Principles of Epidemiology

This exercise illustrates methods of measuring the occurrence and the outcomes of disease in populations. These methods are commonly used in describing the effects of disease in the population (descriptive epidemiology) and in investigations to test a hypothesis about disease occurrence (analytic epidemiology).

Measures of disease occurrence and outcomes are derived from several sources of information, including:

· Death certificates

· Birth certificates

· Disease reporting and disease registers

· Surveys

· Hospital and medical records

· Occupational health records

· Special studies

· Others

Population size and characteristics are derived primarily from a census — the decennial U.S. Federal Census of the Population or a special enumeration of a population group under study.

I. RATES (MEASURES OF RISK)

There are three essential elements in a rate: the event, i.e., disease, cause of death (numerator), the population in which the event occurred (denominator), and time (interval during which the event takes place).

image1.png

A. CRUDE RATES ‑ the number of events occurring in a defined population during a specified interval of time (commonly one year).

image2.png

image3.png

* For purposes of convenience in managing numbers, rates are expressed in terms of the number occurring in 1,000, 10,000, 100,000 or million in the population. It does not matter which multiple is used, but this constant must be stated. Unless otherwise stated, the time period is generally assumed to be a year

– but the specific year should be stated, i.e. “for 2009”.

B. SPECIFIC RATES ‑ the number of events of a specified disease occurring in a defined population

of age, race, or sex during a specified interval.

image4.png

Rates that are disease specific only, the number with that disease (or dying of that disease) is divided by the entire population. However, if rates are also age, sex, and/or race specific rated are divided by the age, sex and/or race sub-population.

C. EXERCISES (2 points per question -40 total points)

1. SHOW YOUR FORMULA AND WORK

2. USE 2 DECIMAL POINTS

1. Crude Mortality rate: There were 5,251 deaths (all causes) during the period of January 1 ‑ December 31, in residents of Marion County among an estimated total population of 864,550.

Calculate the crude mortality rate per 100,000

2. Specific rates ‑ Of 17,349 deaths certified as disease of the heart in Indiana in 2009, 717 occurred among white males aged 45‑54 years. There were 247,401 white males aged 45‑54 in the population.

Calculate an age, race, and sex specific mortality rate per 100,000.

3. Specific rates – During the same year, 61 heart disease deaths occurred among black males, aged 45‑54, in Indiana. There were 18,899 black males aged 45‑54 in the population per 100,000.

Calculate an age specific (45-54 year old) heart disease death rate for black males.

II. PROPORTIONATE MORTALITY RATIO

The proportion of total deaths due to a specific cause. This is not a measure of risk, but used to illustrate the importance of a specific cause of death among total deaths. It is expressed as a percentage of the total; therefore, the multiple “100” is used in the calculation.

Number of deaths from a specific cause x 100

Total number of deaths from all causes

Of 47,300 deaths (all causes) in Indiana during 2009, 10,432 were determined to be due to cancer.

4.) What was the proportionate cancer mortality rate?

A proportionate morbidity could also be calculated using cases of disease rather than cases of death.

Proportionate mortality (or proportionate morbidity) may be calculated as age, race, or sex specific, if that information is known. However, proportionate mortality is more often used when the size, age, race and other characteristics of the population at risk are unknown. Rates are more often used whenever the charac​teristics of cases and the characteristics of the population are known, because rates measure the risk of an event in a defined community.

Case fatality rate ‑ The case fatality rate expresses the risk of death among persons who have a disease, i.e., the proportion of deaths among cases with the disease (it is expressed as a percentage).

Number of deaths due to a specific disease

Case Fatality Rate = ——————————————————– X 100 Number of cases of a specific disease per year

In a recently reported outbreak of 89 cases of Salmonella infection (Salmonellosis) one year in a long-term care facility, 5 died as a result of the infection.

5.) What was the case fatality rate for Salmonellosis for this outbreak?

6.) As a review of information stated so far, what other information would be needed to assess the magnitude of the problem at the long-term care facility, i.e., the institutional risk of the disease?

During 2009, 110 persons were admitted to North Community Hospital and diagnosed with lung cancer. Of these cases, 61 died and 49 were discharged alive.

7.) What was the lung cancer fatality rate for 2009 at North Community Hospital?

Birth rate ‑ The number of live births during a specified period of time (i.e., one calendar year) per 1,000 population.

Number of live births x 1,000

Population

Infant mortality rate ‑ Infant mortality measures the risk of dying during the first year of life among infants who were born alive.

Number of infant deaths between 0‑365 days of life x 1,000

Total number of live births

Neonatal mortality rate ‑ The number of infant deaths from 0 to 28 days of life per 1,000 live births.

