Presentation on theme: "Influenza Surveillance Systems in an International Setting"— Presentation transcript:
1Influenza Surveillance Systems in an International Setting Case Study 1Welcome to the classroom case study exercise on influenza surveillance systems in an international setting.
2Learning ObjectivesDefine the surveillance objectives, methods of hospital selection, and key data collection priorities for sentinel surveillance for seasonal influenza and severe respiratory diseasesList appropriate surveillance strategies and trigger criteria needed for the early detection of Influenza A(H5N1) in hospitals and communitiesList appropriate surveillance strategies and trigger criteria needed for a broader pandemic early warning systemDescribe how a sentinel site surveillance system for influenza provides an important support function for a pandemic early warning systemIdentify five ways to enhance human, avian, and pandemic influenza surveillance activities in areas where there are known Influenza A(H5N1) outbreaks in poultryThe learning objectives for this module are the following:Define the surveillance objectives, methods of hospital selection, and key data collection priorities for sentinel surveillance for seasonal influenza and severe respiratory diseasesList appropriate surveillance strategies and trigger criteria needed for the early detection of Influenza A(H5N1) in hospitals and communitiesList appropriate surveillance strategies and trigger criteria needed for a broader pandemic early warning systemDescribe how a sentinel site surveillance system for influenza provides an important support function for a pandemic early warning systemIdentify five ways to enhance human, avian, and pandemic influenza surveillance activities in areas where there are known Influenza A(H5N1) outbreaks in poultry
3Outline Introduction to the scenario Routine Surveillance for Respiratory Disease and Seasonal InfluenzaInfluenza A(H5N1) and Pandemic Early Warning SurveillanceIn this session, we will work through a case study set in a fictional country. It will be your responsibility to make decisions and recommendations for influenza surveillance in this country as we go along.We will begin by introducing you to the geography and infrastructure given for the fictional country in the scenario. We will then examine routine surveillance for respiratory disease and seasonal influenza.
5Introduction to the Republic of Pegu Developing countrySoutheast Asia21 provincesPopulation: 50 millionYou have been asked to travel to the Republic of Pegu as an international consultant on influenza. Your terms of reference for this trip are to work with the local Ministry of Health and other international partners to provide technical assistance on the development of a comprehensive National Influenza Surveillance System.Pegu is a “developing” country in Southeast Asia that contains 21 provinces and a population of approximately 50 million. Its capital is Anawrahta.
6Geography Monsoon climate Bordered by 5 countries Population: Migrants AnawrahtaBordered by 5 countriesPopulation:75% ruralNo highway access First let’s go over the geography of Pegu. [Instructor: Click for first box] There is a monsoon climate.[CLICK] Pegu is bordered by five other countries, and faces the ocean on its southwestern border.[CLICK] Migrants regularly cross these borders and have prompted attempts at coordinated infectious disease surveillance activities among these countries.[CLICK] Approximately 75% of the population live in rural areas and do not have access to highways. Road conditions become poor or impassible during the monsoon season.Migrants
7Health Care Kinds of facilities Each province Each district Teaching hospitalsSpecialist hospitalsProvincial hospitalsDistrict hospitalsLocal health stationsTraditional clinics12 traditional medicine hospitalsEach province16-50 bed hospitalEach districtMedical officer, public health, and medicinePegu Provincial HospitalsAnawrahtaPelu JaghaiDava GharThe Pegu Ministry of Health is the focal point for planning, organization, financing, regulation and provision of health care for the population. Hospitals within the country include teaching hospitals, specialist hospitals, provincial hospitals, district hospitals, and local health stations/traditional clinics. There are 12 traditional medicine hospitals in the country as well.[CLICK] Every province has a bed hospital, and[CLICK] every larger district has a medical officer who has both public health and curative responsibilities.There is an uneven geographical distribution of health care with poor quality of care and low service utilization in rural areas. In a recent health survey, 30% of rural respondents indicated that they would not take sick family members to local hospitals but instead would go to their village healer.Polio is on the verge of being eradicated, reflecting a well-established grassroots polio eradication infrastructure. This has produced a system of village health committees that provide informal communication mechanisms and some infrastructure for social mobilization.
8Influenza Laboratory Testing National LaboratoryPCR DiagnosisClosest WHO Reference Laboratory is in a neighboring countryRegional LaboratoriesSerological DiagnosisInfluenza laboratory testing capabilities: Provincial hospitals and traditional hospitals have no laboratory capacity. The National Laboratory is the only lab equipped to undertake PCR analyses although serologic diagnostic methods are undertaken at regional laboratories. While this laboratory uses the WHO influenza reagent kit, the closest WHO reference laboratory is in the neighboring country to the southeast.No laboratory testingTraditional HospitalProvincial Hospital
9Surveillance Infrastructure National notifiable disease surveillance systemImmediate reportingDiphtheriaCholeraYellow feverRoutine reporting (3 days)Standard reporting formSurveillance infrastructure:There is a national notifiable disease surveillance system in Pegu. Certain diseases such as diphtheria, cholera and yellow fever require immediate reporting while others are reported routinely (usually within 3 days) through a using a standard reporting form (that measures case demographics, condition being reported, hospitalization status, travel history and reporter information). Reporting is passive and is usually the responsibility of the health care provider who makes the diagnosis of a patient's illness.
10Part I: Routine Surveillance for Respiratory Disease and Seasonal Influenza
11Question 1The main goals of routine (seasonal) influenza surveillance include all of the following EXCEPT:Describe virus circulation and provide virus isolates for vaccine developmentProvide rapid response to seasonal outbreaksDefine the epidemiology and patterns of viral circulationProvide a support mechanism for pandemic early warning and monitoring systemsAnswer: b.What are the main goals of seasonal influenza surveillance?The main goals of routine (seasonal) influenza surveillance include all of the following EXCEPT:Describe virus circulation and provide virus isolates for vaccine developmentProvide rapid response to seasonal outbreaksDefine the epidemiology and patterns of viral circulationProvide a support mechanism for pandemic early warning and monitoring systemsAnswer: b The purpose of a seasonal influenza surveillance system is to:1. Describe virus circulation and provide virus isolates for vaccine development, 2. Define the epidemiology of influenza, patterns of viral circulation, clinical manifestations of influenza, and high risk groups for severe outcomes in order to make better recommendations for prevention of influenza infection--especially regarding the use of vaccines, and, 3. Provide a support mechanism for broader pandemic early warning and monitoring systems.It is important to note that much of what is known about influenza is from data gathered in industrialized countries of the world. Very little is known about the occurrence and impact of influenza in the developing world and the tropics. Such a routine influenza surveillance system also has value for pandemic preparedness in that countries need data specific for their situation to plan appropriate interventions and to make accurate predictions and plans for outbreaks and pandemics. The establishment of a strong influenza surveillance system will support pandemic early warning activities through the enhancement of laboratory capacity for specimen testing, creation of a logistical network for specimen collection and transport and the training of a cadre of laboratorians who routinely work with influenza viruses, and can therefore accurately identify a novel virus. While a routine influenza surveillance system does not replace the need for the strengthening of a broader pandemic early warning network, these surveillance systems will help to strengthen the important relationships between trained epidemiologists and educated and aware health care workers. Finally, the only practically feasible formal surveillance for most countries during a pandemic will be to continue what is already done on a routine basis. A few well chosen and maintained surveillance sites will be sufficient for tracking the course of the pandemic, especially if combined with other data sources.
12Question 2What surveillance approach might be used to achieve these goals?Universal surveillanceSentinel site surveillanceInfluenza registryLaboratory-based reportingAnswer: b.Sentinel site surveillance for:Hospitalizations due to respiratory diseaseOutpatient visits for influenza-like illnessWhat surveillance approach might be used to achieve these goals?Universal surveillanceSentinel site surveillanceInfluenza registryLaboratory-based reportingNote to instructor: You may wish to have the class discuss the answer choices and come to a consensusSuggested answer: b. Sentinel site surveillance for hospitalizations due to respiratory disease and possibly also outpatient visits for influenza-like-illness. If sentinel sites are chosen well, a sample of cases can be tested for influenza to meet these surveillance objectives. For the goals of describing virus circulation and the epidemiology of influenza, obtaining a limited amount of high quality data from a few well-run sites is preferable to obtaining a lot of lower quality data from many sites.
