• FluTrackers.com Inc. does not provide medical advice. Information on this web site is collected from various internet resources, and the FluTrackers board of directors makes no warranty to the safety, efficacy, correctness or completeness of the information posted on this site by any author or poster. The information collated here is for instructional and/or discussion purposes only and is NOT intended to diagnose or treat any disease, illness, or other medical condition. Every individual reader or poster should seek advice from their personal physician/healthcare practitioner before considering or using any interventions that are discussed on this website. By continuing to access this website you agree to consult your personal physican before using any interventions posted on this website, and you agree to hold harmless FluTrackers.com Inc., the board of directors, the members, and all authors and posters for any effects from use of any medication, supplement, vitamin or other substance, device, intervention, etc. mentioned in posts on this website, or other internet venues referenced in posts on this website.
  • We are not asking for any donations. Do not donate to any entity who says they are raising funds for us.

Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication

german-docter

Well-known member
Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

Epidemiological Model of possible Influenza A (H1N1) Epidemic in Germany (SEIR-Model of Robert Koch Institut): Early interventions save time, but total number of cases only minimally reduced.

M?glicher Verlauf einer Epidemie durch das Neue Influenzavirus
A/H1N1 in Deutschland und Auswirkungen pr?ventiver
Ma?nahmen des ?ffentlichen Gesundheitsdienstes

http://www.rki.de/cln_091/nn_200120...ile.pdf/Schweineinfluenza_Epid_vorab_2209.pdf

Aufgrund einer Studie mit mathematisch-epidemiologischen Modellberechnungen (SEIR-Modell) geht das Robert Koch Institut davon aus, dass vor allem durch entschlossene Fr?hma?nahmen des ?ffentlichen Gesundheitswesens ein entscheidender Zeitgewinn hinsichtlich der Ausbreitung der Epidemie der neuen Influenza A (H1N1) in Deutschland erzielt werden kann. Ab einer gewissen Zahl von F?llen w?re dem Modell zufolge die Gesamterkrankungsrate in der Bev?lkerung jedoch nur noch minimal zu beeinflussen.

Das epidemiologische Bulletin des Robert Koch Instituts benutzt f?r sein Epidemiemodell die Fraser-Daten aus Science (http://www.sciencemag.org/cgi/rapidpdf/1176062v2.pdf ) und spielt, ausgehend vom Auftreten des ersten Indexfalles in Deutschland, zun?chst drei sog. SEIR-Modellszenarien mit der Basisreproduktionszahl R0 von 1,34, 1,58 und 2,04 (untere und obere 95-% Konfidenzintervalle) durch.

(Anmerkung 1: Die Basisreproduktionszahl R0 ist ein Ma? f?r die ?bertragbarkeit der Infektion von einem Infektionstr?ger auf weitere Personen).

Das Modell geht von der Annahme einer 75 %-igen Reduktion der ?bertragung durch die Gegenma?nahmen des ?ffentlichen Gesundheitsdienstes aus (sogenannte KfM. KfM bedeutet: Kombination fallbezogener Ma?nahmen) und wird dann in drei Szenarien mit unterschiedlichen Fall-Erkennungsquoten von 10%, 30 % bzw. 50% weitergef?hrt.

Die Hospitalisation des ersten best?tigten Falles in Deutschland war am 24. April 2009.

Hiervon ausgehend ergeben sich f?r die Modelle mit den drei R0 -Werte drei glockenf?rmigen Epidemiekurven mit Pr?valenzpeaks nach 54 Tagen (mit R0 von 1,34: Das w?re der 17. Juni 2009), 81 Tagen (mit R0 von 1,58. Das w?re der 14. Juli 2009) bzw. 120 Tagen (mit R0 von 2,04. Das w?re der 22. August 2009).

In einem zweiten Schritt wurde das Modell fortgef?hrt unter der Annahme von weniger aggressiven und dann auch weniger erfolgreichen Gegenma?nahmen des ?ffentlichen Gesundheitsdienstes ab Fall 500 bzw. Fall 1000 (KfM2). Diese Ma?nahmen beschr?nkten sich wegen der dann eingeschr?nkten Kapazit?ten des Gesundheitsdienstes im Wesentlichen auf die Verminderung der Infektionskontakte im Haushalt der Infizierten. Hierdurch k?me es dann, ausgehend von der Basisreproduktionszahl R0 von 1,58 und den o. g. 81 Tagen ab dem Indexfall, zu einer weiteren Verschiebung der Pr?valenzpeaks der Epidemie-Kurven um zus?tzlich mindestens 6 und h?chstens 56 Tage.

