NU

Measuring Cases of Disease

Session 4

Hussain Alqhtani
BS.Pharm., M.Sc., MPH, Ph.D.
1
 NU

Measuring New Cases of Disease

1
 Measuring Disease Frequency
• Morbidity – frequency of disease
• Two basic morbidity measurements are
incidence and prevalence
– Incidence – occurrence of new disease
– Prevalence – existence of current disease
 Incidence
• Quantifies number of new cases of disease
that develop in a population at risk during a
specified time period
• Three key concepts:
– New disease events
– Population at risk (candidate population)
– Diseases develop over time
• Two incidence measures
– Cumulative incidence (CI)
– Incidence rate (IR)
 Cumulative Incidence

Cumulative Incidence = Number of new cases of disease

Number in candidate population
during a specified period of time

• Cumulative incidence gives an estimate of risk

• Used mainly for fixed populations

 Example: Cumulative incidence of Sudden Infant
Death Syndrome during first year of life

Population 1,000 livebirths

Cases of SIDS 10
Cumulative Incidence 10/1,000 or 1% over one year

Note that all livebirths are ‘at risk’

• Proportion

•The period of time is expressed in words

 Cumulative Incidence
• Can use CI to look at lifetime risk of disease
– e.g., lifetime risk of developing breast cancer for
women is 1 in 8
• 1 of every 8 women will develop breast cancer in her
lifetime
• This means that 12.5% of women will develop breast
cancer at some point in their lives
 Age and Breast Cancer Risk
Age Cumulative Age Cumulative
risk risk
25 One in 19,608 60 One in 24

30 One in 2,525 65 One in 17

35 One in 622 70 One in 14

40 One in 217 75 One in 11

45 One in 93 80 One in 10

50 One in 50 85 One in 9

55 One in 33 Lifetime One in 8

As shown by these cumulative incidence figures, the

biggest single risk factor for breast cancer is age

SOURCE: National Cancer Institute SEER Cancer Statistics Review

 Cumulative Incidence
• CI assumption - you have followed the entire
population for the entire follow-up period
– Often not practical – you can’t always follow all people
over the time period as people move in and out of a
population or become lost
– Then length of follow-up won’t be the same for the whole
population
• Incidence rates do not make the assumption of
complete follow-up
 Incidence Rate
Incidence rate (IR) = # new cases of disease
person-time of observation
in candidate population

•True rate
•1/t (or t-1)
•Values of an incidence rate are
from 0 to infinity
•Person-time is the measure of
time
 Person-Time
• Only includes people who are still at-risk for
a disease
• Recorded only when someone is being
followed over time so it ends if:
– The person dies
– The person is lost to follow-up
– The person gets the disease
Accrual of Person Time
 Accrual of Person-Time

Jan Jan Jan

1980 1989 1999

Subject 1 ------------------x 10 Person-Years (PY)

Subject 2 ------------------x 10 PY

Subject 3 ------------------------------------ 20 PY

X = outcome of interest
 Some Ways to Accrue 100PY
• 100 people followed 1 year each = 100 py
• 10 people followed 10 years each= 100 py
• 50 people followed 1 year plus 25 people followed 2 years
= 100 py

Time unit for person-time = year, month or day

Person-time = person-year, person-month, person-day
 Incidence Rate
Incidence rate (IR) = # new cases of disease
person-time of observation
in candidate population

•True rate
•1/t (or t-1)
•Values of an incidence rate are
from 0 to infinity
•Person-time is the measure of
time
 Cohort study of the risk of breast cancer
among women with hyperthyroidism
• Followed 1,762 women ---> 30,324 py
• Average of 17 years of follow-up per woman
• Ascertained 61 cases of breast cancer
• Incidence rate = 61/30,324 py = .00201/y
= 201/100,000 py (.00201 x 100,000
p/100,000 p)
 Summary of CI and IR
• Measure frequency of disease
• Measure the transition from health to disease
• CI include the assumption of time
• IR includes the actual time
• IR may be more challenging to determine
• IR – best for dynamic populations
• CI – best for fixed population
 NU