13Arrival in Pegu You are to : Evaluate the influenza and respiratory disease surveillance infrastructureWork with the MOH to develop a protocol to implement a sustainable national influenza surveillance systemNow that you have arrived in the Republic of Pegu, your responsibilities include evaluating the influenza and respiratory disease surveillance infrastructure and working with the MOH to develop a protocol to implement a sustainable national influenza surveillance system. You will start Monday with a visit to the Office of the Deputy Director of the Pegu Ministry of Health.
14The Pegu Deputy Director Situation: Due to mass poultry die-offsTeam: Chief Surveillance Officer and the Director of Epidemiology, and youDevelop guidelines for expanding their national pneumonia and influenza surveillance systemUse money from World Bank to develop pandemic early warning networkDue to mass die-offs of poultry, Pegu’s Deputy Director would like you to work with his Chief Surveillance Officer and the Director of Epidemiology to develop guidelines for expanding their national pneumonia and influenza surveillance system to better detect cases of influenza A(H5N1). He also informs you that the World Bank has provided a large amount of money to Pegu to develop a pandemic early warning network.
15Current Case Identification Clinician initiated pnuemonia and influenza surveillance among hospitalized patientsDoctors select hospitalized patientsNo case definitionsNasopharyngeal and serum specimens submitted to regional laboratoriesRegional laboratories test seraNational laboratories test high-priority specimens and confirm positive influenza A results from regional laboratoriesDuring your meeting, Pegu’s surveillance officer joins you in the room and describes the current surveillance system.He tells you that the system uses clinician initiated “pneumonia and influenza surveillance” of hospitalized patients. Doctors select the hospitalized patients that they feel have signs and symptoms compatible with pneumonia or influenza and submit nasopharyngeal and serum specimens to their regional laboratories for confirmation.The regional laboratories will normally undertake serologic testing for confirmation, which necessitates collection of convalescent serum specimens as well. The Pegu Ministry of Health reports that collecting needed convalescent sera from patients for these tests is no problem. Other times, when the specimen is considered to be a priority, the National Laboratory will receive specimens directly, and use real-time RT-PCR analysis.Isolates and specimens associated with all positive influenza A testing results at the regional level are also sent to the National Laboratory for further confirmation. There is adequate surge capacity to undertake PCR subtyping for influenza A, B, H1, H3, and H5 at the National level.
16Current Laboratory Testing Regional laboratories test clinical specimensAcute and convalescent samples for serum specimens90% of specimens tested within 9 daysNational laboratory conducts PCR for severe casesConfirmatory tests within hoursDetailed characterization performed at WHO reference laboratoryNumber of influenza A specimens shared with WHO unknownWhen the regional laboratories undertake serologic testing for confirmation cases convalescent sera are often obtained from cases. When there is a particularly severe or high profile case to be tested, the national laboratory will use PCR analyses for confirmation.Regional laboratories refrigerate specimens and test 90% of specimens within a nine days of receiving them.The National Laboratory undertakes its confirmatory tests within hours of receiving specimens. The maximum biosafety level rating is only level 2 in Pegu. Any detailed antigenic characterization or viral culturing is therefore done at another WHO reference laboratory in a neighboring country.Isolates from all confirmed Influenza A specimens at the regional level are sent to the national laboratory. However, you cannot obtain a consistent answer regarding the number of these isolates which are shared with one of the four WHO collaborating centers.
17Current Surveillance Reports Routine reports monthlyBased on total counts of patients discharged with pneumonia, ARI, or clinician-defined influenzaData presented byAgeGender1% diagnosed with laboratory confirmed influenza3-4% of specimens tested at National Laboratory are influenza positive, annuallyRoutine surveillance reports are prepared regularly and are based on counts and estimated rates of patients discharged with pneumonia, ARI or clinician defined influenza. These data are presented by age and gender. Death or survival is also reported. The reports are typically sent monthly, although the requirement is weekly. In periods of heightened alert, they feel that it would be possible for the reports to be sent on a weekly basis. Weekly or monthly case counts are left blank if no report was received. You are not aware of any training campaigns that have taken place to educate clinicians to report suspected cases of influenza A(H5N1), although some posters were made using NGO funds, and may have been placed in some of the provincial hospitals.The routine surveillance reports suggest that a very small percentage (1%) of these cases are diagnosed as having laboratory confirmed influenza, leading some MOH partners to suggest that influenza is not a very common disease in Pegu. However of specimens tested at the national laboratory, 3-4% were found to be influenza positive annually.
18Current Data Collection on Pneumonia Cases Standard patient-level data form for any patient tested for influenzaConsistent and accurate data entryCase demographicsDate of admissionDate of data entryLimited completeness / updating of fieldsSpecimen collectionDate of illness onsetFeverFinal laboratory resultsCurrently:A standard patient-level data form is completed for any patient that is to be tested for influenza. There is fairly consistent and accurate data entry of case demographics and date of hospital admission into the surveillance database (data entry occurs at regional laboratories).However there is limited completeness of fields reflecting date of specimen collection, date of illness onset, presence of fever, and final laboratory results. Some of this lack of completeness is due to the incomplete and rapid filling of forms by clinicians submitting specimens, and some is apparently due to the lack of timely updating of the database with confirmatory results.
19Your Data Collection All 21 provincial hospitals submit P & I data 4 of 6 you visited report data monthlyCriteria for “pneumonia” discharge are unclearNational laboratory confirms influenza A, B, and subtypesInvolvement with WHO FluNet unclearData do not suggest seasonalityTwo hospitals account for 70% of pneumonia casesFeedback to physicians occurs rarely, if everYou spend the next 3 days visiting six different provincial hospitals, some of which are referral facilities for severe respiratory illnesses. While at the hospitals you notice several things:All 21 of the provincial hospitals have submitted pneumonia and influenza data into the system, but only four of the six hospitals that you visited (those in or near Anawrahta province) appear to be reporting data every month.It is unclear from hospital to hospital what clinical criteria are being used to define a discharge as “pneumonia.”Confirmatory testing for influenza A and B (as well as subtypes H1, H3, and H5) is undertaken at the national laboratory by real time RT-PCR. This is not a WHO reference laboratory and you are unsure how this laboratory is integrated with the WHO FluNet surveillance system.Your limited retrospective review of available records does not suggest any overt seasonality patterns except a bimodal pattern that appears at two hospitals which seem to account for 70 percent of the reported pneumonia cases.Local physicians and record keepers report rarely, if ever, receive summaries of influenza circulation in their area and/or the etiology of the pneumonia cases.This is all you have been able to learn about the existing surveillance system. The Deputy MOH asks you if the existing system will meet objectives for seasonal influenza surveillance.
20Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Before we answer…Question 2: Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Instructor: See next slide for helpful reminders
21Remember….Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Objectives of virologic surveillanceDescribe the epidemiology and burden of disease of influenza, andProvide virologic isolates for vaccine developmentWhen considering this question, first let’s remember the objectives of virologic surveillance:Describe the epidemiology and burden of disease of influenza, andProvide virologic isolates for vaccine developmentKeeping this in mind, let’s answer this question by looking at some specific criteria.
22Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAnswer:Not many metrics have been definedLag of up to 9 days for testing refrigerated specimens is longMay affect influenza confirmation rateSee suggested Lectora formatting at the bottom of the scriptNote to instructor: Lead the class through the discussion of each characteristic, as given on the slides.First, let’s consider the timeliness of the data.Suggested answer:It doesn’t seem like a lot of metrics for timeliness have been defined in this system. However the lag of up to nine days for testing of refrigerated specimens at the regional laboratories may be a little long. This may have something to do with the low influenza confirmation rate being reported.For Lectora:Remember the points reviewed here and determine whether the system does or does not meet the objectives for seasonal, human influenza surveillance.Facts:9 day lag for testing refrigerated specimensNot many timeliness metrics are defined for us to evaluateWhich of the following might be an appropriate indicator for timeliness?Data reporting from the sentinel site to the next administrative level (Example target: 80% of data reports sent within 48 hours of the data reporting deadline) (CORRECT)Data reporting with all fields completed (INCORRECT – this indicator measures data quality, not timeliness)Specimen collection using good laboratory practice (INCORRECT – this is a measure of quality laboratory specimen)Specimen collection to shipment to laboratory for testing (Example target: 80% of specimens shipped within 48 hours of specimen collection) (CORRECT)Processing within the laboratory from receipt of specimen until result is available (Example measurement: the average number of days between receipt of specimens and the reporting of the results) (CORRECT)
23What might be appropriate indicators for timeliness? Question 3What might be appropriate indicators for timeliness?Data reporting, time from:From sentinel site to the next administrative levelFrom administrative level to the national levelTime interval between date of onset of fever and specimen collectionSpecimen testing, time from:Collection to laboratoryReceipt of specimen to test resultLaboratory result to informing referring institution and physician What time intervals are appropriate for quantifying “timeliness”?Instructor: have the class brainstorm the points in the process of surveillance, including both reporting and laboratory testing, where relevant time intervals can be monitored. Click to show suggested answers. Additional discussion points below can be used to get the conversation going, or to conclude the brainstorming session.Suggested AnswersData reporting from the sentinel site to the next administrative levelData reporting from that level to the national levelTime interval between date of onset of fever and specimen collectionSpecimen collection to shipment to laboratory for testingProcessing within the laboratory from receipt of specimen until result is availableResult reporting from laboratory to referring institution and physicianDiscussion points: One method of quantifying timeliness is to calculate the percentage of times that a site achieves targets for specific time intervals, for example, the percentage of times that a site sends reports or specimens to the appropriate place within a specified time frame. A reasonable target might be that 80% of data reports sent within 48 hours of the data reporting deadline or that 80% of specimens shipped within 48 hours of specimen collection. Likewise, for the laboratory, the percentage of samples that are tested and have final results within a target time frame can be calculated. A similar quality metric that can be used is the calculation of the average time to accomplish surveillance activities. For example, a given site that is chronically late in sending data every month might have an average of 4.5 days between the deadline for receipt (the day of the week or month on which reports are due) and actual receipt of data per month over time. For laboratory sample processing, the average number of days between receipt of specimens and the reporting of the results can be measured and followed similarly. Site time averages can be compared to identify sites that are underperforming and to target improvements. Either percentages of sites achieving timeliness targets or time lag averages can also be used as a quality metric to be followed over time.
24Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAcceptabilityAnswer:Lack of feedback to physicians limits acceptability to physiciansQuantitative measures of acceptability can include sentinel site participation rate, interview completion and question refusal rates, completeness of report forms, facility reporting rate, and timeliness of data reporting.Is this system acceptable to stakeholders?Suggested answer:There is apparently very limited feedback of results to the reporting clinicians. Any feedback they would receive would not be likely to be timely enough to affect clinical decision-making. But even regular reports on strain circulation in their region could be diagnostically useful to them. Lack of feedback will limit acceptability to physicians and limit their desire to submit laboratory specimens into the system. It is clear that many already perceive that they are unsure how the data that they report will be used to inform national influenza control policies.For Lectora:Facts to consider:There is limited distribution of surveillance results to reporting cliniciansWhich of the following might be quantitative measures of acceptability?Sentinel site participation rateInterview completion and question refusal ratesCompleteness of report formsFacility reporting rateTimeliness of data reportingAnswer: ALL of these measures indicate a willingness by physicians and facilities to participate in the surveillance system.
25Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAcceptabilityRepresentativenessAnswer:Large % of cases from only 2 hospitalsMany hospitals not reporting regularlyNeed more information to determine representativeness of populationContinuous assessment of acceptability and representativeness should be included in regular surveillance system evaluations. Does this system achieve goals for surveillance in terms of representativeness?Suggested answer:Warning signs for representativeness are that a large percentage of cases are being accounted for by only a two provincial hospitals, although the original system wasn’t overtly designed to be a sentinel system. There appears to be uneven reporting and it is also clear that many provincial hospitals are not regularly reporting data for their provinces into the system. You need to know more about the hospitals and their representativeness of the population at large.To assess representativeness, case characteristics can be compared to census data to ensure appropriate capture of a broad range of society. Furthermore, disease rates determined by special studies or other surveillance activities can be compared to sentinel surveillance data.For Lectora:Facts to consider:A large percentage of reported cases are accounted for by two hospitalsMany provincial hospitals are not regularly reporting data for their provinces into the systemWhich of the following are possible indicators for acceptability?Comparison of case characteristics to census dataCompare disease rates determined by special studies or other surveillance activities to sentinel surveillance data
26Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAcceptabilityRepresentativenessCompletenessAnswer:Some sites over-represented compared to othersLaboratory data variableNeed to re-train clinicians and data-entry staffIndicators of completeness can be determined through analysis of reported data. Suitable completeness indicators may include the percentage of reports received from each site with complete data, percentage of total expected data reports that are received, and the percentage of total expected cases that have specimens submitted to the laboratory. How do you think our system is doing in terms of completeness of reported data?Suggested answer:Data reporting at the site level seems incomplete with an over-representation of some sites relative to others. Completeness of data entered into the surveillance database for cases tested in the laboratory also appears to be variable given a variety of reported empty fields such as date of illness onset and final confirmation results. There is a need for refresher training on form completion by clinicians, and a need to remind data entry staff to update the database more frequently with final laboratory results.For Lectora:Facts to consider –Some sites over-represented compared to othersLaboratory data variableNeed to re-train clinicians and data-entry staffWhich of the following are adequate indicators of completeness of data?Percentage of reports received from each site with complete data (CORRECT)Percentage of total expected data reports that are received (CORRECT)Percentage of reports received by the first administrative level (INCORRECT)Percentage of total expected cases that have specimens submitted to the laboratory (CORRECT)
27Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAcceptabilityRepresentativenessData Validity / Data QualityAnswer:No case definition limits ability toDetermine baselineInterpret trendsEstimate rates of illnessAssess risk factorsIncomplete reporting by most facilitiesLong refrigeration affects specimen qualityWhat about data quality?Suggested answer:A lack of standardization in what is reported (e.g. no case definition) across hospitals limits any ability to determine a baseline of pneumonia or influenza cases by season, and severely limits the possibility of producing interpretable trends in respiratory disease in this population. This is a core component of data validity as without a standard case definition one cannot make estimates of rates of illness, or assess the risk factors for pneumonia/influenza in the population. The apparently incomplete reporting by most of the facilities in the system also limits any ability to draw conclusions about laboratory confirmed influenza as a cause of hospitalized patients in the country.The combination of lack of a standard case definition and fairly long refrigeration times for specimens prior to laboratory testing with DFA may be issues of data quality that are influencing the low influenza positive rate being observed at regional laboratories in the system. Similarly, this limitation may also reduce the utility of the surveillance system to inform influenza control policy (a concern raised by the clinicians in Pegu).Discussion Points: Once a case definition is established, regular field evaluations and audits at a facility level must be a standard component of the surveillance system. This process can determine that cases are being counted appropriately, that reported cases meet the case definition, and that sampling procedures are being employed uniformly without evidence of bias. Data values recorded in the surveillance system can be compared to a gold-standard of chart-review values by a retrospective review of a sample of medical records. If a sampling procedure is employed for specimen collection, audits can ensure procedures are uniform without evidence of bias. Additionally, audits can determine whether clinical specimens are being taken, stored, processed, tested (if appropriate), shipped properly, and shipped in a timely fashion from all those who meet sampling criteria.For Lectora:Facts to consider –-No case definitionIncomplete reporting by most facilities in the systemLong refrigeration times affect specimen qualityWhich of the following methods could be used to increase data validity and/or quality?Regular evaluations of laboratory protocols (INCORRECT)Regular field evaluations and facility audits (CORRECT)Compare data from surveillance system to data obtained from a retrospective sample of gold-standard chart reviews (CORRECT)Regular written exams of participating clinicians (INCORRECT)
28Question 3Does this system achieve the objectives for seasonal/human influenza surveillance that were discussed earlier? Why or why not?Consider:TimelinessAcceptabilityRepresentativenessData Validity /Data QualityFlexibilityAnswer:With appropriate laboratory facilities, the system may be flexible enough to identify respiratory pathogens in circulationWith case definitions, the system could be expanded to capture a wider range of diseasesIs the system flexible enough?Suggested answer:With appropriate laboratory facilities, this system may be flexible enough to also be used to examine the circulation and relative burden of other respiratory pathogens in hospitalized cases. However this can only happen if standard case definitions are implemented. Once the concept of case definitions is introduced in Pegu, this system could be expanded to include a wider range of hospitalized respiratory diseases under surveillance.For Lectora:Facts to consider:-Laboratory facilities are appropriateClinicians report the cases they think they should (no case definitions are used)Which of the following best describes this system?It is currently flexible enoughWith the implementation of case definitions to capture a range of respiratory diseases, it would be flexible enoughThere is too much work to do to make this system flexibleAnswer: B
29Your RecommendationsDevelop a standard case definition for “severe acute respiratory illness”Formally identify sentinel sitesTraining for sentinel site cliniciansRoutinely send influenza isolates to WHO collaborating centers, and enter into WHO/Flu-NetImplement a plan for regular feedback of surveillance information to cliniciansImmediate notification and response for high priority cases and clustersAfter sharing your perspectives on the different objectives for the national surveillance system, and providing the MOH with a draft report of your initial surveillance evaluation, the Deputy MOH agrees with your argument on the need to build a sentinel site surveillance system for severe respiratory disease and seasonal influenza. Several key recommendations from your surveillance evaluation will be implemented in the new surveillance system:A standard case definition for “pneumonia” will be developedSentinel sites in the surveillance system will be formally identifiedClinicians at sentinel sites will receive training on the future case definition, as well as on the importance of complete epidemiologic data capture for all cases testedInfluenza isolates identified by the surveillance system will be routinely sent to the WHO collaborating centers, and entered into WHO/Flu-Net. Refresher trainings on the need for complete and timely data entry into the surveillance data base will also be undertaken.A plan for regular feedback of surveillance information to clinicians will be implemented. During periods of known epidemic influenza circulation this feedback will occur weekly.The surveillance system will add an immediate notification and response component for high priority cases and clusters so they can be quickly investigated at the request of local public health authorities.
