PU520: Principles of Epidemiology Unit 4 Epidemiologic Measurements

Question 1

What are important mathematical terms applied to epidemiologic measures which are all types of ratios?

Answer 1

Rate, proportion (includes risk), and percentage.

Ratio as well.

Question 2

What is defined as the value obtained by dividing one quantity by another?

How is it expressed?

Answer 2

Ratio

Ratios are sometimes expressed as percentages.

Ratio = X/Y

Example l: With respect to AIDS mortality, the sex ratio of deaths (male to female deaths) = X/Y, where: X = 450,451 and Y = 89,895. The sex ratio = 450,451/89,895 = 5 to 1 (approximately).

Question 3

What is a type of ratio in which the numerator is part of the denominator?

How is it expressed?

Answer 3

Proportion

Proportions may be expressed as percentages.

Proportion is expressed as follows: proportion = A/A+B

Example 1: Proportion of AIDS deaths Suppose that

A = the number of male deaths from AIDS
A = 450,451
B = the number of female deaths from AIDS
B = 89,895

The proportion of deaths that occurred among males = 450,451/(450,451 + 89,895) = 0.83

Question 4

What is a proportion that has been multiplied by 100?

How is it expressed?

Answer 4

Percentage

(A/A + B) x 100

Example 1: The percentage of male deaths from AIDS was (0.83 × 100) = 83%.

Question 5

How can proportion (including percentage) be helpful in describing how important a health outcome is?

Answer 5

Because it looks at a health outcome and measures the cases across the population in which they occur.

For example, if 10 dorm residents get mononucleosis, we would want to know how large of a problem these 10 cases represent. If the dorm houses 20 total residents, that is 50% and seems pretty important.

While if the population of the dorm is 500, these cases would represent 2% of the population which paints an entirely different picture.

Clearly, these two scenarios paint a completely different picture of the magnitude of the problem. In this situation, expressing the count as a proportion is indeed helpful. In most situations, it will be informative to have some idea about the size of the denominator. Although the construction of a proportion is straightforward, one of the central concerns of epidemiology is to find and enumerate appropriate denominators to describe and compare groups in a meaningful and useful way.

Question 6

How does rate (also a type of ratio) differ from a proportion because the denominator involves a measure of time?

Answer 6

Rate

Epidemiologic rates are composed of a numerator (the number of events such as health outcomes), a denominator (a population in which the events occur), and a measure of time. This measure of time is the time period during which events in the numerator occur. The denominator consists of the average population in which the events occurred during this same time period.

Question 7

What are rates used to measure in epidemiology? (2)

Answer 7

To measure risks associated with exposures and provide information about the speed of development of a disease.

They are also used to make comparisons among populations.

Question 8

T/F Medical publications may use the terms ratio, proportion, and rate without strict adherence to the mathematical definitions for these terms. Hence, you must be alert regarding how a measure is defined and calculated.

Answer 8

True

Question 9

What types of information does epidemiologic measures provide? (3)

Answer 9

• the frequency of a disease or condition
• associations between exposures and health outcomes
• strength of the relationship between an exposure and a health outcome

Question 10

How is the numerator in epidemiologic measures typically defined?

Answer 10

• Case definition (condition) - For epidemiologic measures to be valid, the case of disease or other health phenomenon being studied must be defined carefully and in a manner that can be replicated by others
• Frequency - How many cases are there
• Severity - Some epidemiologic measures employ morbidity as the number and others use mortality

Question 11

How is the denominator in epidemiologic measures typically defined?

Answer 11

Does the measure make use of the entire population or a subset of the population?

Some measures use the population at risk, defined as those members of the population who are capable of developing a disease, for example, people who are not immune to an infectious disease.

• Existing (all cases) versus new cases.

Question 12

What is the simplest and most frequently performed quantitative measure in epidemiology?

Answer 12

A count and it refers merely to the number of cases of a disease or other health phenomenon being studied.

The reader with the impression that counts, because they are simple measures, are of little value in epidemiology; this is not true, however. In fact, case reports of patients with particularly unusual presentations or combinations of symptoms often spur epidemiologic investigations. In addition, for some diseases even a single case is sufficient to be of public health importance.

For example, if a case of smallpox (now eradicated) or Ebola virus disease were reported, the size of the denominator would be irrelevant. That is, in these instances a single case, regardless of the size of the population at risk, would stimulate an investigation.

