4 Outbreak Response
Outbreak
A higher number of cases of a disease above what is normally expected in a specific population and time period.
The number of cases and the time period used to define an outbreak depend on the characteristics of the disease and how it spreads within the population under surveillance. For some diseases, detection of one case could be considered an outbreak because the disease is so rare, severe or contagious that we want to prevent any cases from occurring (e.g., poliomyelitis). Often an outbreak might be indicated a large number of cases occurring over a short time period, but some outbreaks occur slowly over a longer period of time because of the time it takes for transmission to occur or symptoms to develop (e.g., syphilis).
Outbreak steps
The following table shows eight outbreak investigation steps based on the steps outlined in the Pacific Outbreak Manual. Some countries or resources describe outbreak investigations using more or different steps.
In practice, the steps will usually not occur in order. Some steps may happen at the same time or in a different order, and it may be necessary to move back and forth between steps as new information is available. Even though implementing control measures is one of the last steps described, it is often necessary to start some interim control measures at the beginning of the investigation and to update these measures (including withdrawing them if appropriate) as more is understood about what is occurring.
Outbreak investigations require the input of people with different skills and resources. It is best to form an outbreak investigation team as early as possible to ensure that all relevant people who are able to contribute to the investigation are included. This ensures the outbreak can be controlled effectively and quickly. The number and types of people that should be involved in the team will depend on the size and location of the outbreak and what is causing the outbreak.
Outbreak Step |
Overview of step |
|---|---|
|
The first step of an investigation is to find out if an outbreak is actually occurring. A higher number of cases than usual may not always mean that there is an outbreak, so it is important to consider the information that is available to determine if further investigation is needed.
Questions to consider to verify an outbreak:
|
|
|
|
A standard outbreak case definition
Need to find a balance between being
A case definition should include elements of:
May have different categories based on level of certainty
May classify cases based on exposure
Case definition may change over time as new information is learned in the outbreak |
|
Actively find additional cases
Collect information
|
|
|
|
From your linelist, you can start to analyse the data (e.g., describe person, place, time) and generate ideas about what is causing it (why, how). You can analyse and interpret the data in several ways:
Time
Place
Person
After analysing the data, you now need to interpret it and to generate ideas as to what is happening (hypothesis generation). For each of the following questions, it is important to consider whether the hypothesis make sense/is plausible.
You might decide that you need further information to better understand the disease outbreak and decide on the right actions to take to control the spread.
|
|
Act early to minimise impact and control the risk to other people in the community; do not wait until the end of the investigation. Control measures are often implemented at the same time as the above steps.
Use general measures initially (e.g., general health messaging, alerting the community that the outbreak is occurring) and adapt these according to new information and how the outbreak evolves. Measures may include:
|
|
It is important that you communicate clearly and regularly throughout the investigation. You should communicate with everyone who is involved in the investigation so that good decisions can be made, people do not duplicate tasks, and things do not get missed.
You should also communicate to the community to keep people informed of what is happening, what the risks are, how to avoid getting sick, and what actions are being taken. The way that you communicate findings will depend on your target audience. Communication with the investigation team and health staff
Communication with the publicTimely and targeted public communication can:
Public risk communication should always:
Common mistakes in public risk communication:
|
At the end of an outbreak
An outbreak may be closed when there is no longer any risk to the public, when enough information has been gathered to understand what occurred and/or no further actions can be taken to control the spread. Decisions to close an outbreak should be based on the evidence gathered, analysis undertaken and advice of experts involved in the investigation. If you know the pathogen that caused the outbreak, it is common to wait until at least two incubation periods for that pathogen have passed without new cases being found before closing an investigation.
Example:
You have been investigating an outbreak of diarrhoea among people living in village X. Your analysis suggests that a sewerage leak in the village caused the infections. The leak has now been fixed and despite actively looking, no further cases have been found for two weeks. After speaking with the investigation team, the investigation team leader decides that the investigation can be closed.