Measuring Existing Cases of Disease

2
 Prevalence
• (P) Quantifies number of existing cases of disease in a
population at a point or during a period of time
• Two measures of Prevalence
– Point Prevalence
– Period Prevalence

• P = Number of existing cases of disease

Number in total population (at a point or during a
period of time)
 Point Prevalence
• e.g. City A has 7000 people with arthritis on
Jan 1st, 2014
• Population of City A = 70,000
• Prevalence of arthritis on Jan 1st = 0.10 or
10%

Number of existing cases of disease

During a point in time
Number in total population
 Period Prevalence
• Measures the proportion of the population
that has the disease during a period of time

Number of existing cases of disease

During a period of time
Number in total population
 Prevalence
• Point and Period Prevalence:
– numerator includes all living cases
– denominator includes everyone in the
population
– Prevalence is a proportion
• Values are from 0 to 1 or from 0% to 100%
 Review of Dimensions
Prevalence = people
people no dimension

Cumulative incidence = people

people no dimension

Incidence rate = people

person-time dimension is time –1
 Relationship Between Prevalence and Incidence

• P/(1-P) = IR * D
• P is the proportion of the population with the disease
• 1-P is the proportion of the population without the
disease
• IR is the incidence rate of the disease
• D is the duration of the disease
• Assuming a steady state population (equal number of
people entering and exiting during any unit of time in
population)
• P<0.1, then P ~ IR * D
 Picture It…

Prevalence Incidence

Cure or Death
 Relationship Between Prevalence and Incidence

• Prevalence depends on incidence and duration of disease

(duration lasts from onset of disease to its termination)
• If incidence is low but duration is long - prevalence is
relatively high
• If incidence is high but duration is short - prevalence is
relatively low

Prevalence
Incidence
Cure or Death
 AIDS

Source CDC
Figuring Duration from Prevalence and
Incidence

Lung cancer incidence rate = 45.9/100,000 py

Prevalence of lung cancer = 23/100,000 people

D = P = 23/100,000 p
IR 45.9/100,000 py =0.5 years

Conclusion: Individuals with lung cancer survive 6

months from diagnosis to cure or death
 NU

Other Measures in Epidemiology

3
 Crude Mortality Rate

Crude Mortality Rate (Crude Death Rate)

Calculated as: total # of deaths from all causes

# of people in the population

173,000 deaths from all causes in 2010 in Florida in a population of

18.8 million people, for a crude mortality rate of: 173,000 /18,800,000 = 916/100,000

In the US in 2010 there were 2.5 million deaths from all causes in a population of
308.7 million people, for a crude mortality rate of: 2,500,000 /308,700,000 = 799/100,000

Why was the rate higher for Florida?

 Other Mortality Measures

– Cause-specific Mortality
• # of deaths from a specific cause, per 100,000
population per year
• Denominator should be population at risk
– e.g., denominator for ovarian cancer mortality rate is all
women, not all population
– Age-specific Mortality
• # of deaths within an age group, per 100,000
population per year
Years of Potential Life Lost (YPLL)

– Measures premature mortality

– Death of a condition at a younger age reduces
the number of productive years
– To calculate:
• Age at death is subtracted from a predetermined age
at death (in the US, this is 65 years old)
• YPLL for each person is added up to give a total YPLL
for a condition
 Figure 4-10 Years of potential life lost (YPLL) before age 65 years among children younger
than 20 years from injuries and other diseases, United States, 1986. (Adapted from Centers
for Disease Control and Prevention: Fatal injuries to children: United States, 1986. MMWR
39: 442-451, 1990.)
 Livebirth Rate

Livebirth rate =

total number of livebirths

1,000 people (sometimes women of childbearing age) For one year

*Livebirth is defined a newborn who after separation breathes or shows

any sign of life
 Infant Mortality

Infant mortality :

