8.4 10 Key Steps in Outbreak Investigation

The ten steps are:

  1. Confirming an outbreak
  2. Establishing an outbreak team and developing case definition(s)
  3. Systematic case finding and data collection
  4. Descriptive epidemiology
  5. Hypothesis generation
  6. Hypothesis testing
  7. Microbiological investigations or environmental studies
  8. Control and prevention measures
  9. Surveillance
  10. Communication

8.4.1 Confirming an outbreak

Recall that an outbreak / epidemic happens when the number of cases of a disease is more than expected among a group of people in a given area over time.

Furthermore, a cluster is merely a collection of cases over a particular period of time. Whether the number of cases of a disease is greater than expected is irrelevant.

Nevertheless, in this step, one has to determine whether or not a cluster of disease cases is indeed an outbreak or merely sporadic cases.

8.4.1.1 How do we confirm outbreaks?

There are two ways listed on the lecture slides:

  1. Evaluating surveillance data

    Consequently, this may lead to a change in surveillance methods, a change in case definitions in reporting, and even increased awareness among the general and the scientific community. Better or even novel laboratory tests can be implemented.

  2. Case diagnoses

    These are based on clinical signs and symptoms that are common across infected individuals. Laboratory tests can also help in this setting (notwithstanding reporting errors due to humans).

8.4.2 Establishing an outbreak team and developing case definitions

An outbreak team typically has the following members:

  1. Team leader
  2. Epidemiologist
  3. Laboratory expert
  4. Data manager
  5. Doctor
  6. Interviewer
  7. Communications expert
  8. Health authority
  9. Entomologist
  10. Environmental health specialist

Furthermore, the CDC defines a case definition as:

“A case definition is a standard set of criteria for deciding whether an individual should be classified as having the health condition of interest.”

– CDC themselves

A case definition also encompasses the following aspects:

  1. Clinical criteria
  2. Time restrictions
  3. Place restrictions
  4. Person restrictions

A case definition can also be classified into one of three categories:

  1. Confirmed (i.e., confirmed by laboratory methods)
  2. Probable (i.e., clinical features are present, but there is no laboratory confirmation)
  3. Possible or suspected cases (i.e., for cases that have less clinical features)

However, it musn’t include exposure or risk factors that are suspected to have caused an outbreak.

8.4.3 Systematic case finding and data collection

There are two aspects in this step that are worth mentioning:

  1. Systematic case finding

    Here, increasing reports of small cases (either via passive or active surveillance) of a disease eventually lead to the discovery of an outbreak. However, these small cases are not necessarily representative of all cases (i.e., bias is present).

    Public announcements may also be carried out; case patients may also be interviewed and the entire population may be surveyed.

  2. Data collection form

    Here, we identify one’s personal contact information (i.e., name, address, telephone number, etc). We also ask for demographic information (e.g., sex, race, occupation, etc).

    Such information can also lead to the identification of risk factors.

8.4.4 Descriptive epidemiology

Descriptive epidemiology describes information in terms of people over time.

The objectives here are critical, including but not limited to:

  1. Identifying key demographics at risk for infection
  2. Identifying at-risk populations
  3. Identifying possible source(s) and modes of transmissions
  4. Coming up with control measures
  5. Identifying and correcting errors (i.e., repeated analyses of disease data).

8.4.4.1 Epidemic curves

An epidemic curve is a diagram that displays the magnitude of an epidemic over time. The shape of the curve, the course of the outbreak (i.e., does the curve have a beginning, a middle, and / or an end), and any outliers.

8.4.4.1.1 Drawing an epidemic curve

An epidemic curve must have the following to be plotted:

  1. Time (x-axis)
  2. Number of cases (y-axis)
  3. Date of disease onset
  4. Time of disease onset (if the disease has a short incubation period)
8.4.4.1.2 Interpreting an epidemic curve
Different Shapes of Epidemic Curves

Figure 8.1: Different Shapes of Epidemic Curves

There are two shapes to be aware of:

  1. Point-source epidemic

    These have a log-normal curve (i.e., steep upslope and a gradual downslope). Average and minimum incubation periods are used to inform the likely period of exposure.

    Otherwise, the time of exposure and the time of onset can be used to estimate the incubation period.

  2. Continuous common-source epidemic

    The shape of this graph is plateau-shaped - the duration of exposure is prolonged.

Outliers in the data may be due to errors in data; however, outliers may also provide data on easily identifiable exposures.

8.4.4.1.3 Places in descriptive epidemiology

Spot maps are a simple and useful technique - these maps also help illustrate where cases live, work, and may have been exposed.

Furthermore, spot maps may also identify who is at risk and who are the case patients (along with the geographic extent of the problem).

8.4.5 Hypothesis generation

This begins as soon as data is collected. The hypothesis may involve the following elements:

  1. Source of agent
  2. Modes of transmission
  3. Exposure or risk factors that cause disease

Some methods include for hypothesis generation include:

  1. Talking to initial cases
  2. Visiting patients’ homes
  3. Gathering information from descriptive epidemiology

8.4.5.1 Bioterrorism

Some epidemiological clues of bioterrorism include:

  1. Unusual cases of a disease that have been transmitted by an uncommon source.
  2. Unusual strains of a pathogen.
  3. High morbidity and mortality of a disease.
  4. Unusual disease presentations.
  5. Disease with unusual geographic distributions.
  6. Disease patterns that have been linked to ventilation systems.

There are also more features to note:

  1. A large number of unexplained death / disease.
  2. Compressed point source epidemic curves.
  3. Several unusual / unexplained diseases that coexist in the same patient without other explanations.

8.4.6 Hypothesis testing

Environmental evidence, laboratory data, and epidemiology can be used to test hypotheses.

For instance, one can define a null hypothtesis H0 and an alternate hypothesis H1 along with a few controls, statistical analyses, and also implement study designs:

  1. Retrospective study design

    This is most suitable for an outbreak in a small, well-defined population over a period of time.

  2. Case-control study

    This is most appropriate when a population is not well-defined and when the speed of investigation is crucial.

8.4.6.1 Choosing controls

Control subjects must not have the disease - furthermore, they must also be representative of the disease population.

Controls can be recruited via random telephone calling, neighbors of cases, friends of cases, and even patients from the same healthcare facility with no overt symptoms.

A maximum of four controls is usually adopted per case - controls should also not be used to test hypotheses!

8.4.7 Microbiological investigations or environmental studies

Laboratory findings and environmental studies are crucial for confirming epidemiological findings and are also important for certain pathogens and settings.

Medication and medical equipment sampling may also need to be considered.

8.4.8 Control and prevention measures

Confidentilality in these measures should be implemented as early as possible - not only is this the primary aim of outbreak investigations, but it should also be implemented as soon as possible.

Control measures may target:

  1. Sources / agents
  2. Modes of transmissions
  3. Portal of entries (e.g., bed nets, etc.)
  4. Hosts (e.g., vaccinations, etc.)

8.4.9 Surveillance

Active surveillance should be continued to encourage case finding. The surveillance of post-control measures monitors:

  1. Effectiveness of control measures
  2. Geographical areas that have been affected by a disease

8.4.10 Communication

Communications plans may entail communication within a team and with the affected and the general public (including the media).

Outbreak reports also summarize the findings of investigations and also help send information to the relevant parties.