After an investigation has been closed, it is important to reflect on how the investigation went so that improvements can be made for future investigations. A review should be undertaken to identify positive aspects of the investigation that should be done again and the things that can be avoided or improved next time. Outbreak reviews can also help to improve routine surveillance and how it is used to identify outbreaks.
The type of review undertaken will depend on the size, length and seriousness of the outbreak. Reviews can range from a reflection in an outbreak report to a debrief with the investigation team to a full After Action Review where feedback is gathered from all stakeholders through different formats. Whichever method is chosen, it is important to document the conclusions of the review so that they can be incorporated into future work. More information on how to conduct an after action review can be found in WHO’s Guidance for after action review.
Some of the things that may be considered in a review are:
- Could the outbreak have been identified sooner?
- Were the right people involved in the investigation team?
- Was communication clear and timely?
- Were decisions made based on evidence?
- Were decisions and actions recorded?
- Was the right public health action taken and was it implemented at the right time?
Common Source Outbreak
A common-source outbreak is one where people are exposed to an infectious agent from the same source.
There are three types of common source outbreaks:
Point source outbreak
A type of common source outbreak where people are exposed over a relatively short period of time, so that all cases occur within one incubation period for the pathogen that caused the outbreak. This occurs when the infectious agent was only present for a defined period of time, such as at an event attended by the cases.
Example:
Contaminated food source at a dinner party with 100 guests and 65 guests become cases.
An epicurve in a point source outbreak will have a sharp increase in cases followed by a slower decrease over time.
Continuous common source outbreak
A type of common source outbreak where people are exposed to the source over time, with cases occurring longer than the time of one incubation period.
Example:
Contaminated water source to a village where access is not stopped, and people continue to drink the contaminated water.
An epicurve in a continuous common source outbreak will have multiple peaks and declines. There are cases as long as people remain exposed to the common source.
Intermittent common source outbreak
A type of common source outbreak where people are exposed to an infectious agent at irregular time periods; sometimes they are exposed, and sometimes not.
Example:
If a restaurant has five food handlers and one is positive for Hepatitis A, we would expect the number of Hepatitis A cases to have eaten at the restaurant only when the infected food handler was working.
An epicurve in an intermittent common source outbreak will show rapid increases and decreases, reflecting the presence of the source. The length of time between peaks is not related to incubation period, but is instead related to times when the source is present.
Propagated outbreak
An outbreak that does not have a common source, but instead spreads from person to person.
Example:
Measles, where each infected person can infect 12- 18 other non-immune people, who can then each infect up to another 12-18 other non-immune people.
An epicurve in a propagated outbreak usually has a number of irregular peaks reflecting the generations of infection, with peaks separated by approximately one incubation period.
Cluster
A cluster is group of cases with a similar disease or set of symptoms reported over a specified time and in a defined area that is suspected to be higher than expected. The word ‘cluster’ is often used to describe a situation prior to a formal outbreak being declared.
Example:
Syndromic surveillance may identify a cluster of influenza-like-illness (ILI) cases in a particular village.
Contact
Someone who has been exposed to a source of an infection or pathogen.
Example:
A person who has spent time with a confirmed COVID-19 case is known as a contact. It means that they were exposed to a confirmed case while the case was infectious. They are not currently a case but have the potential to become a case.
Linelist
A linelist is a way of organising data into a table where each ‘row’ contains information about one case.
The variables (e.g., Age, sex, clinical characteristics) about each case are in the columns. It is important there is only one row per person and one column per variable (i.e., do not combine multiple symptoms in one column as this cannot be easily analysed). Linelists are commonly created in Excel, but can also be done on paper. A linelist makes it easy to analyse and interpret your data.
