DARM DA study exercise group 3

This page is a knowledge crystal of subtype assessment. The page identifier is Op_en4968
Moderator:Carmen Gil (see all)
Upload data {{#opasnet_base_link:Op_en4968}}

For some guidance see the discussion page: D↷

Introduction

Since travel restriction for an indetermined period of time is an unrealistic measure, more feasible tools to control the spreading of the swine flu to Finland are needed. We propose the use of thermal-image scanners combined with a PCR flu test.

The symptoms of swine flu are similar to regular flu symptoms. The most common symptoms include a sudden fever, a sore throat, cough and a runny nose. The incubation period for influenza (the time between infection and appearance of symptoms) is usually 2–3 days, but symptoms may start 1–7 days from infection. Because the fever is one of the most common symtoms of swine flu, the idea is using the thermal-image scanners to prevent swine flu spreading with passengers arriving from abroad in Finland. The scanners could be set up at airports and other border control points for checking all incoming passengers. Thermal scanning is a quick and non-intrusive system for mass screening of people from a distance of more than one metre. The scanners use thermal imaging to assess the skin temperatures of people as they pass through a checkpoint, and transform that data into a coloured image on a screen. The scanners will be set up so that when someone with raised body temperature, passes through the testing area, his/hers features will be highlighted in a particular colour. The threshold temperature of scanner will be above normal body temperature (e.g. to 38 C). Those who are monitoring the passengers passing by the scanner would then pull over the persons who are suspected to have a fever for further investigation. Whether the observed higher body temperature is really the result of fever, is needed to check and confirm by a thermometer before further tests.

If the fever is confirmed, the person will be quarantined/hospitalized for the time of laboratory tests that are required to confirm whether the person has really been infected with influenza A(H1N1)virus. There are several methods for testing swine flu, but use of PCR-test (real-time reverse transcriptase-polymerase chain reaction, rRT-PCR) is recommended for confirmation of novel cases, because PCR-test is recognized to be one of the most effective and sensitive tests (http://www.cdc.gov/h1n1flu/specimencollection.htm). PCR-test uses a device to amplify copies of genes so researchers can easily compare a sample taken from a sick person to the genetic material of the potentially pandemic infection. Running of PCR-test in laboratory takes few days (48-96 hours: http://www.cdc.gov/h1n1flu/guidance/diagnostic_tests.htm) and during that time person who is tested will be quarantied. Although the accuracy of PCR-test is quite good: high sensitivity 86-100% (http://www.cdc.gov/h1n1flu/guidance/diagnostic_tests.htm) and specificity 99 %, the test can also provide "false negative" and "false positive" results. "False negative" is in cases where the person actually has the virus and the a negative result does not, by itself, exclude the possibility of swine flu virus infection. Another problem is that a positive result only indicates that the patient is presumptively infected with swine flu virus, but not the stage of infection.

When the positive swine flu case is detected, the patient will be kept in quarantine at the hospital where the treatment with drugs will be started. By keeping the swine flu patients in quarantine will be prevented the swine flu speading to the Finnish population.

Purpose

The question to be answered by this assessment is: can the use of thermal scanners combined with PCR tests prevent the spreading of swine flu to Finland if all passengers arriving to from abroad will be scanned at the border control points?

Scope

This assessment is carried out to find out whether the installation and use of thermal scanners (and performance of PCR swine flu tests if fever is detected) to the Finnish airports, harbours and other border controls is a relevant system to prevent the swine flu spreading to Finland.

Boundaries

Our assessment takes place at the moment when swine flu was starting to spread from Mexico, but no cases had yet been detected in Finland (end of April 2009). At that time, The Finnish Ministry of Foreign Affairs was not issued any travel restrictions to the affected areas. The scanners would be placed in the Finnish airports, harbours and other border control points and all the passengers arriving to Finland from abroad would be scanned. The scanning of passangers would be continued untill the decision about swine flu vaccination campaign in Finland will be made.

Scenarios

1. Scanners are used and the passangers with a raised body temperature (temp. above normal temperature 37±0.5 C) are tested for swine flu infection through the PCR test and quarantined/hospitalized if test is positive.

2. No scanners are used and the passangers can arrive to Finland freely without monitoring and controlling possible swine flu infection.

Intended users

The intended users of assessment will be:

The Finnish Ministry of Social Affairs and Health

The Finnish Ministry of Foreign Affairs

The Finnish Ministry of Transport and Communications

The Finnish Ministry of Employment and Economy

The Finnish Customs

Participants

The participants doing this assessments are: Bate Shadrack, Carmen Gil, and Minna Ruokolainen from the group 3 of Darm-course 2011

Definition

Add a legend for you diagram.

Add a legend for you diagram.

Decision variables

Decision variable are

Using thermal scanners combined with PCR test

Not using thermal scanners

Indicators

Passengers infected with influenza virus H1N1 get through the passenger control points on their arrival to Finland

Other variables

• Swine flu incubation: 48 hours on average
• Infected individuals are infectious for up to 10 days
• Sensitivity of thermal scanner: 70%
• Specificity of thermal scanner: 92%
• Sensitivity of the PCR swine flu test: 86-100%
• Specificity of the PCR swine flu test: 99 %
• Number and origin of passengers travelling to Finland per day: 150000 [1]
• Prevalence of swine flu in different countries
• Intake of antipyretic drugs (e.g. painkillers)
• Spreading of the virus in Finland [2]

Calculation

1. To estimate the fraction of arriving passengers that could be infected, we need to know the departure country of passengers, the numbers of passengers per day, and the prevalence of swine flu in each departure country
2. To estimate the accuracy of thermal scanner determining if the passenger has fever or not, we need to know the sensitivity and specificity of the scanners. The detectability of the fever is also dependent on the likelihood that passenger is using antipyretic drugs. We also need to take into account the incubation time for swine flu, because during this period infected people do not yet have fever. In this step we also need to know the prevalence of swine flu. As well it is needed to know the body temperature of population, which has a normal distribution.
3. To estimate accuracy of PCR-test in determining whether the fever is caused by swine flu or not, we need to know the sensitivity and specificity of the PCR-test.
4. The outcomes will be expressed as expected utilities. We will assign the utility 0 for swine flu and utility 1 for healthy.

----#: . One critical things in the assessment is the amount of swine flu cases that need to enter Finland to probably cause a wide spread of swine flu. How do you estimate that? Also, do you have a time dimension here, i.e. are you happy if you can postpone the epidemic in Finland for a month, even if it finally comes? --Jouni 09:03, 29 March 2011 (EEST) (type: truth; paradigms: science: comment)

Analyses

EXTRA
* Analyses: statistical and other analyses that contain two or more variables, e.g. optimizing.

Result

To make a final decision we should calculate the expected utilities of using thermal scanners combined with PCR-tests or not. To make the decision of using scanners we would need the expected utility to be very close to 1. Otherwise there would be too high probability of swine flu spreading to Finland and the scanners would be useless.