Incidence refers to the number of new cases of a certain disease or event in a population over a specified period of time. There are 2 ways of measuring incidence: cumulative incidence (CI) and incidence rate (IR). Both are measures of the risk of contracting a disease or condition.
- Cumulative incidence per 1,000 = (Number of new cases of disease occurring in a population over a specified period of time) / (Number of total individuals in the population who are at risk for developing the disease during the specified period of time) x 1,000:
- CI is also known as cumulative incidence proportion, incidence proportion, risk, attack rate, or probability of developing the disease.
- CI is not a true rate but a proportion, as the people included in the numerator are also included in the denominator (total population)
- The denominator should include only “at-risk“ individuals. Individuals who already have the disease or are incapable of contracting the disease must be excluded from the population.
- The number used to multiply depends on the frequency of the disease. For example, 1,000 may be sufficient to use for a common disease, but 100,000 or 1,000,000 would be required for a rare disease. Using 100 as the multiplier makes the CI proportion a percentage.
- CI does NOT take into account:
- People who became lost to follow-up
- People who die from another cause (a “competing risk”) during the observation period
- When people developed the disease (i.e., at the beginning, middle, or late in the observation period)
- The synonym attack rate is commonly used during outbreaks of disease and in hypothetical predictions.
- Food-specific attack rate per 100 = (Number of susceptible or at-risk individuals who were exposed to a pathogen or suspect food who became ill) / (Number of total individuals who were exposed to that pathogen or food)
- Example: If 30 people developed gastroenteritis after eating potato salad, and 100 people in total ate the salad, then the AR = 30 / 100 x 100 = 30%, which is the risk of becoming ill after eating the salad.
- CI is used mostly used in situations where the follow-up time is relatively short and there is relatively little loss to follow-up. Otherwise, epidemiologists generally use the incidence rate.
- Incidence rate per 1,000 = (Number of new cases of disease occurring in the population during a specified period of time) / (Total person-time, or the sum of the time periods of observation of each person who was observed) x 1,000
- IR is a true rate and is also known as person-time rate or incidence density.
- Total person-time: based on the eligibility criteria for the study population that an individual must be at risk (disease-free) and actively followed. Once a person develops the disease or is lost to follow-up, their contribution of time at risk ends. If their contribution ends after 1 year, then that person has contributed 1 person-year (py) to the study. For example:
- 1 individual at risk who is in a 1-year study = 1 py
- 5 individuals at risk who are in a 5-year study = 25 py
- 5 individuals at risk who are in a 5-year study, with 2 being lost to follow-up or disease after 1 year while the remaining 3 continue to be at risk for the full 5 years = 2 py + (3 x 5) = 17 py
Over the past year at a particular hospital, 500,000 patients were admitted for hip or knee replacements. Five thousand of those patients over that year had a venous thromboembolism (VTE) during their admission. What is the risk of developing VTE over the past year at this hospital?
- There are 5,000 new cases of VTE over the year in a population at risk of 500,000.
- Therefore, 1% of those admitted, or 1 out of 100 patients, experienced VTE during their hospitalization.
Over the time period from 2000 to 2009, in Quebec, Canada, there were 91,000 cases of VTE over a population, with a disease-free time over that period that totals 74,200,000 py. What is the IR of VTE in Quebec over this time period?
Therefore, the IR is 0.0012 cases per py. By convention, this is multiplied by 10,000 to get 12 cases per 10,000 py of observation.
- Prevalence is the number of affected individuals present in a population at a particular point in time divided by the number of total individuals in the population at that point in time (point prevalence) or over a period of time (period prevalence). Prevalence is typically expressed as a proportion or percentage.
- Point prevalence is like a “snapshot” of the population that determines who has the disease and who does not; much more commonly used than period prevalence.
- Does not measure risk, but provides an important measure of the burden of disease in a community, which is necessary for resource allocation.
- Prevalence cannot provide a risk estimation, as it does not give information about when the disease developed. Some individuals may have developed the disease a week ago, some last year, and some 20 years ago.
- There is a relationship between prevalence and incidence (Prevalence = Incidence × Duration of disease) as long as:
- The population is in a steady state.
- The disease rate is constant.
- The prevalence is not too high.
Prevalence versus incidence
- In 2 hypothetical cities, with 1,000 inhabitants in each, there is a contagious disease with a prevalence of 100 / 1,000 in City A and a prevalence of 60 / 1,000 in City B.
- There is not enough information to determine which city would be “less risky” to live in without considering incidence.
- However, if City B has an incidence of 20 cases / year, while City A has only 4 cases / year, then City B is riskier.
- The prevalence in City B is lower because the average survival of a diseased person in City B is only 3 years, while a diseased person in City A lives an average of 25 years.
- For City A: 4 / year x 25 years = 100 / 1,000
- For City B: 20 / year x 3 years = 60 / 1,000
Over the past year, there were 12 million patients in the Medicare database, and 200,000 Medicare patients suffered a VTE. What is the period prevalence of VTE in Medicare patients in the United States over the past year?
- The period prevalence for the year is a measure of the proportion of Medicare patients who had VTE. There were 12 million patients in the Medicare database; 200,000 of these patients had VTE.
- Therefore, the period prevalence of VTE in Medicare patients for the year was 1.6%.
- Greenberg, R. S. (2014). Epidemiologic measures. R. S. Greenberg (Ed.), Medical Epidemiology: Population health and effective health care, 5e (). New York, NY: McGraw-Hill Education. accessmedicine.mhmedical.com/content.aspx?aid=1108589766
- Thacker, S. B. et al. (2006). Measuring the public’s health. Public health reports (Washington, D.C.: 1974), 121(1), 14–22. DOI:10.1177/003335490612100107
- Celentano, David D., ScD., M.H.S., & Szklo, Moyses, MD, M.P.H., DrP.H. (2019). The occurrence of disease: I. disease surveillance and measures of morbidity. Celentano, David D., ScD, MHS, & Szklo, Moyses, MD, MPH,DrPH (Eds.), Gordis epidemiology(pp. 41-64) https://www.clinicalkey.es/#!/content/3-s2.0-B9780323552295000036
- Celentano, David D., ScD., M.H.S., & Szklo, Moyses, MD, M.P.H., DrP.H. (2019). The occurrence of disease: II. mortality and other measures of disease impact. Celentano, David D., ScD, MHS, & Szklo, Moyses, MD, MPH,DrPH (Eds.), Gordis epidemiology (pp. 65-93) from https://www.clinicalkey.es/#!/content/3-s2.0-B9780323552295000048
- CDC. Section 1: Frequency Measures, Lesson 3: Measures of Risk. Principles of Epidemiology in Public Health Practice, Third Edition. CDC. Retrieved August 1, 2020, from https://www.cdc.gov/csels/dsepd/ss1978/lesson3/section1.html
- Boston University School of Public Health. Measures of Disease Frequency. Retrieved on August 1, 2020, from https://sphweb.bumc.bu.edu/otlt/MPH-Modules/EP/EP713_DiseaseFrequency/index.html