Author: Lauren Bishop
Morbidity, which refers to disease, injury, and disability, broadly indicates a departure from the state of wellbeing, whether physiological or psychological.
Incidence and prevalence are the key measures of morbidity frequency that aim to quantify population health in terms of health events and health status, respectively.
| Incidence | The occurrence of a health event within an at-risk population during a specific time interval. |
| Prevalence | The proportion of health events (new and preexisting events) in an at-risk population at a specific timepoint or within a specific time interval. |
A population at risk refers to the individuals who are susceptible to the health event.
| Example Only those with cervices can logically be at risk for cervical cancer. Therefore, the at-risk population for cervical cancer might consist of women ages 20+ (the age restriction is imposed here since it is extremely rare for women to be diagnosed with cervical cancer before the age of 20). |
Incidence refers to the proportion of at-risk individuals who develop the health event in question, whereas prevalence refers to the proportion of individuals who, at that point in time, have the health event in question.
The two concepts are related, but that relationship may look different depending on which health event you are studying.
| Example For type II diabetes, there may be low incidence (few people develop type II diabetes) but a high prevalence (many people already have type II diabetes) during the observation period. Conversely, for a disease like seasonal influenza, the incidence may be very high (many people develop the flu), but the prevalence may be low (people do not usually have the flu for very long). |
| Note Depending on the epidemiologist, incidence may refer to either the number of new events in a specific population or the number of new events per unit of a specific population. |
We will now look at measures of incidence and prevalence in a bit more detail. Incidence is often measured using the incidence proportion or the incidence rate.
Incidence proportion
The incidence proportion refers to the risk of occurrence of a health event (e.g. developing a disease, becoming injured or disabled) during a specific time interval among those who are at risk at the beginning of the time interval.
Also sometimes referred to as cumulative incidence, the incidence proportion is calculated by dividing the number of new events within the at-risk population (numerator) by the entire at-risk population (denominator).
Example
| Example |
 |  |
| 860,000 | 70,000,000 |
| Let us jump into a time machine and return to the 14th century to examine the incidence proportion of bubonic plague (the black death, caused by Yersinia Pestis) in Europe. If the at-risk population in the year 1347 was 70,000,000 and, of those, 860,000 individuals developed bubonic plague during the month under observation, what is the incidence proportion? The incidence proportion is the number of at-risk individuals who developed bubonic plague divided by the total population at risk during that specific month: 860,000/70,000,000 = 0.01 = 1.2% |
Incidence rate
The incidence rate has the same numerator as the incidence proportion: the number of new events within the at-risk population. However, the denominator is different.
For the incidence rate, the denominator is the time each at-risk individual was observed, summed for all at-risk individuals.
In other words, the denominator is the total time the population was at risk for the health event during the time interval under study. Incidence rates are often used to measure the speed at which a health event is distributed within a population.
| Example |
 |  |
| 15 | 100 |
| In a magical world where all individuals in the study are followed for the same amount of time and none are lost to follow-up, we have a study population of 100 people, each of whom is followed for 10 years. Within this population, 15 are newly diagnosed with heart disease. We must first calculate the total time at risk. In this example, the unit of time at risk is measured in person-years. We have 100 people, each followed for 10 years. 100 people followed for 10 years = 1,000 person-years. Now that we have our denominator, we can calculate the incidence rate. Number of health events/time at risk = incidence rate. 15/1,000 person-years = 0.015 heart disease events per person-year. Since health events are often reported per 100,000 person-years, we could also report the incidence rate as 1,500 heart disease events per 100,000 person-years. |
| Note Measures of person-time will be discussed in much more detail in Cox regression. |
Prevalence
Prevalence can be calculated at a single point in time (point prevalence) or during a specific time interval (period prevalence).
Calculating prevalence can be quite straightforward: we divide the number of people who have the health event (numerator) by the number of people in the at-risk population (denominator) at a specific time or during a specific time interval.
| Example |
 |  |
| 18,800 | 2,000,000 |
| We have an at-risk population of 2,000,000 adults between ages 50 and older who live in Europe. Of those, 18,800 have been diagnosed with dementia. The prevalence of dementia within this population at this exact point in time is the number of at-risk people with dementia divided by the total population of persons at risk: 18,800/2,000,000 = 0.009 = 0.9% |
| Note Though prevalence is a relatively straightforward calculation, it is important to consider the factors that influence prevalence when interpreting your results, or results you read elsewhere. Prevalence may be influenced by, for example, how long the health event lasts (e.g., the duration individuals may live with a non-communicable versus a communicable disease), improved reporting, prolongation of life with the health event, or an increase or decrease in incidence of the health event.. |
| Note Morbidity may also be reported in terms of risk ratios and odds ratios, which are discussed in more detail in Risks and odds. |