Formal risk assessment is the process of systematic review of evidence that define or estimate a risk in the population. The purpose is to guide risk management (e.g. preventive and control measures).  The concept of 'risk' is here defined as the [XXX hazard/impact * probability of occurrence.... to be completed..]

The differences with Rapid Risk Assessments are:

  • the timeline in which the assessment takes place: instead of 24-28 hours, a formal risk assessment may take weeks or even months
  • the focus of the risk scenarios: instead of focus on the immediate problem, the focus includes possible future evolutions of the hazard under assessment
Health risks may arise in the population (emerging or newly identified diseases), or may already exist (yet with changing epidemiological pattern or changes in risk factors). It is important to be aware that risks do not remain static; formal risk assessments take development-scenarios into account.

Problem formulation

The scope of the assessment is the starting point. This could be the risk of introduction of a disease agent, or the risk of spread of a disease. It could cover threats to one sector (health) or many (e.g. agriculture, food, security). Once the scope is defined, the problem is formulated, with related objectives of the assessment (SMART). This will also clarify what sectors of society are affected by the problem.

Constituting a Formal Risk assessment group

Many hazards that are assessed are cross cutting through different sectors and disciplines in health. Therefore the team needs to reflect this multidisciplinary and multisectoral aspect. The group should be large enough to cover all areas, include representatives from health and other relevant sectors. Additional experts will be contacted to provide expert input.

Defining the probability of occurrence: an example

When defining the risk of introduction of West Nile Virus in a country, then a first issue to address is estimating the probability of introduction of the virus in the population. This includes studying migratory routes of birds that could carry the virus and the presence of vectors (Aedes albopictus) in the country. Specifically the interest will be in the frequency of occurrence of specific migratory birds in the country (this can be informed via veterinarians and wild life societies) and the presence and future spread of the mosquito (this can be informed by environmental specialists and entomologists and in addition specialists in climate change).

Socio-economic factors (e.g. agricultural development, technological development, movement of people) also need to be taken into account

Defining the impact

The scope of the risk assessment defines also the scope of the impact to describe: only health, or also other areas such as economics, travel, agriculture, security etc. The impact depends on various hazard factors.

Hazard factors for infectious agents are:

In addition, there will be several host factors that are relevant to describe:

Characteristics of the system are essential to describe: the structure and capacity of the health care system, laboratory capacity, treatment capacity, prevention and control capacity. For bloodborne diseases (such as WNV), it is important to describe the blood donation process, options and costs for screening (depending on scenario of establishment of the disease in the country)

Modeling the risks

Scenario tree modeling describes the chain of events that will lead to possible risks. This requires to describe a tree of events (each event step will be a relevant condition for the final risk). Then for each of these steps, the probability is assigned. Finally a sensitivity analysis is performed.

Mapping the risks

It is usually relevant for risk managers to have a visual representation of risks according to geographical region of a country (map). This will help to set priorities for risk reduction strategies. Risk maps can also aggregate information from different factors that are considered predictors of probability of an event.

Sources of data

Probability and impact need to be as much as possible fact and evidence based. Surveillance data (human, animal, environmental) are useful, as are specific surveys, published research. If information is missing from those sources, then expert opinion can be recruited to complement the required information.