30Your RecommendationsPerformance indicators for objective monitoring and evaluationIncrease laboratory PCR testingAdditional laboratory quality controlPerformance indicators for objective monitoring and evaluation of sentinel sites will be developed. Site visits and data quality audits will be routinely undertaken.The national laboratory will increase the number of specimens tested annually by PCR methods. They will provide additional quality control and quality assurance on the methods of specimen collection, packaging, transport and testing that are being implemented at regional laboratories. They will provide random “blinded panels” of influenza specimens to the regional labs to assess the quality of their confirmatory methods, and will also test random samples of “influenza negative” specimens submitted by the regional laboratories.
31Your Next Task Work with MOH of Pegu Write a formal set of national guidelinesOutline the approach to establish sentinel surveillanceStandard case definition of SARI among hospitalized inpatientsStandard case definition if ILI among outpatientsNow that you have finished your initial task, your next assignment is to work with the MOH of Pegu to develop a formal set of national guidelines. These guidelines will outline the approach to establish sentinel surveillance, and to provide standard case definitions of SARI among hospitalized inpatients and less severe ILI among outpatients.
32Question 4What criteria will you use to decide where sentinel hospitals should be located?Answer:Representative of a defined populationReasonable logistics within the hospital forCase identificationSpecimen collectionSpecimen transportationPolitically acceptablePractically FeasibleAdded benefit: Location in “high risk” locationNumber of facilities selected will be based on local resourcesEach facility should have a focal point to oversee collection and reporting of data and specimensQuestion 3: What criteria will you use to decide where sentinel hospitals should be located?Discussion Points: Ideally, selected sites would be representative of a wide cross-section of ethnic and socioeconomic groups of the country or region. In addition, because influenza virus activity varies with climate, it is important to include sites from different climatic regions. Placement of sites in areas where the population denominator can be estimated will facilitate burden of disease estimates. Ultimately, the choice of sentinel hospitals will often be based on practical issues of feasibility such as human resources, communication infrastructure, and the availability of specimen transport. There is no ideal number of surveillance sites, and the number chosen by a particular country will depend in part on resources available. There should be a focal point at each hospital that oversees collection and reporting of data and specimens.Suggested answer:Important characteristics of chosen sentinel sitesRepresentative of a defined populationReasonable hospital logistics for case identification, specimen collection, and transportPolitically acceptablePractically feasible! An “added benefit”: Surveillance sites can also be positioned to attempt to detect human cases of novel influenza virus infection (for example, near areas of intensive poultry husbandry). If the characteristics of the chosen sentinel sites listed above are also achieved then this may provide added support to any existing pandemic early warning systems.For Lectora:Which of the following are criteria that you should consider when deciding which hospitals to use in the sentinel system? (Choose all that apply)Is the facility nationally recognized as a prestigious institution?Are the hospital /clinic logistics sufficient for case identification, specimen collection, and transport?Does the facility collect at least 20 respiratory samples per day on average?Is the patient population representative of the region or country?Is this choice practically feasible and likely to succeed?Answers: b, d, and e.
33CDC/WHO SARI Case Definition for persons > 5 years old Lower respiratory tract illness consisting of ALL of the following:Sudden onset of fever over 38°C, ANDCough or sore throat, ANDShortness of breath or difficulty breathing, ANDRequiring hospital admissionSince there is no formal case definition for “pneumonia” in PEGU, you offer to help improve standardization of reporting by offering a CDC/WHO case definition for Severe Acute Respiratory Illness (SARI) for consideration:SARI case definition for persons > 5 years old:ALL OF THE FOLLOWINGSudden onset of fever over 38°C, ANDCough or sore throat, ANDShortness of breath or difficulty breathing, ANDRequiring hospital admission
34CDC/WHO Case Definition for persons < 5 years old Any child 2 months to 5 years of age with cough or difficult breathing and:breathing faster than 50 breaths / minute (2 – 12 months)breathing faster than 40 breaths / minute ( 1 – 5 years)or, Any child 2 months to 5 years of age with cough or difficulty breathing and any of the following general danger signs:Unable to drink or breastfeedVomits everythingConvulsionsLethargic or unconsciousChest indrawing or stridor in a calm childAND Requiring hospitalizationSARI case definition for persons <= 5 years old:Any child 2 months to 5 years of age with cough or difficult breathing and:breathing faster than 50 breaths / minute (2 – 12 months)breathing faster than 40 breaths / minute ( 1 – 5 years)(Infants less than 2 months with fast breathing 60 breaths or more per minute are considered serious bacterial infection).or,Any child 2 months to 5 years of age with cough or difficulty breathing and any of the following general danger signs:unable to drink or breastfeedvomits everythingconvulsionslethargic or unconsciouschest indrawing or stridor in a calm child, ANDRequiring hospital admission
35Question 5Which of the following are reasons why good SARI case definitions are a key data collection priority in Pegu?The use of a SARI case definition provides some standardization of reporting across hospitals and regions.Testing defined SARI cases will yield circulating pathogens and strainsSurveillance using a good SARI case definition will yield a better understanding of epidemiology and burden of respiratory diseaseIt could detect emergence of a new pathogenAll of the aboveAnswer: e.Which of the following are reasons why good SARI case definitions are a key data collection priority in Pegu?The use of a SARI case definition provides some standardization of reporting across hospitals and regions.Testing defined SARI cases will yield circulating pathogens and strainsSurveillance using a good SARI case definition will yield a better understanding of epidemiology and burden of respiratory diseaseCould detect emergence of a new pathogenAll of the aboveAnswer: e. All of the above.Classroom discussion points:For seasonal influenza surveillance, even just testing a random sample of inpatient SARI cases during influenza season should give a sense of the strains that are circulating, the pathogens causing SARI, and the risk factors for severe illness. The use of a SARI case definition will provide some standardization of reporting across hospitals and regions. Surveillance for SARI can be used to understand the epidemiology and burden of other respiratory pathogens beyond influenza. It is a very important clinical entity in its own right and is the second leading cause of mortality in low income countries. It should be better understood so it can be better prevented. Finally, changes in the circulation of SARI could represent the emergence of a new respiratory pathogen in the population.
36Question 6 True or False: One drawback of Pegu’s case definition is that it is not sensitive enoughAnswer:False. Pegu’s case definition is sensitive. This is actually a drawback because the country’s single national laboratory could become overwhelmed with casesQuestion 6:True/False: One drawback of Pegu’s case definition (above) is that it is not sensitive enough.Answer:False. Pegu’s case definition is sensitive. This is actually a drawback because the country’s single national laboratory could become overwhelmed if it received influenza testing requests for SARI cases from too many locations.
37Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected and why?Consider:General informationAnswer:Unique identification number*Medical record number*Name (and parent’s name, if a minor)*Date of Birth*Sex*Address*Date of onset of symptoms*Date of collection of epidemiologic data*Part of an outbreak investigationInpatient or outpatientQuestion 7: What kinds of data should be collected from the SARI cases from which specimens are being collected?First, consider what general information you need.Instructor: Encourage students to think in terms of the six broad categories of surveillance variables, which are given on this and the subsequent slides. These data will be used to describe those persons most at risk for a serious respiratory illness. As an additional benefit, and given some existing concerns about pandemic influenza, questions about risk factors for Influenza A(H5N1) can be easily integrated into the data collection forms to identify pneumonia cases with possible exposure to avian influenza. Some of the possible variables that could be collected in the system include:Suggested answer:Unique identification number*Medical record number*Name (of patient and parent’s name, if a minor)*Date of Birth*Sex*Address*Date of onset of symptoms*Date of collection of epidemiologic data*Part of an outbreak investigationInpatient or outpatientNote that the starred (*) data are the recommended essential data for Severe Acute Respiratory Illness Surveillance.Question 5 for Lectora:There are sixbroad categories of surveillance data that should be collected from the SARI cases from which specimens are being collected. They are: General Information; Specimen Collection Information; Clinical Signs and Symptoms; Risk Factor Information; Pre-Existing Conditions, and Treatment History. List each variable below under its general category of surveillance data. Hospital or clinic phone numbersDate of symptom onsetDOBCase IDUnderlying conditionsDoctor, nurse or laboratory phone numbersDate of reportNameSexSignsSymptomsLaboratory or diagnostic testing informationExposure t:o sick/ill poultry, wild birds, pets, other animals, sick humansAddressRace/ethnicityKnown Contacts in last 7 daysDispositionOccupationPhone numberOutcomeOccupational exposuresAnswers:Identifying information:Demographic information:Clinical InformationRisk Factor InformationReporter information:Hospital or clinic phone numbers
38Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected?Consider:General informationSpecimenNow think about the information you need about specimen collection:Suggested answer:Throat swab specimen – date of collection*Nasal swab specimen – date of collectionOther specimen (if collected) – date of collectionAnswer:Throat swab – date of collection*Nasal swab– date of collectionOther specimen (if collected) – date of collection
39Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected?Consider:General informationSpecimenClinical signs, symptomsAnswer:Fever >38*Cough*Sore throat*SOB/Difficulty breathing*IMCI danger signs (per WHO protocols)*DiarrheaWhat kinds of clinical signs and symptoms will be most useful to know?Suggested answer:Fever >38*Cough*Sore throat*SOB/Difficulty breathing*Integrated Management of Childhood Illness danger signs (per WHO protocols)*Diarrhea
40Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected?Consider:General informationSpecimenClinical signs, symptomsRisk factor informationAnswer:Occupation *Contact with:Suspected H5N1 casesSick or dead poultry or wild birds*Severe respiratory illness casesTravelEating raw or undercooked poultry productsWhat do you need to know about possible exposures to H5N1 or other pandemic pathogens?Suggested answer:Occupation of patient*Suspected H5N1 case (per WHO protocols)Contact with sick or dead poultry or wild birds*Contact with a friend or family who has severe respiratory illnessTravel in an area known to have endemic circulation of avian influenza H5N1Eating raw or undercooked poultry products in an area of H5 N1 virus circulation
41Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected?Consider:General informationSpecimenClinical signs, symptomsRisk factor informationPre-existing medicalAnswer:Liver disease*Kidney disease*Immune compromised state*Neuromuscular dysfunction*Diabetes*Heart disease*Lung disease*Smoking history*What information should you collect about the reporting physician or health care facility?Suggested answer:Liver disease*Kidney disease*AIDS, cancer, or other immune compromised state*Neuromuscular dysfunction*Diabetes*Heart disease*Lung disease*Smoking history*
42Question 7What kinds of data should be collected from the SARI cases from which specimens are being collected?Consider:General informationSpecimenClinical signs, symptomsRisk factor informationPre-existing medicalTreatment historyWhat information should you collect about pre-existing medical conditions?Suggested answer:Vaccination against influenza within the past year*Currently taking anti viral medicineAnswer:Vaccination against influenza within the past year*Currently taking anti viral medicine
43Question 7: Key Points Laboratory-Epidemiology link is critical There must be a system in place where the same unique identifier is place on both sets of dataThe importance of a lab-epi link must also be emphasized as critical. There must be a system in place where the same unique identifier is placed on both the laboratory specimen and epidemiologic data collection forms.
44Chief Surveillance Officer Response Concerned about having too many hospitals report too many SARI cases, overwhelming the laboratoryRandom sampling at hospitals may be complicated for staffInstead suggests sampling all SARI cases from a few hospitalsThe Chief Surveillance Officer recognizes the value of this case definition to detect cases of severe influenza, pneumonia, and even to measure the contribution of influenza to exacerbations of illnesses such as chronic lung disease, heart disease and diabetes. She recognizes that standardization of reporting through the use of a case definition will help the Pegu MOH describe the epidemiology of severe seasonal influenza, and provide some isolates for identification of viruses. However, she raises the concern that if too many hospitals report SARI cases the laboratory and response components of the surveillance system will become overwhelmed. She feels that randomly sampling SARI cases at multiple hospitals would provide good representation of the population, but also thinks that this would be complicated for hospital staff that are new to the concept of this kind of surveillance. Instead, she asks if Pegu might just sample all of the SARI cases from a few hospitals—this way surveillance staff wouldn’t have to be taught the complexities of random sampling. You nod in agreement to her suggestion and agree that “quality data can be obtained from a few well run sites.” You both feel that practically speaking, small amounts of good data are better than large amounts of bad data.” You will initially pilot the SARI surveillance in five geographically diverse hospitals. Two of these hospitals serve as regional reference centers for severe respiratory illness, and three others are a bit smaller but have fairly well defined catchment areas of populations-served.
45Surveillance for Less Severe Influenza Chief Surveillance Officer would like to include less severe, more common, influenza cases in the systemCan be provided by outpatient surveillanceWHO criteria for influenza-like illnessSudden onset of fever over 38CCough or sore throatAbsence of other diagnosesThe next morning the Chief Surveillance Officer greets you with an additional question. She says she would like the surveillance system to also capture some virus isolates and risk factor information from less severe, more common, influenza cases. She notes that certain influenza types and sub-types of influenza (for example influenza B and possibly influenza A (H1N1) may produce less severe disease than influenza A(H3N2). She thinks that only sampling hospitalized patients in the surveillance system could under-represent infections caused by influenza B, and possibly influenza A(H1N1) in the surveillance system.Surveillance for Influenza-Like-Illness (ILI) in outpatients would provide information about trends in less severe influenza-like-illness in Pegu.ILI is defined according to WHO criteria as:Sudden onset of a fever over 38°C, ANDCough or sore throat, ANDAn absence of other diagnoses.
46Question 8How could the sentinel site system be expanded to include some less severe influenza cases?Answer:Implement ILI surveillance in outpatient clinics of SARI sentinel-site hospitalsWeekly counts of ILI outpatient visits testing positive for influenzaChoose small sample of cases for specimen and epidemiologic data collectionQuestion 8: How could the sentinel site system be expanded to include some less severe influenza cases?Suggested answer:The ILI case definition does not require hospitalization or difficulty breathing. If ILI surveillance were implemented in the outpatient clinics of the selected SARI sentinel site hospitals, one could then possibly compare the epidemiologic and demographic risk factors for mild and severe disease.Aggregate weekly counts of ILI outpatient visits and the proportion of weekly ILI cases testing positive for influenza have been shown to rise and fall with influenza virus circulation in a population. As the number of cases presenting to ambulatory care sites is likely to be very large, case counts would be aggregate, and only a small sample of cases would have clinical specimen and epidemiologic data collected. Weekly case counts should be categorized by age group according to well-studied and locally standard age range categories (6 months to 23 months, 2-4 years; 5 – 18 years; years; years; and ≥65 years). Cases chosen to have detailed epidemiologic data and clinical specimens collected should be selected in as unbiased a manner as possible.The selection protocol must take into account local health seeking behaviors such as differential use of evening and weekend clinics. Ideally, the weekly total number of patients seen by clinics by age group will also be collected to allow for proportion of ILI to be calculated. As with SARI Surveillance, ILI Surveillance should emphasize quality data from a few well-run sites, over broad system expansion.For Lectora:Which of the following is a way to expand the sentinel site system to include less severe influenza cases?Implement ILI surveillance in outpatient clinicsInclude severity of illness as a field on the hospital reporting formConduct weekly counts of ILI outpatient visits testing positive for influenzaConstruct an algorithm to identify less severe hospital casesAnswer: a and c.Answer feedback (for any answer):Because the ILI case definition does not require hospitalization or difficulty breathing, it could be implemented in the outpatient clinics of selected SARI sentinel site hospitals; this would give information about epidemiologic and demographic risk factors for mild as well as severe disease.Aggregate weekly counts of ILI outpatient visits and the proportion of weekly ILI cases testing positive for influenza have been shown to rise and fall with influenza virus circulation in a population. This can provide an indication of the timing and severity of the current flu season relative to past seasons. Useful strain surveillance data can also be obtained by taking specimens from a systematic sample of outpatient ILI cases (e.g. the first two or three cases each day) for laboratory testing.