(10% erkannt: + 6 Tage, peak am 20. Juli 2009, 30 % erkannt: + 22 Tage, peak am 5. August 2009, 50% erkannt: + 56 Tage, peak am 8. September 2009).

Es wird jedoch ausdr?cklich betont, dass wegen der vielen m?glichen Einflussfaktoren die Modellberechnungen f?r Prognosen grunds?tzlich nicht geeignet seien.

?Die Modelle sollen aber verdeutlichen, welchen Einfluss getroffene Ma?nahmen des ?ffentlichen Gesundheitswesens auf den Verlauf einer Epidemie durch das neue Influenzavirus A (H1N1) haben k?nnen.?

Interessant ist dabei die Feststellung, dass sich ab einer gewissen Durchdringung der Infektion die absolute Anzahl der Erkrankten kaum mehr ?ndern lasse, es komme dann nur noch zu einer zeitlichen Verschiebung der Epidemiekurve, bzw. des Kurvenpeaks.

?W?hrend sich die Gesamterkrankungsrate nur minimal ?ndert, wird der Gipfel um 6, 22 und 56 Tage sp?ter erreicht.?

(Anmerkung 2: Dieses Modell geht nat?rlich von einer praktisch fehlenden Immunit?t der Bev?lkerung und einem Szenario ohne Impfma?nahmen aus).

Interessant ist vor allem, dass in diesem Szenario die entschlossenen Erstma?nahmen zur Verminderung der ?bertragungsdynamik (KfM1 f?r die ersten 1000 F?lle) bereits den entscheidenden Zeitgewinn bedingen (5 von 6 Tagen, 16 von 22 Tagen bzw. 40 von 56 Tagen).

Das f?hrt auch dazu, dass im Falle eines niedrigeren R0 (von z.B. 1,34 statt 1,58) dieser Zeitgewinn ganz erheblich ansteigt (11 statt 5 Tage, 39 statt 16 Tage und 121 statt 40 Tage).

Das Fazit ist daher:

?Insgesamt zeigt sich, dass die Betreuung der fr?hen F?lle einen besonders gro?en Zeitgewinn bewirkt...?

Dieser Effekt (durch die fr?hen Gegenma?nahmen KfM1) werde andererseits jedoch auch relativ schnell aufgebraucht, wenn die Zahl der t?glich neu importierten F?lle deutlich h?her sei als angenommen (im Modell wurden 2 von au?en neu nach Deutschland importierte Influenzaf?lle pro Tag kalkuliert; bei 5 ?Neuimporten? pro Tag geht der initiale Zeitgewinn bei einem R0 von 1,34 von 5 auf 4 Tage zur?ck (10 % Surveillance-Sensitivit?t), bzw. von 16 auf 10 Tage (30 % Surveillance-Sensitivit?t) und von 40 auf 26 Tage (50 % Surveillance-Sensitivit?t) ??und bringen immer weniger Zeitgewinn, w?hrend die sp?teren Gegenma?nahmen nach KfM2 wichtig bleiben.?

Anmerkung 3: Erl?uterung der Kombination fallbezogener Ma?nahmen (KfM)

KfM in der ersten, fr?hen Phase (KfM1): Zu diesen geh?ren die Therapie des Falles, die Isolation des Falles, aktive Suche von Ansteckungsverd?chtigen durch den ?ffentlichen Gesundheitsdienst, Absonderung und ggf. Fr?htherapie von Kontaktpersonen (entsprechend den aktuell bestehenden Empfehlungen), wobei angenommen wird, dass die Kombination der Ma?nahmen so wirksam sind, dass die Zahl der angesteckten Personen um 75 % reduziert wird. Da KfM1 sehr ressourcenintensiv sind, wird in den Berechnungen angenommen, dass diese Ma?nahmen nur f?r eine geringe Anzahl von F?llen, z. B. die ersten 500 oder 1.000 F?lle durchgef?hrt werden.

KfM in einer nachfolgenden Phase (KfM2): In einer nachfolgenden Phase wird angenommen, dass die Gesundheits?mter keine aktive Suche von Ansteckungsverd?chtigen mehr durchf?hren, sondern fallbezogene Ma?nahmen nur noch f?r die im gleichen Haushalt lebenden Kontaktpersonen umsetzen (KfM2).
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

> Early interventions save time, but total number of cases only minimally reduced.

