**Estimating Risk:**

Is there an Association?

Is there an Association?

**Not everything that can be counted counts,**

and not everything that counts can be counted.

—William Bruce Cameron, 1963

and not everything that counts can be counted.

—William Bruce Cameron, 1963

**Frequency Measures Used in Epidemiology**

**NRS 411 - Epidemiology**

CDC. (2006). Applied epidemiology for public health nurses. Module 2: Frequency measures used in epidemiology. From

Principles of epidemiology: An introduction to applied epidemiology and biostatistics

(2nd ed.).

Gordis, L. (2014).

Epidemiology

(5th ed.). St. Louis, MO: Elsevier Saunders.

Powell, T. (2008).

Confidence intervals.

Retreived from http://www.in.gov/isdh/24228.htm

Powell, T. (2008).

The p-value.

Retrieved from http://www.in.gov/isdh/24207.htm

References

**Measures**

of Morbidity

of Morbidity

http://www.oregonlive.com/education/index.ssf/2016/04/norovirus_outbreak_confirmed_o.html

Using

rates

and

proportions

to express

the extent of morbidity resulting from a disease

Incidence rate

- New cases/population at risk~time period

Attack rate

- Sick exposed folks/Well exposed folks

Prevalence

- Affected folks/Total population~spot in time

Identify New Cases to Calculate Incidence

1. Spot in time

2. Period of time

Point & Period Prevalence

and

Cumulative Incidence

Interview Question

“Do you currently have asthma?”

“Have you had asthma during the last [n] years?"

“Have you ever had asthma?”

Type of Measure

Point prevalence

Period prevalence

Cumulative incidence

What is the numerator for

incidence

in 2012?

What is the numerator for

point prevalence

in 2012?

May

July

September

November

Relationship Between Incidence & Prevalence

Prevalence in the population

Increased incidence = increased prevalence

Decreased prevalence from death or cure

Trends in Data

&

Quality of Data

Trends in prevalence of obesity, by state, United States, 1990, 1995, 2000, 2005, and 2010,

based on self-reported height and weight

. Obesity was defined by BMI (body mass index) ≥30, or ~30 lbs overweight for a 5′4″ person. (Adapted from Centers for Disease Control and Prevention, based in part on data from the Behavioral Risk Factor Surveillance System.)

**Measures**

of Mortality

of Mortality

Expressing mortality in quantitative terms:

Can pinpoint

differences in the risk of dying

from a disease between people in different geographic areas and subgroups in the population

Can serve as measures of

disease severity

Can help determine whether

treatment

for a disease has become more effective over time

May serve as

surrogate for incidence rate

when the disease being studied is a severe and lethal one -

but not if the disease is mild and non-fatal

Mortality Rate

vs

Proportion

Proportion

of deaths

from heart disease

Rate

of deaths from

heart disease and cancer

3 Basic Study Designs in Epidemiology

Randomized clinical trial

Cohort study

Case-control study

Relative Risk

Relative risk

=

Risk in the exposed

Risk in the nonexposed

=

Incidence in the exposed

Incidence in the nonexposed

Negative association or a protective effect

Positive association, may be causal

Odds Ratio

(Relative Odds)

- The ratio of

the odds that the disease will develop in an

exposed person

to the odds it will develop in a

nonexposed person

Cohort Study

Case-Control Study

Odds ratio in both a cohort and a case-control study

The

odds ratio

is also known as the

cross-products ratio

, because it can be obtained by multiplying both diagonal cells in a 2 × 2 table and then dividing ad/bc:

Confidence Intervals

p-Value

When testing a hypothesis, the

p-Value

helps determine the significance of your results/validity of your claim/strength of the evidence

Your statement about what is currently believed about a population is the

"null hypothesis"

Versus the

"alternative hypothesis"

if the null hypothesis is untrue

This is your proposed hypothesis, or that which you wish to test

The p-value is a number between 0-1 comparing your

observed findings

to your

expected findings

, and lets you determine if there is a significant correlation

Interpreting the p-value:

A

small p-value (< 0.05

or 5%

)

is statistically significant (not likely attributable to chance), so you

reject the null hypothesis

Successfully shown a possible correlation related to your alternate hypothesis

A

large p-value (> 0.05)

indicates the events occur more commonly and are considered insignificant, so you

fail to reject the null hypothesis

p-Values very close to the cutoff (0.05) are considered to be marginal (could go either way)

All who drink of this treatment recover in a short time,

Except those whom it does not help, who all die,

It is obvious, therefore, that it fails only in incurable cases.

—Galen1 (129–c. 199 ce)

**The Importance**

of

Randomized Trials

and

Case Controls

of

Randomized Trials

and

Case Controls

One day when I was a junior medical student, a very important Boston surgeon visited the school and delivered a great treatise on a large number of patients who had undergone successful operations for vascular reconstruction. At the end of the lecture, a young student at the back of the room timidly asked, “Do you have any controls?” Well, the great surgeon drew himself up to his full height, hit the desk, and said, “Do you mean did I not operate on half of the patients?” The hall grew very quiet then. The voice at the back of the room very hesitantly replied, “Yes, that’s what I had in mind.” Then the visitor’s fist really came down as he thundered, “Of course not. That would have doomed half of them to their death.” God, it was quiet then, and one could scarcely hear the small voice ask, “Which half?” - E. Peacock

Studies Without Comparison

"Results can always be improved by the omission of controls."

- Professor Hugo Muensch's Second Law,

Harvard University

Experimental

Study

Observational

Study

Comparison

is an essential component of epidemiologic investigation and is well exemplified by the case-control study design

Calculating Relative Risk

Example - The Framingham (Cohort) Study

Only the

odds ratio

can be calculated as a measure of association; but we can estimate the relative risk in a case-control study from the odds ratio

Either the

relative risk

or the

odds ratio

is a valid measure of association

**CDC Case Study -**

Norovirus in Vermont

Norovirus in Vermont

Interpreting the Odds Ratio

It is the same as interpretation of the relative risk:

If the exposure is

not related

to the disease

The odds ratio will be

= 1

If the exposure is

positively related

to the disease

The odds ratio will be

> 1

If the exposure is

negatively related

to the disease

The odds ratio will be

< 1