47SARI and ILI Surveillance Outpatient ILI surveillance at 5 SARI sentinel site hospitalsSystematically select first 2 cases each day for laboratory and epidemiologic investigationSentinel hospitals will provide weekly tally of total ILI cases at facilitiesTogether you decide to also implement outpatient ILI surveillance at the five SARI sentinel site hospitals. You agree that systematically selecting the first two cases of ILI each day from the general adult and pediatric clinic for laboratory testing and epidemiologic questioning will be understandable to local staff. The sentinel site hospitals will also provide weekly tallies of the total number of ILI cases presenting to the sentinel site facilities. You spend the next 3 days developing reporting forms and databases and training staff at the first proposed sentinel site. You are amazed at the speed and efficiency with which public health action can take place in Republic of Pegu! It seems that the new SARI surveillance system will soon begin to operate. The CSO proudly provides the new draft guidelines at the morning briefing to the Deputy Minister of Health.
48Part II: Influenza A(H5N1) and Pandemic Early Warning Surveillance
49Media Reports Mass deaths of flocks of chickens, geese, waterfowl Southeastern PeguMinistry of Agriculture investigation3 chicken samples ‘weakly positive’ for Influenza A (H5N1)No systematic avian surveillance existsThat evening at dinner the Deputy Minister of Health indicates that over the last year there have been a series of reports from different media sources indicating mass deaths of flocks of chickens as well as geese and other waterfowl. These deaths have primarily occurred in the southeastern region of Pegu. The Ministry of Agriculture (MOA) recently sent investigators into this region to follow-up on a particularly large outbreak and has reported test results from three samples collected from dead chickens sent to the national laboratory in Anawrahta to be “weakly positive” for avian Influenza A (H5N1). There is no systematic surveillance for avian Influenza A (H5N1) in poultry, wild bird or other animal populations. He is feeling pressure that high level political officials are demanding to hear from about his current plans for “detecting bird flu”, and he now needs you to develop surveillance guidelines for Pegu’s pandemic early warning system. He asks you based on your site visits, “if a human case H5N1 was admitted to a surveillance hospital or to another hospital in Pegu, do you think it would be immediately identified and reported?”
50Question 9Are you confident that a hospitalized human case of Influenza A(H5N1) would be recognized and responded to? Why or why not?Answer: No.There is no system of 24-hr SARI notification and prioritization for influenza A (H5N1) testingRapid detection is needed: Treatment is most effective if given within 48 hours, but infectiousness may occur 24 hrs prior to onset – need to quickly identify cases and contactsQuestion 9: Are you confident that a hospitalized human case of Influenza A(H5N1) would be recognized and responded to? Why or why not?Instructor: pose this question to the students, and allow them to offer their opinions and rationale. If no one is forthcoming to answer, begin first with “Yes or No.” Once someone offers an answer or a consensus is reached (right or wrong), ask why. You can remind them of the structure of the system that has been put in place (sentinel hospitals with all inpatient SARI cases with reported, including specimen collection, and weekly tallies of outpatient ILI, to include 2 specimens tested per day.Suggested Answer:NoThere is not yet any system of immediate notification (within 24 hours) of specific SARI or severe respiratory illness cases that need to be prioritized and tested for the presence of Influenza A(H5N1).The incubation period for avian influenza usually ranges from 2-7 days, but treatment appears to be most effective when it occurs during the first 48 hours of exposure. Cases may also be infectious up to 24 hours prior to symptom onset so there is an urgent need to quickly identify close contacts of any suspected or confirmed cases.For Lectora:True/False. A hospitalized human case of Influenza A (H5N1)in Pegu would trigger the early detection of influenza A(H5N1) in hospitals and communities.Answer: False. The Pegu surveillance system does not have a system of immediate notification (within 24 hours) of specific SARI cases that need to be prioritized and tested for the presence of Influenza A(H5N1).
51Question 10How might surveillance for seasonal influenza support efforts to recognize an emerging pandemic or detect human cases of Influenza A (H5N1)? By counting casesBy creating a logistical networkBy establishing case definitions and reporting criteriaBy tapping into Pegu’s health budgetBy enhancing laboratory capacityNone of the aboveAnswer: b, d.Question 10: How might surveillance for seasonal influenza support efforts to recognize an emerging pandemic or detect human cases of Influenza A (H5N1)? By counting casesBy creating a logistical networkBy establishing case definitions and reporting criteriaBy tapping into Pegu’s health budgetBy enhancing laboratory capacityNone of the aboveAnswer: b, d.The routine surveillance system should also provide a support function to pandemic early warning systems. The recently adopted International Health Regulations (2005) call for strengthened surveillance for all events which may constitute a Public Health Emergency of International Concern, including human cases of novel influenza virus infection. The establishment of a strong influenza sentinel surveillance system will support pandemic early warning activities through:the enhancement of laboratory capacity for specimen testing,creation of a logistical network for specimen collection and transport,the training of a cadre of laboratorians who routinely work with influenza viruses, and can therefore accurately identify a novel virus.These surveillance systems will also strengthen the important relationships between trained epidemiologists and educated and aware health care workers.Students may also consider (next slide)
52Question 10: Additional Answers During a pandemic, data from the routine sentinel site surveillance system will help describe the:Changing geographic location of the virusTrend in casesSeverity of the pandemicA severe pandemic will also place significant strains on health care infrastructure, limiting countries’ capacity to collect and report data. As a result, monitoring the course of the pandemic will need to make use of existing sources of information that are being reported in the inter-pandemic phase as part of routine influenza surveillance, and so is a continuation of current surveillance activities. During a pandemic, data from the routine sentinel site surveillance system will help describe the:changing geographic location of the virus,the trend in cases, and,the severity of the pandemic.
53Question 10: Key PointsAs routine SARI surveillance is instituted, data will be more complete and standardizedSentinel-based surveillance is feasible for most countries to track a pandemicKEY POINTSAs more countries institute routine SARI surveillance at sentinel sites, the data set that may be used to monitor a future pandemic will become more complete and standardized, and thus more useful. The only practically feasible formal surveillance for most countries during a pandemic will be sentinel-based and a few well chosen and maintained sentinel sites will be sufficient for tracking the course of the pandemic, especially if combined with other data sources.
54Your RecommendationsClinicians at sentinel hospitals and non sentinel-site hospitals need to be trained in influenza A (H5N1) screening criteriaCriteria can elevate index of suspicion about SARI casesSurveillance officer agrees that trigger criteria could help prioritize SARI cases for immediate laboratory testingYou suggest that in order to ensure that any human case of Influenza A (H5N1) is rapidly detected, clinicians who work in the sentinel site hospitals, and also those working at non sentinel-site hospitals, need to be trained in additional Influenza A(H5N1) screening criteria. You add that because the clinical presentation of influenza A(H5N1) is similar to severe pneumonia caused by many other pathogens, these screening criteria would have to include epidemiologic risk factors that will help doctors elevate their index of suspicion that some cases of SARI may be more likely than others to be due to infection with influenza A(H5N1). You add that these screening criteria have been called “triggers” (or “signal events”) in other countries’ pandemic early warning systems. The CSO replies that the use of epidemiologic criteria would be a useful way to prioritize certain cases of SARI for immediate alert, investigation and laboratory testing for influenza A(H5N1) , seasonal strains, and possibly other pathogens as well. As you work on the pandemic early warning portion of the surveillance guidelines that morning, you provide the CSO with examples of trigger criteria for Influenza A(H5N1) that were used in a neighboring country.