> Ab einer gewissen Zahl von Fällen wäre dem Modell zufolge die Gesamterkrankungsrate
> in der Bevölkerung jedoch nur noch minimal zu beeinflussen.

Ab = zeitlich. Also early interventions can be successful. Wrong translation ?

> http://www.sciencemag.org/cgi/rapidpdf/1176062v2.pdf

werde/plane das später zu lesen...


ich erinnere:

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0002606

http://www.advisorybodies.doh.gov.uk/spi/minutes/modelling-implications-summary-jan07.pdf


es hängt natürlich davon ab, wie gefährlich das virus ist.
Für eine CFR von 0.1% wird man nicht die Wirtschaft lahm legen


effect of school closure in Mex:
http://www.flutrackers.com/forum/showpost.php?p=237824
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

Any chance we can get an English translation, or at least a summary of the key findings?

Respectful apologies in advance for not being multi-lingual :oops:

Snick
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

gsgs, thanks for the links ! Very interesting.

Snicklefritz, here's to you:

?Epidemiologisches Bulletin? is published by the Robert Koch Institut to provide current data and information of infectious diseases and public health in Germany.

Robert Koch Institut (RKI): Epidemiologisches Bulletin, Nr. 22 (June 1, 2009, preliminary published on May 20, 2009): Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

http://www.rki.de/cln_100/nn_200120...ile.pdf/Schweineinfluenza_Epid_vorab_2209.pdf

Based on the data of Fraser et al. (Pandemic Potential of a Strain of Influenza A (H1N1): Early Findings), recently published in Science, RKI calculated different models of a possible Influenza A (H1N1) epidemic in Germany.

The purpose of the study was to point out the possible effects on the course of the Epidemic within Germany due to different types of preventive measures of the public health service.

Simulations of the (first) epidemic wave (prevalence over time) are made under simplifying presumptions as follows:


There is no pre-existing immunity in the German population (and within the model vaccination is no matter of option, of course).

There will be an ongoing ?input? of new cases from outside Germany of at least 2 infected persons per day ( and 5 cases per day, respectively).

This is important, because under these prevailing circumstances containment by no means can be an option of the model.

Estimated basic reproduction number R0 for Influenza Epidemic was set R0 = 1,58 (according the Fraser-paper; modified models: R0 = 1,34 and 2,04 respectively, i.e. limits of a 95% confidence interval)

Surveillance-Sensitivity (detecting new cases) was set: 0 %, 10 %, 30 %, and 50 %, respectively.

Two types of interventions of the public health service were defined:

Type 1: Early interventions of the public health service (so called ?Combination of case-related measures 1? = KfM 1), i.e. therapy, isolation, active search of suspected cases, isolation of contact persons and early preventive therapy, eventually.

These measures were assumed to diminish transmission by 75 %.

Type 2: Late interventions of the public health service (KfM 2), i.e. measures limited only on contact persons, who are living in the same household.
These measures were assumed to diminish transmission only by 50 %.

Type 1 (early) interventions will need resources , that could be provided only for a small number of cases, that is up to 500 or 1000 cases. Then public health service will switch to the less effective Type 2-interventions.

Results: Without any measures (Surveillance Sensitivity = 0%) in the first setting with R0 = 1,58 the apex (angular point) of the epidemic wave will be reached on day 81 after the first infection case occurred (day 54 with R0 =1,34 and day 120 with R0 = 2,04). There will be a shift of the epidemic curve to the right for an additional time period of 6, 22 and 56 days, respectively, if the Surveillance Sensitivity is set to 10 %, 30 % or 50 %.

In any case, the main effect in the delay of time is due to the early interventions of the public health service (type 1, KfM 1). The total count of cases will be changed only minimally.
The shape of the epidemic wave will be shifted to the right and it will be less high but broader instead.

In other words: The Epidemic within Germany couldn?t be stopped, as long as the Pandemic is going on ?outside?. So, after containment efforts failed, you can only change the number of cases at a certain period of time, but you can?t significantly reduce the total number of cases.

Nevertheless that time delay may be helpful. The epidemic event will not be overwhelming the capacities of health care facilities, the results of therapeutic measures will become better with time and experience. The numbers of workers will not go below essential limits. In this model of the first epidemic wave in any population without pre-existing immunity, you will not have time enough to gain protection provided by vaccination.