# deaths of infants under 1 year of age For one year

1,000 live-births

• Can calculate neonatal deaths (occurring during the first 27 days after
livebirth)
• Can calculate post-neonatal deaths (occurring between day 28 and 12
months after livebirth)
 Congenital Anomaly Rate

Congenital anomaly rate (birth defects):

# children born with defects

10,000 births

• Numerator and denominator often include all births, not just livebirths
• Can calculate rate for all causes or cause-specific
 Attack Rate

Attack Rate =

# cases of disease that develop during defined period

# in population at risk at start of period

• Usually calculated for a short time period

• At-risk population had potential of exposure
 Case Fatality

Case fatality:

# of deaths___
for a defined period of time
# cases of disease

• This is a cumulative incidence measure so you specify a time period

• e.g., 5-year case fatality rate
•Also often used for infectious disease outbreaks
 Survival Rate

Survival rate:
# living cases for a defined period of time
# cases of disease

• This is a cumulative incidence measure so you specify a time period

• Other side of the case fatality rate
• This is often what is reported as prognosis of a disease, especially cancer
 Boston Globe Report on Celebrity
Skiing Deaths
“Although skiing has inherent risks, it isn’t more dangerous
than other common activities. The nationwide comparisons
below do not reflect differing numbers of participants.”

Is this statement accurate? What are the numbers in the

following charts? Incidence? Prevalence? Or something else?

Skiing deaths (one season)…………..………36

Skiing deaths (15 year average)……..………34
Accidental Deaths: How does skiing compare to
other sports? Remember to consider the missing
denominators when making your assessment.

Deaths in other sports

Parachuting…………………………………39

Scuba diving…………………………..…...104

Recreational boating………………………716

Drowning in swimming, boating

and water sports ……….………………...4,500
Bicycling ……………………………………800

SOURCE: News reports, National Ski Patrol

 NU

Q&A

2
NU

Screening in Public Health

Practice
Session 5

Hussain Alqhtani
BS.Pharm., M.Sc., MPH, Ph.D.
3
 What is Screening?
• Allows early detection of disease
• Followed by clinical diagnosis and treatment
• Useful if early detection will improve
treatment outcomes
 What is Screening?

Have disease Treatment

Likely to have disease Diagnostic tests

Screen asymptomatic population

Don’t have disease

Unlikely to have disease

 Natural History of Disease
Age (years)

40 Preclinical Stage
50 60

Pathological Clinical Remission Relapse Death

Onset Symptoms,
diagnosis, and
treatment
 Epi and Prevention
• Identification of high risk groups

Preventative
Efforts

Prevention
Screening
Resources Tailored to
Tests
that group
 Epi and Prevention
Identification
of high risk
groups

Risk Factors

Non-
Modifiable
Modifiable

Race Sex Diet Body size Exercise

Age
 Types of Prevention
• Prevention can be categorized into three types

Primary Immunization,
(Healthy person) avoiding
pollutants, dietary
modification,
exercise
Secondary
Screening (Preclinical
disease)
Tertiary Treatment,
(Clinical
disease) rehabilitation
 Types of Prevention vs. National
History of Disease
Age (years)