Question 13

What does the term incidence refer to?

Answer 13

The number of instances of illness commencing, or of persons falling ill, during a given period in a specified population.

More generally, the number of new health-related events in a defined population within a specified period of time.

Question 14

What are the different ways to express incidence?

Answer 14

Incidence rate, cumulative incidence, incidence density, and attack rate.

Question 15

How is incidence rate defined and expressed?

Answer 15

The rate at which new events occur in a population.

The new events are usually new cases of disease but can be other health outcomes. The incidence rate is a rate because a time period during which the new cases occur is specified and the population at risk is observed.

The incidence rate denotes a rate formed by dividing the number of new cases that occur during a time period by the average number of individuals in the population at risk during the same time period times a multiplier.

Question 16

When calculating incidence rate, why is it important to always choose an average number for the denominator when talking about a population during a specific time?

Answer 16

The midpoint of the population during the timeframe is chosen because of migration and how often populations change.

To overcome this challenge, the population at the midpoint of the year is used as the denominator and is considered to be the average population at risk.

Question 17

When is an incidence rate called a cumulative incidence (incidence proportion)?

Answer 17

When ALL individuals in the population are at risk throughout the time period during which they were observed.

An example of a population in which all members of the population are at risk is a fixed or closed population (such as the participants in a cohort study) in which no new members are allowed to enter the study after it begins.

Here is a hypothetical calculation example for cumula-tive incidence: An epidemiologist studies cardiovascular disease among 23,502 male middle-aged alumni of an Ivy League university. Initial medical examinations certify that the alumni have never had a heart attack in the past. During the first year of the research, 111 alums have heart attacks.

Cumulative incidence = 111/23,502 = .005 (or if a percentage, 0.5%)

Question 18

What is a variation of an incidence rate that is used when the time periods of observation of the members of a population vary from person to person?

Answer 18

Incidence density.

During a study that takes place over an extended period of time (for example, a cohort study, which is described later in the text), participants may be observed for varying periods of time because some drop out before the study is completed.

In order to make use of all participants’ data, we calculate incidence density according to the formula shown in the box. The numerator is the number of new cases during the time period and the denominator is the total person-time of observation.

Person-time is the total period of time that each individual at risk has been observed. For example, one person-year means that one subject has been observed for one year.

Question 19

What is a type of incidence rate used when the occurrence of disease among a population at risk increases greatly over a short period of time?

Answer 19

Attack Rate

This is often related to a specific exposure, and is frequently used to describe the occurrence of foodborne illness, infectious diseases, and other acute epidemics.

An attack rate is not a true rate because the time dimension is often uncertain.

Question 20

What, expressed as a proportion, refers to the number of existing cases of a disease or health condition, or deaths in a population at some designated time divided by the number of persons in that population?

What are the two types of this term?

Answer 20

PrevALence

Point prevalence - refers to all cases of a disease, health condition, or deaths that exist at a particular point in time relative to a specific population from which the cases are derived.

For example, if we are referring to illness (morbidity) in a group of people, the formula for point prevalence is shown in the following box.

Point prevalence = # ill at a point in time / total # in the group at a point in time

Point prevalence may be expressed as a proportion formed by dividing the number of cases that occur in a population by the size of the population in which the cases occurred times a multiplier. Note that point prevalence is a proportion and not a rate. If the value of 100 is used as the multiplier, prevalence becomes a percentage.

Period prevalence denotes the total number of cases of a disease that exist during a specific period of time (e.g., a week, month, year, or other interval).

An example of period prevalence is asthma period prevalence. The numerator for asthma period prevalence reflects whether a respondent answered affirmatively to the question “during the past 12 months, did anyone in the family have asthma?”
The time period for this measure is the past year.

Question 21

Food For Thought

What does lifetime prevalence denote?

Answer 21

Lifetime prevalence denotes cases of disease diagnosed at any time during the person’s lifetime.

Question 22

What does prevalence measures used to describe

Answer 22

Prevalence measures are used to describe the scope and distribution of health outcomes in the population.

The scope or amount of disease is called the burden of disease in the population.

By offering a snapshot of disease occurrence, prevalence data contribute to the accomplishment of two of the primary functions of descriptive epidemiology: to assess variations in the occurrence of disease and to aid in the development of hypotheses that can be followed up by analytic studies.