By the way, gsgs, that?s one of the main results in the UK-SAG Subgroup-Paper:
First wave: Delay of 1 or 2 weeks to 3 to 4 months. There would be little impact on the total number of cases.


http://www.advisorybodies.doh.gov.uk/spi/minutes/modelling-implications-summary-jan07.pdf


Quote:
" 2.1 The initial outbreak
? for R0 of up to about 1.9. Even if the strategy fails to contain the disease, it should delay its progress by around a month.

2.2 International spread

Imposing a 90% restriction on all air travel to the UK would delay the peak of a pandemic wave by only 1 to 2 weeks. On the other hand a 99.9% travel restriction might delay a pandemic wave by 2 months.

If restrictions on travel from all countries which had epidemics of pandemic flu were put in place internationally the effect could be somewhat greater: a 90% reduction might delay the spread by 3 to 4 weeks and a 99.9% effective ban by 3 to 4 months.

For all practical levels of restriction, there is little probability of a country missing the pandemic altogether due to travel restrictions; however some poorly connected countries might miss an epidemic given a 99.9% ban.

2.3 Geographical spread within the UK


There would be little impact on the total number of cases.

2.4 Spread among, and impact on, the UK population


However, mass treatment of clinical cases with antivirals would flatten this temporal profile, lowering the peak and lengthening the base.

2.5 The second wave


The impact of vaccination with a pandemic-specific vaccine, if it were available, is entirely dependent on the timing and size of any second and subsequent waves in relation to the fist wave and hence inherently difficult to estimate. "


The Davay-Paper, referring to the proposed US guidance, is different in various aspects and focuses on some other items.

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0002606

Quote:
"Unfortunately, stopping the global spread of an influenza
pandemic may be close to impossible.

This exploration enables us to identify and
choose community mitigation strategies to implement that
minimize illness, death and loss of workforce regardless of
transmission to our community from outside ?


Based on the findings of our study, we recommend pandemic
policy in 3 areas that focus on the critical enablers of resilient
community containment: priority for the preparation and
implementation of interventions; regional vs. local application of
interventions; and targeted administration of pre-pandemic
vaccine.
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

That's awesome Doctor. Thank you very much!
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

:oops:Correction:

" ... Results: Without any measures (Surveillance Sensitivity = 0%) in the first setting with R0 = 1,58 the apex (angular point) of the epidemic wave will be reached on day 81 after the first infection case occurred (day 54 with R0 =1,34 and day 120 with R0 = 2,04). There will be a shift of the epidemic curve to the right for an additional time period of 6, 22 and 56 days, respectively, if the Surveillance Sensitivity is set to 10 %, 30 % or 50 %. ..."

must be the other way round, of course:


"... day 54 with R0 = 2,04 and day 120 with R0 =1,34 ..."
 
Re: Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

?Epidemiologisches Bulletin? is published by the Robert Koch Institut to provide current data and information of infectious diseases and public health in Germany.

Robert Koch Institut (RKI): Epidemiologisches Bulletin, Nr. 22 (June 1, 2009, preliminary published on May 20, 2009): Possible Course of an Influenza A (H1N1) Epidemic in Germany and Effects of Preventive Measures of the Public Health Service (preliminary publication)

link (original article in German) has moved to:

http://www.rki.de/cln_100/nn_137849...Id=raw,property=publicationFile.pdf/22_09.pdf

Based on the data of Fraser et al. (Pandemic Potential of a Strain of Influenza A (H1N1): Early Findings)
http://www.sciencemag.org/cgi/rapidpdf/1176062v2.pdf)

, recently published in Science, RKI calculated different models of a possible Influenza A (H1N1) epidemic in Germany.

The purpose of the study was to point out the possible effects on the course of the Epidemic within Germany due to different types of preventive measures of the public health service.

Simulations of the (first) epidemic wave (prevalence over time) are made under simplifying presumptions as follows:


There is no pre-existing immunity in the German population (and within the model vaccination is no matter of option, of course).

There will be an ongoing ?input? of new cases from outside Germany of at least 2 infected persons per day ( and 5 cases per day, respectively).

This is important, because under these prevailing circumstances containment by no means can be an option of the model.

Estimated basic reproduction number R0 for Influenza Epidemic was set R0 = 1,58 (according the Fraser-paper; modified models: R0 = 1,34 and 2,04 respectively, i.e. limits of a 95% confidence interval)

Surveillance-Sensitivity (detecting new cases) was set: 0 %, 10 %, 30 %, and 50 %, respectively.

Two types of interventions of the public health service were defined:

Type 1: Early interventions of the public health service (so called ?Combination of case-related measures 1? = KfM 1), i.e. therapy, isolation, active search of suspected cases, isolation of contact persons and early preventive therapy, eventually.