40 50 60

A B C D E
Pathological Clinical Remission Relapse Death
Onset Symptoms

Primary Secondary Tertiary

Prevention Prevention Prevention
 Which Diseases can be Screened?
• Characteristics of diseases that are good
candidates for screening tests:
– Serious illness, progressive disease
– Detectable preclinical phase (DPCP)
– DPCP is prevalent in the population with a long
duration
Characteristics of a Screening Test
• A screening test will be more effective if it is:
– Cost-effective
– Doesn’t hurt too much, isn’t too risky
– Acceptable to the population
– Valid and reliable
 Validity
• We screen for disease to determine who has
a pre-clinical disease
• The test should be positive for people who
do have a pre-clinical disease and negative
for those who don’t
• We can measure the validity of a screening
test as its sensitivity and its specificity
 Reliability
• Reliability means that you can administer the
test more than once and it will give the same
result for an individual
• Reliability is affected by:
– Changes in the physiologic condition of the
person being tested
– Technical aspects of the test
– Intraobserver agreement
– Interobserver agreement
False Positives and False Negatives
• False Positive test result – when the test indicates
someone has the disease but they actually don’t
– Lead to unnecessary treatment, worrisome, and lifestyle
change
– Stigma
• False Negative test result - when the test indicates
someone does not have the disease but they actually
do
– Failure to treatment
– Opt to no lifestyle change
– Spread of disease
Criterion of Positivity
Criterion of Positivity
 Sensitivity
• The probability that the test will correctly
identify a person with disease as being
positive
• A percentage (proportion)
Number of individuals with Pre-clinical Disease who test positive
X 100
Number of individuals with Pre-clinical Disease
 Specificity
• The probability that the test will correctly
identify a person without disease as being
negative

Number of individuals without Pre-clinical Disease who test negative

X 100
Number of individuals without Pre-clinical Disease
 The 2x2 Table…
• To calculate sensitivity and specificity you create a
2x2 table with your screening results
Population
Results of Disease No Disease Total
Screening

Positive a b a+b

Negative c d c+d

Total a+c b+d a+b+c+d

Sensitivity = a/a+c Specificity = d/d+b

 Validity of Screening Tests
True Characteristics in
the Population
Results of Screening Disease No Disease Total

Positive True Positives (TP) False Positives (FP) TP +FP

Negative False Negatives (FN) True Negatives (TN) FN + TN

Total TP + FN FP + TN TP + FP + FN + TN

Sensitivity = TP/TP+FN

Specificity = TN/TN+FP
 True Characteristics in
the Population
Results of Test Disease No Disease Total
Positive 80 100 180
Negative 20 800 820
Total 100 900
1,000

Sensitivity =

Specificity =
 True Characteristics in
the Population
Results of Test Disease No Disease Total
Positive 80 100 180
Negative 20 800 820
Total 100 900
1,000

Sensitivity = 80/100*100 = 80%

Specificity = 800/900*100 = 89%

 True Characteristics in
the Population
Results of Test Disease No Disease Total
Positive 80 100 180
Negative 20 800 820
Total 100 900
1,000
Sensitivity = 80/100*100 = 80%

Specificity = 800/900*100 = 89%

Sensitivity – The probability that the screening test can correctly identify people with
pre-clinical disease as positive

Specificity - The probability that the screening test can correctly identify people
without pre-clinical disease as negative
 Predictive Value of a Test
• Clinically, the question may be if a person tests
positive, how likely is it that they truly have the
disease?
– The Predictive Value Positive (PVP)* of a test tells us
what proportion of the patients who screen positive truly
have the disease
– The Predictive Value Negative (PVN)* of a test tells us
what proportion of patients who screen negative truly
don’t have the disease

* Some textbooks refer them as Positive Predictive Value (PPV)

and Negative Predictive Value (NPV)
 Predictive Value
• PVP - The proportion of individuals with a
positive test who have pre-clinical disease
Number of individuals with Pre-clinical Disease who test positive
PVP = X 100
Number of individuals who test positive

• PVN - The proportion of individuals without

pre-clinical disease who test negative
Number of individuals without Pre-clinical Disease who test negative
PVN = X 100
Number of individuals who test negative
 The 2x2 Table…
Population
Results of Disease No Disease Total
Screening

Positive a b a +b
Negative c d c+d
Total a+c b+d a+b+c+d

Sensitivity = a/a+c PVP = a/a+b or TP/TP+FP

Specificity = d/b+d PVN = d/c+d or TN/FN+TN

Predictive Value of a Test
True Characteristics in
the Population
Results of Test Disease No Disease Total
Positive 80 100 180
Negative 20 800 820
Total 100 900
1,000