55Question 11True or False: The proposed epidemiologic “trigger criteria” below could be used to prioritize SARI cases for immediate reporting and laboratory testing for Influenza A (H5N1)Travel within last 3 weeks to an area with known H5N1 circulationHospitalized for SARIMeets the WHO suspect, probable, or confirmed H5N1case definitionClose contact with WHO suspect, probable, or confirmed caseOccupational exposureSARI in a previously healthy individualConsumption of raw /undercooked poultry or wild bird productsHandling samples (animal or human) suspected of containing H5N1 virus in a laboratory or other settingAnswers1. False2. False3. True4. True5. True6. False7. True8. TrueQuestion 11: True or False: The proposed epidemiologic “trigger criteria” below could be used to prioritize SARI cases for immediate reporting and laboratory testing for Influenza A (H5N1)Instructor: Once all of the potential criteria are discussed, click to show the answers.Travel within last 3 weeks to an area with known H5N1 circulation (False)Hospitalized for SARI(False)Meets the WHO suspect, probable, or confirmed H5N1case definition (True)Close contact with WHO suspect, probable, or confirmed case (True)Occupational exposure (True)SARI in a previously healthy individual (False)Consumption of raw /undercooked poultry or wild bird products (True)Handling samples (animal or human) suspected of containing H5N1 virus in a laboratory or other setting (True)
56Chief Surveillance Officer Response “What if the next pandemic isn’t caused by Influenza A (H5N1), but some other respiratory pathogen that isn’t associated with poultry or wild bird exposure?“We should learn our lesson from SARS and design a system that can also detect a respiratory pathogen that is spreading between humans and causing severe disease.”The CSO agrees with these criteria but asks the question “what if the next pandemic isn’t caused by Influenza A (H5N1), but some other respiratory pathogen that isn’t associated with poultry or wild bird exposure?” She adds that “we should learn our lesson from SARS and design a system that can also detect a respiratory pathogen that is spreading between humans and causing severe disease.”
57Question 12Consider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.Clusters of 2 or more SARI cases occurring within 7-10 days of each other are suspicious under all circumstances EXCEPT:If they are in a familyIf they all have a social connectionIf they all ate cooked chickenIf they all have an occupational connectionAnswer: CConsider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.Clusters of 2 or more SARI cases occurring within 7-10 days of each other are suspicious under all circumstances EXCEPT:If they are in a familyIf they all have a social connectionIf they all ate cooked chickenIf they all have an occupational connectionAnswer: C
58Question 13Consider the following series of questions about potential epidemiologic trigger criteria…SARI in health care workers who care for patients with ______.PneumoniaChronic respiratory diseasePoultry exposurePrevious hospitalizationAnswer: a. PneumoniaConsider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.SARI in health care workers who care for patients with ______.PneumoniaChronic respiratory diseasePoultry exposurePrevious hospitalizationAnswer: a. Pneumonia
59Question 14Consider the following series of questions about potential epidemiologic trigger criteria…Changes in the _________of SARI cases such as a shift in the age group affected or changes in mortality ratesseverityrecommended treatmentepidemiologyclinical presentationAnswer: e. epidemiologyConsider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.Changes in the _________of SARI cases such as a shift in the age group affected or changes in mortality ratesseverityrecommended treatmentepidemiologyclinical presentationAnswer: e. epidemiology
60Question 15Consider the following series of questions about potential epidemiologic trigger criteria…Any unexplained death due to SARI in persons _____________.aged 5-40aged < 5without underlying medical conditionsin countries with known circulation of possible pandemic respiratory virusesAnswer: a.Consider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.Any unexplained death due to SARI in persons _____________.aged 5-40aged < 5without underlying medical conditionsin countries with known circulation of possible pandemic respiratory virusesAnswer: a. aged 5-40 years.
61Question 16 Answer: a. True Answer: a. True Consider the following series of questions about potential epidemiologic trigger criteria…An increase in the numbers of cases occurring in a facility compared to the same season in a previous year is considered a potential trigger for raising the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.TrueFalseAnswer: a. TrueConsider the following series of questions about epidemiologic trigger criteria that might raise the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.An increase in the numbers of SARI cases occurring in a facility compared to an established baseline is considered a potential trigger for raising the index of suspicion about whether a respiratory pathogen of pandemic potential could be circulating in the population.TrueFalseAnswer: a. True
62Questions 12-16: Key Points Surveillance for CLUSTERS of SARI is criticalEven for an influenza A(H5N1) pandemic, most cases would not have a poultry linkSee Trigger Criteria Summary HandoutKEY POINTS: Surveillance for CLUSTERS of SARI is also a critical component of detecting a new respiratory pathogen with pandemic potential. Even if the next pandemic were to be caused by Avian Influenza A (H5N1) it is quite possible that most cases would not have epidemiologic links to sick or dying poultry.Instructor: at this point pass out the Case Study 1 Early Warning Trigger Criteria Handout. This will serve as a summary of the trigger criteria discussion, for student reference.
63Trigger Criteria Decisions National clinician education about trigger criteria and reportingAt hospitals within the sentinel systemAt hospitals outside the sentinel systemCases meeting criteriaImmediate notification to Provincial Medical Officer via toll-free phone numberOropharyngeal and nasopharyngeal swabsTogether you decide that there will be a national program to educate clinicians about these “trigger criteria”. It is important to note that clinicians who work at hospitals that are included within sentinel surveillance system as well as those who work at hospitals that are not in the sentinel surveillance system will receive sensitization training on these trigger criteria. Cases that meet these criteria will require immediate notification to the Provincial Medical Officer (who will investigate, and if necessary, request deployment of a regional Rapid Response Team). Oropharyngeal and nasopharyngeal swabs will also be collected and tested with a full panel of influenza diagnostics at the national laboratory. The regional Rapid Response Teams will make use of specimen collection and transport infrastructure located in the sentinel site hospitals in order to assure timely transport and testing of specimens. Clinicians in Pegu will also receive training on the mechanism for reporting cases meeting these trigger criteria. This reporting system will make use of a toll-free telephone number for reporting cases to the office of the Provincial Medical Officer.
64Finalize the Guidelines Trigger cases can facilitate timely diagnosis of other respiratory pathogens of pandemic potential, if negative for influenzaDetailed laboratory testing algorithm is plannedWHO case definitions used for international reporting purposesYou recognize that these “trigger cases” will be initially be tested for influenza in the laboratory. However you also feel that the surveillance system could facilitate timely diagnostic testing of other respiratory pathogens of pandemic potential if the initial laboratory diagnostics for influenza A, B, A/H1, A/H3, and A/H5 are negative. The CSO makes a plan to develop a more detailed laboratory testing algorithm for this surveillance system following meetings with the director of the national laboratory.Finally, you add to the guidelines that the WHO case definitions must still be still used for standard international reporting purposes (under the International Health Regulations, 2005, all persons meeting the probable or confirmed WHO case criteria must be immediately reported to WHO).
65Laboratory Samples H5-positive poultry specimens Province is rural From southeastern province, Pelu JaghaiSent to WHO reference laboratoryProvince is ruralHospital care may not be soughtCommunity-level surveillance neededThe next morning you are taken on a tour of the Pegu national laboratory. While at the laboratory you learn that the poultry specimens that tested positive for H5 by PCR were all from a southeastern province named Pelu Jaghai. Specimens were also sent to a nearby WHO reference laboratory. However you recall that Pelu Jaghai is a very rural area, and most of the population in that region does not regularly go to hospitals for urgent health needs. The CSO worries that a person sick with influenza A(H5N1) infection might not go to the hospital in Pelu Jaghai, and additional community-level surveillance is needed.
