6.5: Incidence vs. Prevalence
When epidemiologists track diseases or conditions, there are two numbers that can express the current effect of the disease or condition on the population. Incidence refers to the number of new cases diagnosed during a specific time period. These cases are added to the total number of current cases for a specific disease. The total number of cases at a specific point in time or period of time is called prevalence, which counts anyone with a current diagnosed case of the disease. Of course, cases also exit this count when they are cured or deceased. That is why knowing the time component for both prevalence and incidence is very important.
A good way to understand the difference between incidence and prevalence is to imagine you’re standing at a riverbank looking at a river. The river itself can represent time, because it is constantly flowing past you. You count the fish that you can see at that point in the river - this is the prevalence of fish at that point in the river. A few yards upstream, you notice a Fish and Wildlife officer dumping buckets of fish into the water. This is the incidence - the number of new fish entering the river. How many of the fish you counted originally had been there for a while? How many had just been dumped in? You wouldn’t know whether the fish you just counted had just been added to the river or if they had lived there a long time already. You also don’t know how many people are standing on the bank downriver from you with their lines in the water, catching those fish out of the river. If you’re assessing a point-in-time prevalence, then all you can see are the fish that you are counting in the portion of the river right in front of you. See Fig. \(\PageIndex{1}\) for a diagram of this analogy.
Another way to think of prevalence is like a pool. Swimmers can enter the pool (get a diagnosis) or exit the pool (disease is cured or patient is deceased). The number that enter the pool in a given timeframe are considered the incidence of that disease. Those that are in the pool in a given timeframe constitute the prevalence. This prevalence is then highly influenced by the severity of the disease and how treatable it is - do patients live with this disease for a long period of time? Does the disease kill its victims quickly? Do we have medicines and treatments that extend the life of those diagnosed? See Fig. \(\PageIndex{2}\) for a visual diagram of this analogy.
As an example, let’s examine the incidence and prevalence of a chronic disease. In 2021, 1.2 million new cases of diabetes were diagnosed in the United States. Those 1.2 million are the incidence for the year 2021. The prevalence of diabetes was estimated at 38.1 million for the year 2021, including both diagnosed and assumed undiagnosed cases ( National Diabetes Statistics Report , 2023). Diabetes is a chronic disease, in which the body’s ability to utilize glucose (sugar) is decreased. This causes hyperglycemia (high blood sugar), which can be damaging to the nervous and cardiovascular systems, and could eventually cause kidney failure. Diabetes is also a risk factor for heart disease. So the prevalence “pool” may be reduced by some cases who are deceased within that year from various causes. Although there is no cure for diabetes, many people who have diabetes can manage it, or potentially reduce the severity of the disease with medicines and lifestyle behaviors like diet and exercise - thus essentially going into remission. There is the potential that some cases could exit the “pool” if their blood sugar is managed at below-diabetes levels, and the disease is prevented from progressing.
Knowing the prevalence of diabetes is helpful because we know how many people are currently living with the condition. These people may require treatment, medicine, and ongoing monitoring, and it’s helpful to know how to dedicate resources. If the prevalence of a disease gets smaller over time, it can indicate different things. Perhaps fewer people are getting diagnosed with diabetes - so the incidence is decreasing. If the incidence is decreasing and that is causing a decrease in prevalence, then perhaps public health efforts to prevent diabetes are working, which would be a positive outcome. Or, another possibility is that less screening is taking place, and therefore more cases are simply going undiagnosed. But let’s say that the prevalence is decreasing, but the incidence remains the same - or is perhaps higher than in years past. This could be for different reasons - perhaps more people are dying from diabetes, or it is becoming more difficult to treat. Or perhaps we are finding better medicines to reverse diabetes, but we are not able to prevent it. As this example demonstrates, knowing both the incidence and the prevalence helps an epidemiologist understand the disease better, but isn’t enough information alone to identify problems or solutions for a particular disease.
Now let’s look at the incidence and prevalence of an infectious disease with a much shorter time frame. Salmonellosis is caused by infection with the salmonella bacteria, often caused by eating contaminated foods like undercooked eggs or poultry. Most commonly this infection causes fever, diarrhea and stomach pain within several hours to a few days after exposure. The time between exposure to the pathogen (in this case, bacteria) and the symptoms is called the incubation period . Symptoms can last 4-7 days, and some people experience symptoms up to several weeks ( Questions and Answers , 2023). The period of time when someone can transmit the disease to others is termed the infectious period . For this type of disease, knowing the prevalence on some random day of the year doesn’t really do us much good. That’s because people don’t live with salmonellosis for a long time - they usually recover within days to weeks, although they may remain infectious for several months. The prevalence of people with salmonellosis on a Tuesday in March of 2023 doesn’t tell us how long they have been sick, or help us identify where they got the infection. But knowing the incidence of salmonellosis over a period of time is very helpful. A rapid increase in the incidence over the course of several days would indicate a potential outbreak - which would require immediate action to locate the source of contaminated food and prevent anyone else from eating it. It’s also valuable to know if there is a year-over-year increasing incidence of salmonellosis, which may indicate that food safety regulations are not effective or being enforced.