Early neonatal mortality rate ‑ The number of infant deaths from 0 to 7 days of life per 1,000 live births.

Late neonatal mortality rate ‑ The number of infant deaths from 7 to 28 days of life per 1,000 live births.

Perinatal mortality rate ‑ The number of deaths between 28 weeks gestation to 7 days after birth per 1,000 live births plus fetal deaths.

Fetal mortality rate ‑ The number of fetal deaths after 20 weeks gestation per 1,000 live births plus fetal deaths.

Fetal mortality ratio – The number of fetal deaths divided by the number of live births times 1,000.

Maternal mortality rate ‑ The number of maternal deaths ascribed to prenatal, natal, or postnatal causes per 100,000 live births.

8.) In 2009, there were 105,531 births in Indiana (approximate population 5,547,000). What was the birth rate for Indiana in 2009?

9.) In 2008, the infant mortality rate for the State was 8.8 per 1000 (5.5 per 1000 neonatal, 3.3 per 1000 post-neonatal). Calculate the following assuming 105,531 births in Indiana in 2009.

HINT – you are calculating the total number of infant deaths.

a. How many infants (total) would you expect to die in 2009?

b. How many infants (total) would you expect to die in the neonatal period in 2009?

c. How many infants (total) would you expect to die in the post-neonatal period in 2009?

10.) In 2009 there were 299 fetal deaths and only 7 maternal deaths. Assume 105,531 births in Indiana in 2009.

a. What was the fetal death rate for 2009?

b. What was the fetal death ratio for 2009?

c. What was the maternal mortality rate for 2009?

Morbidity – The Center for disease control require reporting of certain diseases to monitor outbreaks and determine national priorities. These rates are also useful on a local or state level to assess need and monitor changes in disease rates. While reporting of cases may be incomplete in some places and for many diseases, systematic reporting permits surveillance of secular trends, seasonal distribution, and distribution by location, age, race, sex, economic status, etc.

Reported case rate ‑ The number of cases of a notifiable disease reported during a specified period of time per 100,000 population (a crude rate). Age, race, and sex specific reported case rates may be obtained in the same manner as crude death rates.

During the period January 1 ‑ October 31, 2009, 116cases of syphilis were reported to the Marion County Health Department (the estimated population of Marion County was 862,311 in 2006).

11.) What is the syphilis reported case rate ?

12.) What is your estimate of accuracy of this rate?

(HINT – Based on what you know about syphilis and other sexually transmitted infections (STIs), do you THINK the estimate you calculated in #11 is accurate? Yes or No AND briefly justify your answer.)

13.) During the same period presented above in the case for #11 and #12) , 37 cases of tuberculosis were reported. Do you think an estimate of case reporting for tuberculosis is more or less complete as compared to syphilis? Why would one be more completely recorded than the other?

(HINT – Based on what you know about tuberculosis and syphilis, identify which estimate is more or less complete – syphilis or tuberculosis b riefly justify your answer.)

Sex ratio ‑ Occasionally, the sex ratio is reported to express differences in the occurrence of an event among males and females.

Number of male cases

Number of female cases

There were 18,252 deaths certified as diseases of the heart in Indiana in 2009; 10,109 occurred among males and 8,143 occurred among females.

14.) Calculate the sex ratio for heart disease in Indiana for 2009. Be sure to WRITE the result as a ratio .

Prevalence and incidence are important measures of disease in a population, but they measure disease rates differently.

Prevalence is a measure of the total number of living cases of a specified disease (existing and new cases) present in a defined population at a given point in time:

Number of total cases of a disease at a specific time period x 1,000

Total Population

Incidence is a measure of the new cases of a specific disease occurring in a population during some time interval:

Number of new cases of a disease at a specific time period x 1,000

Population at risk

A study starting on January 1, 2009 of a 10% random sample of state employees revealed 67 of the 2,589 individuals in the study had been diagnosed with hypertension, 22 of whom had been diagnosed during 2008.

15.) What was the prevalence of hypertension in the sample?

16.) What was the incidence of hypertension in the sample during 2008?

Note that prevalence is a measure of the amount of a disease existing in a population –

it is not a measure of the risk of acquiring the disease. All individuals with the disease are included in the numerator of prevalence, including those who migrated into the popula​tion while diseased. Only living cases are included. Individuals are excluded from the prevalence numerator who died or recovered or migrated out the population before the study period.

Only new cases are included in the incidence numerator. The incidence denominator is the population at risk at the beginning of the time period during which incidence is to be measured, that is persons without the disease, but who are able to contract the disease. This exact measure for the denomination is often difficult to obtain. In that case, the denominator would be the population of the area minus those who already have the disease (prevalence), if that information is known.