66Question 17How could Influenza A (H5N1) and pandemic early warning surveillance be expanded beyond the hospitals in Pegu? Match the method on the left with its description on the right.MethodDescriptionEnhanced passive surveillancea. Passive identification from media reports, the public, professional groups, and the WHO surveillance networkCommunication with the general populationb. Include these facilities in education and awareness training to report trigger criteriaInvolvement of traditional medicine hospitalsc. Outreach to health care gatekeepers - health care providers, laboratorians, drug dispensers, traditional healers, religious leaders, and others.Rumor surveillanced. Public service messages in print, radio, and televisionQuestion 17: How could Influenza A (H5N1) and pandemic early warning surveillance be expanded into the communities in Pegu? Match the method on the left with its description on the right.Methods:Enhanced passive surveillanceCommunication with the general publicInvolvement of traditional medicine hospitalsRumor surveillanceInstructor: Click to show answers.Answers:Method: enhanced passive surveillance. Description: C.Discussion points: This implementation of enhanced passive surveillance might also consider outreach to, diagnostic laboratorians, coroners, local veterinarians, and drug dispensers. And it may also include others with potential awareness of individuals with severe or atypical respiratory infections such as traditional healers, religious leaders, or unlicensed medical practitioners. An efficient reporting system will make use of existing public health infrastructure such as village volunteers who have relationships with surrounding families and are aware when anyone in this area becomes ill. Thus, education on the recognition and reporting of trigger cases/clusters should target any important gatekeepers to health care in the country.Method: Communication with the general population. Description: D.Discussion points: Messages can be targeted to the general population via various media including print, radio and television. These can improve awareness of severe respiratory infections that also meet trigger criteria. These public service messages should address general disease information including risk factors for infection, methods to reduce individual risk, trigger events that should be reported to public or animal health authorities, and a suggested mechanism for reporting.Method: involve traditional medicine hospitals. Description: B.Discussion points: Given that there are 12 traditional medicine hospitals in Pegu, these probably should be included in education and awareness training to report trigger criteria.Method: Rumor surveillance. Description: A.Discussion points: Rumor surveillance is a passive process, where rumors are identified from media reports, professional groups, the public, and persons in the WHO network (which is made up of WHO headquarters, country offices, and WHO Collaborating Centers). In an enhanced system, rumor surveillance is intensified by actively seeking out rumors and undertaking more rigorous follow up. This surveillance includes analyzing more media sources and regularly requesting information from the WHO network about outbreak events. The importance of rumor surveillance is likely to increase as the international community considers the revised draft of the International Health Regulations (IHR). Article 8 of the IHR Working Paper states, "WHO, in consultation with the health administration of the State concerned, shall verify rumors of public health risks which may involve or result in international spread of disease." An important part of rumor surveillance is the timely dissemination of accurate information to reduce misunderstanding and unwarranted concern, especially for rumors reported in the media. One example was the need to address the international concern that arose about the rumor that pigs were infected with avian influenza. If the rumor had not been reported to be incorrect publicly after the verification process, health authorities may have misdirected limited surveillance resources.
67Pegu AccomplishmentsMOH will train a team in each province, using polio surveillance officer as “bird flu person”Surveillance foundations in placeCase definitionsWHO reportingEarly warning system plansTrigger criteria for laboratory testing and public health investigationGatekeeper trainingSentinel sitesThe MOH indicates that they would like to use some available World Bank Funds to have one Rapid Response Team trained in each province by the end of the year, with the provincial AFP/Polio surveillance officer also becoming the principle surveillance officer for influenza A(H5N1). He will be known as the provincial “bird flu person” and T-shirts will be printed.You spend the next few days formally revising the surveillance guidelines with your MOH partners. You also provide several short trainings to the surveillance “bird flu persons” on the detection and reporting of trigger cases and clusters, appropriate infection control practices, data analysis and outbreak investigation.As your time in Pegu draws to its close, you pack your bags for the airport feeling like you accomplished a great deal during your short visit. You feel that the foundation has been put in place for a routine seasonal influenza sentinel surveillance system that includes case definitions and specimen collection protocols for SARI and ILI. The MOH has committed to sending isolates from confirmed influenza A positive cases to the WHO collaborating centers. You also feel that plans are well underway to implement the key components of a pandemic early warning system. Your colleagues have formally defined trigger criteria for laboratory specimen collection and public health investigation. The ongoing education and awareness trainings of hospital-based clinicians will help assure that trigger cases will be reported by clinicians in hospitals internal and external to the sentinel system. The training of additional health care “gatekeepers” and key members of the community using village health volunteers will further enhance the sensitivity of the pandemic early warning system. The establishment of the proposed system will have to occur in stages, and likely require several more visits in the future, but you are off to a good start.
68The next day… Pelu Jaghai reports another large poultry die-off In farms in backyard populationsSpecimens from Ministry of Agriculture sent to the National LaboratoryMinistry of Health wants to establish active human surveillanceHowever the next morning, the Deputy Director of the MOH calls you and tells you that Pelu Jaghai is reportedly experiencing yet another large poultry die-off in farms and backyard populations. Specimens have been collected by the Ministry of Agriculture and sent to the National Laboratory for testing. The MOH would like his district staff to establish active surveillance for human cases of Influenza A (H5N1) in the affected region. He asks you for advice.
69Question 18Which of the following are surveillance enhancements for human disease that could be recommended for the affected province?Teach Pegu’s traditional healing methods to it regional epidemiologistsDoor to door surveillance for ill people and chickensInitiate school-wide influenza shotsSARI surveillance among healthcare workers at local facilitiesActive case finding among the occupationally exposedDismiss rumors of clusters within health care workers, families or village contactsRecruit private practices, NGO’s, religious institutions, and schools into the surveillance system for H5N1 and pandemic trigger criteriaRefresher training on reporting proceduresConfirm availability of telephone reporting hotlinesQuestion 18: Which of the following are surveillance enhancements for human disease that could be recommended for the affected province?Teach Pegu’s traditional healing methods to it regional epidemiologistsDoor to door surveillance for ill people and chickensInitiate school-wide influenza shotsSARI surveillance among healthcare workers at local healthcare facilitiesActive case finding among the occupationally exposedDismiss rumors of clusters within health care workers, families or village contactsRecruit private practices, NGO’s, religious institutions, and schools into the surveillance system for H5N1 and pandemic trigger criteriaRefresher training on reporting proceduresConfirm availability of telephone reporting hotlinesInstructor: Click to see correct answers, also listed belowDoor-to-door surveys for ill people and chickensSARI surveillance among health care workersAdditional active case finding among the occupationally exposed, including those personnel involved in the veterinary response to the outbreakRecruit private practices, NGO, religious institutions, and schools into the surveillance systemInclude refresher training on reporting procedures. The reporting mechanism (e.g. toll free number) must be clear!Telephone reporting hotlinesDiscussion points: Some additional methods not listed include:Sensitization of community to report illness (including risk reduction messages)Temporarily expand/enhance SARI surveillance to local hospitals, traditional healers and dispensaries.Active daily networking with village health monitors
70Question 18: Key Points Active surveillance is key in this context Make surveillance more active in the hospital and community settingsBackyard poultry husbandry may not be known to authoritiesPoultry workers and community need to know importance of seeking treatmentMake healthcare facilities aware of community education and reporting mechanismsVillage health monitors and leaders can be important sources of information for outside investigatorsKEY POINTS: Active nature of surveillance in this context should be emphasized. Consider what must be done to make surveillance more active in both the hospital and community setting. This would include an active search in poultry workers and door-to-door surveys in the surrounding community. In much of the world, most poultry husbandry is done in backyards and may not be known to the authorities. Poultry workers and the community need to be made aware of the importance of seeking treatment if they become ill. Instruct workers to be vigilant for the development of fever and respiratory symptoms for 1 week after last exposure to avian influenza-infected or exposed birds or to potentially avian influenza-contaminated environmental surfaces. Health care facilities should be aware that this education activity is occurring and know whom to notify if they detect a possible case. Facilities should implement screening / triage / reporting procedures for patients presenting with influenza-like illness. Consider adding non-public facilities delivering healthcare to the reporting network including private practices, NGO and religious institutions. Expand SARI and/or trigger surveillance to local hospitals, traditional healers and dispensaries and train all in the reporting network on procedures (forms, time for reporting, where, etc.). Village health monitors and leaders can be important sources of information for outside investigators. If the technology is available, telephone hotlines can be useful for rapid reporting of suspect events.
71Later that evening…You learn of two possible human cases in neighboring Dava Ghar provinceNo poultry outbreaks have ever been reported here!You are asked to extend your stay and participate in the outbreak investigationYou travel with the District Epidemiologist and two local officials to Dava Ghar ProvinceThat evening you additionally also learn of two possible human cases in neighboring Dava Ghar province. Surprisingly, no poultry outbreaks have ever been reported in that province! Despite being homesick you are convinced to extend your stay and participate in the outbreak investigation. With your heart beating rapidly you travel with the District Epidemiologist and two local officials to Dava Ghar Province.