Real-time Research Response

Loading...

來自 The University of Tokyo 的課程

Global Health Policy

75 個評分

The University of Tokyo

Global Health Policy

75 個評分

In this course, learners will become familiar with principles and theories of global health problems, and major challenges and controversies in improving global population health as well as practical applications of quantitative methods to analyze and interpret issues and challenges for policy. Topics will include health and foreign policy, health governance, acute disease surveillance, non-communicable diseases, burden of disease, universal health coverage, health systems strengthening, health financing, and human resources for health and ageing.

從本節課中

Pandemics and Health Security Responses

In this module, we will explore the relationship between universal health coverage, disasters and pandemics. We will learn how to measure and understand the spread of infectious diseases, and how to respond to them. We will also consider the conditions necessary to ensure an effective and timely response to both infectious disease emergencies and disasters, and identify the common role that universal health coverage plays in preparing societies for these unexpected threats to human well-being.

Lessons from Pandemic H1N1-200910:34

Real-time Research Response9:02

與講師見面

Hiroshi Nishiura
Professor, Graduate School of Medicine, Hokkaido University
(Former Affiliation) Associate Professor, Graduate School of Medicine, The University of Tokyo
Stuart Gilmour
Professor, Division of Biostatistics and Bioinformatics, Graduate School of Public Health, St. Luke's International University
(Former Affiliation) Associate Professor, Graduate School of Medicine, The University of Tokyo
Kenji Shibuya
Professor and Chair
Department of Global Health Policy, Graduate School of Medicine

So, first subject is a real time research response.

So, in addition to active surveyance, as an expert, we are expected to

offer some response to the auto reply in the form of the research.

And we would like to focus on the severity measurement or

the case fatality risk as an example.

Okay, in advance, there was a 2009 pandemic, there was an index that

is referred to as Pandemic Severity Index, introduced by the Center for

Disease Control and Prevention in the United States.

United States categorized the severity of the pandemic,

relying on what we call the CFR or the proportion of death among cases.

And based on the CFR, the pandemic, or

the expected pandemic, was categorized into five different types.

The one was a seasonal influenza equivalent or category one,

in which the case fatality should be less than 0.1%.

Category two is a little severe one and

comparable to Asia and Hong Kong influenza pandemic.

The case fatality ratio should be from 0.1 to 0.5%.

So, category three, 0.5 to 1%, category four, 1% to 2%.

And the category five was considered as the most severe for

action of the pandemic, which is equivalent to Spanish influenza, in which

the case fatality risk maybe 2% or greater.

And the following that the Pandemic Severity Index,

the first study that was offered in

May 2009 was most influential to decide our policy-making.

So, this is a extract of the abstract of the Fraser, et al,

published in Science in May 2009.

This was a real-time researcher response, analyzing the case data in Mexico,

especially in a rural area, La Gloria data, estimating the transimissibility,

as well as the severity of the H1N1-2009 pandemic.

So, based on the confirmed case data, including a suspected case data and

measuring how many died among them, the case fatality risk or

what they call the case fatality ratio was estimated at 0.4%.

The resulting conclusion it's going, although substantial uncertainty remains,

clinical severity appears less than that seen in the 1918 influenza pandemic but

comparable with that seen in the 1957 pandemic.

In other words, in the list of the Pandemic Severity Index

the study indicated that we might be around the category 2.

And the two other published studies

offer the similar interpretation of the severity of the pandemic.

And we were forced to accept that the forthcoming pandemic may be as

bad as Asian influenza.

However, as time goes by, another study appeared that carefully

looked at the bias that may be introduced by the diagnosis.

So, the confirmed suspected cases may represent only

the tip of the iceberg and that study similarly illustrated

the concept by using a two layers of the populations.

So here this triangle shows the two samplings that took place in two

different locations.

One is at New York and the other is in Milwaukee.

Among the hospitalized the patients in New York,

the proportion of dead individuals was measured.

And in Milwaukee, among the cases of influenca who required medical attendance,

the proportion of the hospitalized cases was measured.

And among it applying the IDs as the proportions,

the researchers obtain the estimate of what they call the sCFR,

symptomatic of case fatality risk, which was estimated to be

0.048%, almost one-tenth of the initial estimate.

What this study indicated was

that according to the Pandemic Severity Index, we are not at a category 2 or 3.

But the H1N1 2009 was truly a milder infectious disease and

it's comparable to seasonal influenza.

And some study for the same in notion at

the later stage of the 2009 pandemic.

So what we studied from that Pandemic Severity Index issue was that

the influenza epidemic, the number of confirmed cases will be

much smaller than the total number of infections.

So many infected individuals do not develop symptoms or even provided you

might develop symptoms, you do not see a doctor, just stay in the home.

So most of the infections are very mild and don't result in medical attendance.

And that has caused the problem of the calculating wrong the CFR.

The CFR was prone to ascertainment bias,

the bias introduced by the limited frequency of diagnosis.

And from that issue that we have a study that we need a seroepidemiology or

the measurement of the serological data of the population

to grasp the overall magnitude of the infection.

For instance, this proportion, especially the bolder line shows age

specific proportion over infected individuals measured by serum

samples obtained from the Japanese population across all age groups.

As we observe here, among the school children the infection was most common.

And all the others, the infection was less common.

And are not only capturing the overall magnitude

of the epidemic that estimated epidemic size appeared to be helpful

to offer to improve the measurement of the risk of this given infection,

which is referred to as the infection fatality risk.

Infection fatality risk Is a proportion of deaths among all infected individuals.

Namely, dividing that number of deaths by the estimated number of

infected individuals, those animated here.

The infection fatality risk among 100,000 infections was obtained.

As we observed among children and the young adults it is almost

certain that the infection do not result in any deaths,

due to H1N1-2009 influenza.

However, among elderly, especially 60s who have probably the underlying co-morbidity,

it's much more common to experience the risk of death of influenca.

So findings from the seroepidemiological study can be animated for us here.

So the seroepidemiological study enable us

to infer the cumulative incidence of H1N1-2009.

And the age-dependent pattern of the infection was

seen high cumulative incidence was seen in children or

the incidents among the older adults was very low.

And that allowed to measure the proportion of death among adults

without experiencing the ascertaiment of bias.

So age-specific severity profile was estimated on a per-infection basis and

that infer what we call the infection fatality score, IFR.

However, there is a drawback.

It's difficult to assess the magnitude of an infection in real-time

during the very early stage of an epidemic.

Seroepidemiological study requires a large number of samples, and

the real-time seroepidemiological studies needs some science preparedness.

Namely the preparation of the survey in advance of the emergence

of the pandemic, which is a challenge for the next study gate.

So in summary, we have seen that ascertainment bias could

lead to overestimate the risk of death given infection.

Age-specific seroepidemiological survey has enabled us to

obtain age-specific infection profile directory.

And that has improved the estimation of the risk of death on per-infection basis.

However, the estimation was not attained in real-time.

Thank you.

探索我們的目錄

免費加入並獲得個性化推薦、更新和優惠。

Coursera 致力於普及全世界最好的教育，它與全球一流大學和機構合作提供在線課程。

© 2019 Coursera Inc. 保留所有權利。

通過 App Store 下載

通過 Google Play 獲取