These measures were assumed to diminish transmission by 75 %.

Type 2: Late interventions of the public health service (KfM 2), i.e. measures limited only on contact persons, who are living in the same household.
These measures were assumed to diminish transmission only by 50 %.

Type 1 (early) interventions will need resources , that could be provided only for a small number of cases, that is up to 500 or 1000 cases. Then public health service will switch to the less effective Type 2-interventions.

Results: Without any measures (Surveillance Sensitivity = 0%) in the first setting with R0 = 1,58 the apex (angular point) of the epidemic wave will be reached on day 81 after the first infection case occurred (day 54 with R0 = 2,04 and day 120 with R0 =1,34 [= correction]). There will be a shift of the epidemic curve to the right for an additional time period of 6, 22 and 56 days, respectively, if the Surveillance Sensitivity is set to 10 %, 30 % or 50 %.

In any case, the main effect in the delay of time is due to the early interventions of the public health service (type 1, KfM 1). The total count of cases will be changed only minimally.
The shape of the epidemic wave will be shifted to the right and it will be less high but broader instead.

In other words: The Epidemic within Germany couldn?t be stopped, as long as the Pandemic is going on ?outside?. So, after containment efforts failed, you can only change the number of cases at a certain period of time, but you can?t significantly reduce the total number of cases.

Nevertheless that time delay may be helpful. The epidemic event will not be overwhelming the capacities of health care facilities, the results of therapeutic measures will become better with time and experience. The numbers of workers will not go below essential limits. In this model of the first epidemic wave in any population without pre-existing immunity, you will not have time enough to gain protection provided by vaccination.

By the way, gsgs, that?s one of the main results in the UK-SAG Subgroup-Paper:
First wave: Delay of 1 or 2 weeks to 3 to 4 months. There would be little impact on the total number of cases.


http://www.advisorybodies.doh.gov.uk/spi/minutes/modelling-implications-summary-jan07.pdf


Quote:
" 2.1 The initial outbreak
? for R0 of up to about 1.9. Even if the strategy fails to contain the disease, it should delay its progress by around a month.

2.2 International spread

Imposing a 90% restriction on all air travel to the UK would delay the peak of a pandemic wave by only 1 to 2 weeks. On the other hand a 99.9% travel restriction might delay a pandemic wave by 2 months.

If restrictions on travel from all countries which had epidemics of pandemic flu were put in place internationally the effect could be somewhat greater: a 90% reduction might delay the spread by 3 to 4 weeks and a 99.9% effective ban by 3 to 4 months.

For all practical levels of restriction, there is little probability of a country missing the pandemic altogether due to travel restrictions; however some poorly connected countries might miss an epidemic given a 99.9% ban.

2.3 Geographical spread within the UK


There would be little impact on the total number of cases.

2.4 Spread among, and impact on, the UK population


However, mass treatment of clinical cases with antivirals would flatten this temporal profile, lowering the peak and lengthening the base.

2.5 The second wave


The impact of vaccination with a pandemic-specific vaccine, if it were available, is entirely dependent on the timing and size of any second and subsequent waves in relation to the fist wave and hence inherently difficult to estimate. "


The Davay-Paper, referring to the proposed US guidance, is different in various aspects and focuses on some other items.

http://www.plosone.org/article/info:doi/10.1371/journal.pone.0002606

Quote:
"Unfortunately, stopping the global spread of an influenza
pandemic may be close to impossible.

This exploration enables us to identify and
choose community mitigation strategies to implement that
minimize illness, death and loss of workforce regardless of
transmission to our community from outside ?


Based on the findings of our study, we recommend pandemic
policy in 3 areas that focus on the critical enablers of resilient
community containment: priority for the preparation and
implementation of interventions; regional vs. local application of
interventions; and targeted administration of pre-pandemic
vaccine.

Comment:
I noticed a quite similar scientific paper regarding Italy that was posted by ironorehopper (Thank you, ironorehopper ! :applause: ) and wanted to have a second look on the German paper, quoted in the thread above.
Because RKI in the meanwhile had moved the link, I posted it again with the new link and added another link to the original Fraser-paper (SEIR-model).


For the Italian data see thread:
http://www.flutrackers.com/forum/showthread.php?t=118144
BMC Infect Dis. Age-prioritized use of antivirals during an influenza pandemic

Pdf at:
http://www.biomedcentral.com/content/pdf/1471-2334-9-117.pdf
 
Back
Top