Populations that differ in size cannot be compared directly by using frequency data, i.e., just the numbers of cases. In order to make such comparisons, prevalence proportions need to be calculated. Then it is possible to compare the proportions of health outcomes among different geographic areas. See image.

Question 23

How is incidence and prevalence interrelated concepts?

Answer 23

Prevalence of disease (P) is proportional to the incidence of the disease (I) times the duration of the disease (D).

P = ID

Consequently, when the incidence of a disease increases, the prevalence also increases.

Question 24

What are other factors that cause prevalence of a disease to increase?

Answer 24

Its duration, in-migration of new cases, and development of treatments for the disease, including methods for extending the lives of patients who may not actually be cured.

An example of how the dura-tion of a disease affects its prevalence would be two diseases (A—long duration and B—short duration) that have similar incidence rates; we would expect disease A to have a higher prevalence than disease B. In the United States, the incidence of HIV infections has tended to remain constant over time and is much lower than HIV prevalence. Because more people with HIV infection are surviving for longer time periods, the prevalence of HIV is much greater than its incidence.

Question 25

What term refers to the number of years that a person is expected to live, at any particular year?

Answer 25

Life expectancy

Life expectancy at birth represents the average number of years that a group of infants would live if the group was to experience throughout life the age-specific death rates present in the year of birth.

Question 26

What are the three general categories of rate in which it can be broken down in?

Answer 26

Crude rate = a type of rate that has not been modified to take into account any of the factors, such as the demographic makeup of the population, that may affect the observed rate.

Crude rates are summary rates based on the actual number of events in a population over a given time period. The numerator consists of the frequency of a disease (or other health-related outcome) over a specified period of time, and the denominator is a unit size of population.

An example is the crude death rate, which approximates the portion of a population that dies during a time period of interest.

The denominator is also termed the reference population, which is defined as the population from which cases of a disease have been taken.

Question 27

Rate Information (No Flipside)

Rates improve our ability to make comparisons, although they also have limitations. For example, rates of mortality for a specific disease (see the section on cause-specific mortality rates later in this chapter) reduce the standard of comparison to a common denominator, the unit size of population.

To illustrate, the U.S. crude death rate for diseases of the heart in 2013 was 193.3 per 100,000. One also might calculate the heart disease death rate for geographic subdivisions of the country (also expressed as frequency per 100,000 individuals).

These rates then could be compared with one another and with the rate for the United States for judging whether the rates found in each geographic area are higher or lower. For example, the crude death rates in 2013 for diseases of the heart in New York and Texas were 224.1 and 152.0 per 100,000, respectively.

On the basis of the crude death rates, it would appear that the death rate was much higher in New York than in Texas or the United States as a whole. This may be a specious conclusion, however, because there may be important differences in population composition (e.g., age differences between populations) that would affect mortality experience.

Rates can be expressed in terms of any unit size of population that is convenient (e.g., per 1,000, per 100,000, or per 1,000,000). Many of the rates that are published and routinely used as indicators of public health are expressed according to a particular convention. For example, cancer rates are typically expressed per 100,000 population, and infant mortality is expressed per 1,000 live births.

One of the determinants of the size of the denominator is whether the numerator is large enough to permit the rate to be expressed as an integer or an integer plus a trailing deci-mal (e.g., 4 or 4.2). For example, it would be preferable to describe the occurrence of disease as 4 per 100,000 rather than 0.04 per 1,000, even though both are perfectly correct. Throughout this chapter, the multiplier for a given morbidity or mortality statistic is provided.

Answer 27

None

Question 28

What refers to the number of deaths due to disease that occur among people who are afflicted with the disease?

Answer 28

Case Fatality Rate (CFR)

The CFR(%), which provides a measure of the lethality of a disease, is defined as the number of deaths due to a specific disease within a specified time period divided by the number of cases of that disease during the same time period multiplied by 100.

The numerator and denominator refer to the same time period.

Question 29

What is the number of deaths within a population due to a specific disease or cause divided by the total number of deaths in the population (and multiplied by 100)?

Answer 29

Proportional Mortality Ratio

Question 30

What is a statistic referring to a particular subgroup of the population defined in terms of race, age, or sex?

Answer 30

Specific rate

A specific rate also may refer to the entire population but is specific for some single cause of death or illness.

