Study power can be thought of as how likely your study will find a statistically significant difference between unexposed and exposed groups, if the exposure is actually causing this difference in disease rates. When calculating the study power for a health study looking at the relationship between exposure and disease, epidemiologists estimate the incidence rate for the unexposed and exposed groups.

Different Types of Disease Rates
When trying to find estimates for your study power formula, it is helpful to know that epidemiologists use two different types of measurements to assess how many people are affected by a disease in a population.
 Type of Measurement Description How to Calculate It Question It Answers incidence percentage of a population that develops a particular disease over a period of time dividing the number of new cases by the number of people in the population at risk over a period of time how many people have been newly diagnosed with the disease during this period of time? prevalence percentage of a population that has a particular disease at a given point in time dividing the number of existing cases by the number of people in the population at risk at a given point in time how many people have the disease at this point in time?
When to Use Incidence
Epidemiologists often use incidence when trying to study the causes of disease, or factors which increase risk for a certain health outcome. Because incidence tells you how often a health outcome occurs over a period of time, it can be used to estimate the risk of person developing the health outcome.

Incidence is useful for certain health outcomes that happen as a single episode or event, such as an asthma hospitalization or a cancer diagnosis. Incidence is useful during investigations of diseases with relatively short durations, such as a food borne illness or infectious disease.

When to Use Prevalence
Researchers commonly use prevalence to assess how much a health outcome impacts a community and what types of health service needs exist in a community. Prevalence tells you how many people have a certain health outcome or condition at a given point in time in a community. It accounts for people who have just developed the health outcome, as well as people who have had the health outcome for a long time.

Prevalence is useful for measuring how many people have a chronic disease of long duration and relatively stable conditions, such as asthma or diabetes. Prevalence is also applicable for rarer chronic diseases in which it would be difficult to accumulate a large number of new cases developing in a specific time period, such as multiple sclerosis.

Epidemiologists must be careful when comparing prevalence between an exposed and unexposed group, and then trying to draw associations between exposure and disease. With prevalence, it is often hard to distinguish between factors that affect occurrence of the disease, versus those which increase the number of people who survive having a disease. For example, an increase in prevalence of a disease in a population may be due to improved access to optimal care and treatment for this disease, and not due to decreased exposure to a hazardous substance.

Examples of Prevalence and Incidence Rates
Here are examples of diseases and the available prevalence and annual incidence rates for specific populations.
 Disease Specific Population Annual Incidence:New cases per100,000 populationper year Prevalence:Existing cases per100,000 population brain cancer US population 6 38 01 breast cancer US population 128 803 02 leukemia US population 12 70 03 diabetes (type 2) US population 698 6,933 04 parkinson's disease US population 17 167 05 alzheimer's disease US population 120 1,667 06 chronic kidney disease US population:over 20 years old --- 8,667 07 end stage renal disease US population:over 20 years old 34 --- 08 infertility US population:reproductive age --- 10,000 09 birth defects ormental retardation at birth California population:births 3,030 --- 10 asthma California population:children --- 8,600 11 asthma hospitalization California population 100 --- 12

Data on both prevalence and incidence rates for certain diseases, such as different types of cancers, are routinely collected on a nationwide basis. For other health outcomes, such as infertility, only prevalence data is widely collected for a large population; while for other health outcomes, such as asthma hospitalization, only incidence data is collected.

Comparing Disease Rates Among Exposed and Unexposed Groups
When comparing incidence rates with each other, epidemiologists often calculate relative risk. Relative risk is calculated by dividing the incidence rate in the exposed group by the incidence rate in the unexposed group. Relative risk is how many times greater the disease occurrence in the exposed group is compared to the unexposed group. Another way to think about relative risk is that it answers the question, "What is the ratio of risk of disease among the exposed group to the risk of disease among the unexposed group?" For example, in one study, the relative risk for developing mesothelioma due to household asbestos exposure was 8. This means that the rate of mesothelioma was 8 times higher in the group exposed to household asbestos compared to the group that was not exposed. In other words, there was a 700% greater risk of mesothelioma for the group exposed to household asbestos compared to the group that was not exposed.

When calculating study power, epidemiologists predict incidence rates for exposed versus unexposed groups. Epidemiologists sometimes base these predictions on relative risks reported in previous studies.

Epidemiologists must be careful when comparing prevalence between an exposed and unexposed group, and then trying to draw associations between exposure and disease. With prevalence, it is often hard to distinguish between factors that affect occurrence of the disease, versus those which increase the number of people who survive having a disease. For example, an increase in prevalence of a disease in a population may be due improved access to optimal care and treatment for this disease, and not due to decreased exposure to a hazardous substance.

Here are specific examples of how much the rate of disease increased due to exposure according to health studies conducted in the past. The "% Greater Risk" column compares the risk of disease in exposed versus unexposed populations.

 Exposure Disease Relative Risk % Greater Risk asbestos (household) mesothelioma 8 700 13 asbestos (neighborhood) mesothelioma 7 600 13 smoking lung cancer (men) 23 2,200 14 smoking lung cancer (women) 12.7 1,170 14 secondhand smoke lung cancer 1.26 26 15 secondhand smoke SIDS 1.4 40 16 maternal smoking SIDS 4.7 370 16 benzene (occupational) leukemia 3.37 237 17 benzene (occupational) multiple myeloma 4.09 309 17
References
Disease Prevalence and Incidence
Relative Risks
12 - Asbestos and Mesothelioma
Bourdès V, Boffetta P, Pisani P. Environmental exposure to asbestos and risk of pleural mesothelioma: review and meta-analysis. European Journal of Epidemiology. 2000 May;16(5):411-7
13 - Secondhand Smoke and Lung Cancer
Hackshaw AK, Law MR, Wald NJ. The accumulated evidence on lung cancer and environmental tobacco smoke. British Medical Journal. 1997;315:980-988 (18 October)
14 - Secondhand Smoke and Maternal Smoking and SIDS
15 - Benzene and Leukemia