the, the stringency of, of, of measures. And what, what it, what it shows is that, is that the dangerous frame, created a sense of, of urgency and justified the inflammation of, precautionary measures, but without however, relying or using as a justification, evidence. So it excuses that when political decision makers are at the helm, a a perceived threat is sufficient to, act,
00:14:00 - But, in contrasting in, in Sweden at the beginning of the pandemic, as we just saw with the, the, the, the closure of school, well, they let, they let schools open, but the, the case of school, they initially adopted a, a more moderate frame. So the, the other definition that we just talked about that balance the health risk spur by COVID-19 with, with the societal impacts of, of stringent measures. But in interestingly, well, Sweden also implemented, stringent measures, for instance, they implemented, restrictions in, in, in restaurants. and what we, what we show in that study is that
00:14:43 - stringent me measures. But compared to Quebec, only when the level evidence is high, the use of these frames have evolved over Time. what, what I detailed more, more specifically in my thesis, but what these findings clearly demonstrate is that the different relationship with evidence, but also with uncertainty in the two jurisdiction, and how the organization of the public health, advisory system, were organized, affected the use of scientific evidence and, and, and the interpretation of uncertainty and the policy that were implemented. So when political decision maker are central in that ary systems, they tend
00:15:47 - or justify less, scientifically their, their policies. If, if a problem is framed as being a potential threat to human health, but when scientists are more essential, such as in Sweden, more evidence is needed to implement, stringent policies such as school closure, even if a problem is framed as being a threat to human health, which, which is, very interesting in, in that light.

Faculti AI can make mistakes

Research

Policy

Analysis