Question 31

What are the three most specific rates you will use and encounter and how are they expressed?

Answer 31

Cause-specific rate - is a measure that refers to mortality (or frequency of a given disease) divided by the population size at the midpoint of a time period times a multiplier.

An example of a cause-specific rate is the cause-specific mortality rate, which, as the name implies, is the rate associated with a specific cause of death.

Mortality (or frequency of a given disease) / population size at midpoint of time period x 100,000

Age-specific rate - refers to the number of cases of dis-ease per age group of the population during a specified time period.

Age-specific rates help in making comparisons regarding a cause of morbidity or mortality across age groups. A more precise definition of an age-specific rate is the frequency of a disease (or health condition) in a particular age stratum divided by the total number of persons within that age stratum during a time period. See Image.

Sex-specific rate - refers to the frequency of a disease in a sex group divided by the total number of persons within that sex group during a time period times a multiplier.

Number of deaths in a sex group / total number of persons in the sex group x 100,000

Question 32

What is a rate of morbidity or mortality in a population in which statistical procedures have been applied to permit FAIR comparisons across populations by removing the effect of differences in the composition of various populations?

Answer 32

Adjusted Rates

A factor in rate adjustment is age adjustment. Calculation of age-adjusted rates is a much more involved procedure than that required for crude rates. A weighting process entails the use of detailed information about the age structure of the population for which the rates are being age adjusted.

For example, “age-adjusted death rates are constructs that show what the level of mortality would be if no changes occurred in the age composition of the population from year to year.

The direct method of age adjustment involves multiplying the age-specific rates for each subgroup of a population to be standardized by the number in a comparable subgroup of a standard population.

To age adjust the crude mortality rate in the United States, we would use the standard population, which for the United States is the year 2000 population.

For example, suppose you wanted to standardize the crude mortality data for the United States for 2003. You would multiply the age-specific death rate for the population under age 1 (700.0 per 100,000) in 2003 by the number in the year 2000 standard population under age 1 (3,794,301). This calculation would need to be repeated for each age stratum. The results for each stratum would then be summed to create a weighted average—the age-adjusted death rate.

Question 33

What does maternal mortality encompass?

What is maternal mortality rate?

Answer 33

Deaths that result from causes associated with pregnancy.

Among the factors related to maternal mortality are race, insufficient healthcare access, and social disadvantage.

The maternal mortality rate is the number of maternal deaths ascribed to childbirth divided by the number of live births times 100,000 live births during a year.

In 2005, the maternal mortality rate was 15.1 deaths per 100,000 live births (623 total deaths in 2005). The respective maternal mortality rates per 100,000 live births for black and white women were 36.5 and 11.1; the rate for black women was about 3.3 times that for white women.

Question 34

What is defined as the number of infant deaths among infants age 0 to 365 days during a year divided by the number of live births during the same year (expressed as the rate per 1,000 live births)?

Answer 34

Infant mortality rate

The two categories of infant mortality rate are neonatal mortality rate and postneonatal mortality rate.

The neonatal mortality rate is the number of infant deaths under 28 days of age divided by the number of live births during a year.

The postneonatal mortality rate refers to the number of infant deaths from 28 days to 365 days after birth divided by the number of live births minus neonatal deaths during a year (expressed as rate per 1,000 live births).

Question 35

What is refined as the death of a fetus when it is in the uterus and before it has been delivered?

What are the two measures of this?

Answer 35

Fetal Mortality

Fetal death rate and late fetal death rate.

Question 36

What is a crude birth rate?

Answer 36

The crude birth rate refers to the number of live births during a specified period such as a year per the resident population at the midpoint of the year. The birth rate affects the total size of the population.

Question 37

What is general fertility rate?

Answer 37

Related to birth rates is the general fertility rate, which refers to the number of live births reported in an area during a given time interval divided by the number of women age 15 to 44 years in the area (expressed as rate per 1,000 women age 15 to 44 years). The general fertility rate is referred to more broadly as the fertility rate.

Question 38

What is defined as the number of late fetal deaths (after 28 weeks or more gestation) plus infant deaths within 7 days of birth divided by the number of live births plus the number of late fetal deaths during a year (expressed as rate per 1,000 live births and fetal deaths)?

Answer 38

Perinatal Mortality