An example of a linelist is below.
| ID | First Name | Last Name | Sex | Age (Years) | Village | Province | Date of illness onset | Nausea | Vomiting | Laboratory confirmation |
|---|---|---|---|---|---|---|---|---|---|---|
| 1 | Name | Name | M | 36 | Village X | Province A | 4/12/2022 | Yes | Yes | No |
| 2 | Name | Name | M | 42 | Village X | Province A | 4/12/2022 | Yes | Yes | Yes |
| 3 | Name | Name | M | 63 | Village Y | Province A | 2/12/2022 | Yes | Yes | No |
| 4 | Name | Name | F | 38 | Village Y | Province A | 1/12/2022 | No | Yes | Yes |
| 5 | Name | Name | F | 25 | Village X | Province A | 5/12/2022 | Yes | Yes | No |
| 6 | Name | Name | M | 34 | Village Y | Province A | 7/12/2022 | Yes | No | Yes |
| 7 | Name | Name | F | 43 | Village Y | Province A | 2/12/2022 | No | Yes | No |
| 8 | Name | Name | M | 52 | Village X | Province A | 3/12/2022 | Yes | Yes | Yes |
| 9 | Name | Name | F | 51 | Village X | Province A | 5/12/2022 | No | No | No |
| 10 | Name | Name | M | 40 | Village Y | Province A | 4/12/2022 | Yes | Yes | No |
Endemic
The constant presence and spread of a disease or infectious agent within a given geographic area or population group in a way that is expected.
Example:
Malaria is endemic in some Pacific Island nations, in parts of Africa, parts of Asia, and parts of South and Central America.
Epidemic
The occurrence of more cases of disease than expected in a given area or among a specific group of people over a particular period of time. The increase is often higher than an outbreak or over a longer time period.
Example:
Whilst dengue fever can be endemic to a country, there may be a dengue fever epidemic when cases increase over a long period of time in a particular area.
Pandemic
An epidemic occurring worldwide, or over a very wide area, crossing international boundaries (multiple continents) and usually affecting a large number of people.
Example:
COVID-19 was declared a pandemic because the SARS-CoV-2 virus was a new virus spreading between large numbers of people on multiple continents.
Sensitivity and specificity
Sensitivity and specificity are measures of how accurately a person is identified as either having or not having a disease (or health condition under investigation). The principles of sensitivity and specificity apply both to diagnostic tests that identify if a person is infected with a pathogen and more broadly to case definitions that establish the criteria for a person to be considered as having a disease.
Sensitivity
Sensitivity is how likely it is that a person that truly has a condition will be correctly classified as having the condition (true positive). High sensitivity means that as many true cases will be counted as possible and there will be less cases missed. This is important for serious diseases that require an immediate response or in the early stages of an outbreak when you do not yet have much information about what is causing the outbreak and it is important to identify as many cases as possible.
Example:
- Sensitivity of a laboratory test
- A blood test with 95% sensitivity means that the test will detect the pathogen in 95% of people who are infected. The other 5% will be incorrectly classified as not being infected.
- A sensitive case definition is one that is broad to try to include as many cases as possible:
- Three of more loose stools in a 24-hour period
AND
Resident or visitor of Province B
AND
Symptom onset on or after January 15, 2022.
- Three of more loose stools in a 24-hour period
Specificity
Specificity is how likely it is that a person who does not have a condition will be correctly classified as not having the condition (true negative). High specificity means that less people will be incorrectly classified as having the condition when they do not have it. This is important for conditions that might cause serious distress or harm to a person if incorrectly diagnosed as having the condition (such as one with stigma, or a fatal disease that cannot be treated).
Outbreak case definitions are usually changed to be more specific during an investigation when more information is available about what is causing the outbreak and who is affected.
Example:
- Specificity of a laboratory test
- A PCR test that is 90% specific means that a pathogen will not be detected in 90% of people who are not infected. The test will incorrectly find that the pathogen is present in the remaining 10% of people who are not infected.
- A specific case definition is one that is narrow to be certain that someone counted as a case really has the disease:
- Detection of E.coli in a clinical specimen
AND
Resident or visitor of Village C
AND
Symptom onset on or after January 15, 2022.
- Detection of E.coli in a clinical specimen
Sensitivity and specificity usually need to be balanced because having high sensitivity often reduces specificity and the other way around. This is particularly true when deciding on a case definition. A sensitive (broad) case definition will likely count some people as cases when they do not have the disease. A specific (narrow) case definition may miss counting some people who have the disease. Each condition or outbreak will have different priorities for sensitivity and specificity and these priorities can change over time.