Predictive Value Positive =

Predictive Value Negative =

Predictive Value of a Test
True Characteristics in
the Population
Results of Test Disease No Disease Total
Positive 80 100 180
Negative 20 800 820
Total 100 900
1,000

Predictive Value Positive = 80/180*100 = 44.4%

Predictive Value Negative = 800/820*100 = 97.6%

 Prevalence and PVP
Sensitivity = 99%, Specificity = 95%
Disease Test Results Sick Not Sick Totals PVP
Prevalence
+ 99 495 594
99/594 =
1% - 1 9,405 9,406
17%
Totals 100 9,900 10,000
+ 495 475 970
495/970 =
5% - 5 9,025 9,030
51%
Totals 500 9,500 10,000
Sensitivity, Specificity, and PVP
Disease
A.
+ -

+ 250 250 500

Test

- 250 250 500

500 500 1000

Prevalence = 50%
Sensitivity = 50%
Specificity = 50%
PVP = 250/500 = 50%
 A. Disease B. Disease
+ - + -

+ 250 250 500 + 100 400 500

Test Test

- 250 250 500 - 100 400 500

500 500 1000 200 800 1000

Prevalence = 50% Prevalence = 20%
Sensitivity = 50% Sensitivity = 50%
Specificity = 50% Specificity = 50%
PVP = 250/500 = 50% PVP = 100/500 = 20%
 Disease Disease
A. B.
+ - + -

+ 250 250 500 + 100 400 500

Test Test

- 250 250 500 - 100 400 500

500 500 1000 200 800 1000

Prevalence = 50% Prevalence = 20%
Sensitivity = 50% Sensitivity = 50%
Specificity = 50% Specificity = 50%
PVP = 250/500 = 50% PVP = 100/500 = 20%

Disease
C. + -

+ 180 400 580

Test

- 20 400 420

200 800 1000

Prevalence = 20%
Sensitivity = 90%
Specificity = 50%
PVP = 180/580 = 31%
 Disease Disease
A. B.
+ - + -

+ 250 250 500 + 100 400 500

Test Test

- 250 250 500 - 100 400 500

500 500 1000 200 800 1000

Prevalence = 50% Prevalence = 20%
Sensitivity = 50% Sensitivity = 50%
Specificity = 50% Specificity = 50%
PVP = 250/500 = 50% PVP = 100/500 = 20%

Disease Disease
C. + - D. + -

+ 180 400 580 + 100 80 180

Test Test

- 20 400 420 - 100 720 820

200 800 1000 200 800 1000

Prevalence = 20% Prevalence = 20%
Sensitivity = 90% Sensitivity = 50%
Specificity = 50% Specificity = 90%
PVP = 180/580 = 31% PVP = 100/180 = 56%
 Sensitivity, Specificity, and PVP
• Remember that specificity deals with people who
do not have disease while sensitivity deals with
people who do have disease
• Specificity affects PPV more than sensitivity does
because most people in the population do not have
the disease
 Expressing Prognosis
• Prognosis is important to:
– Establish disease control and treatment
priorities
– Allow the patient to understand what to expect
from the disease
– Allow comparison of the disease before and
after treatments/interventions
Natural History of Disease

The natural history of disease in a patient.

Downloaded from: StudentConsult

© 2005 Elsevier
 Expressing Prognosis
• When do you start calculating survival time?
– Time of onset of disease?
• Not usually known
– Time of onset of symptoms?
• Not always know; subjective
– Time of diagnosis?
• Different people seek care at different times
• Diagnosis relies on how good diagnostic test is
• Some people die before they are diagnosed
 Expressing Prognosis
• Prognosis can be expressed in terms of
survival with disease or in terms of death
from disease
• Could also express time to recurrence of
disease, time to disability from disease
• For now, we are talking about expressing
prognosis from diagnosis to death
 Case-Fatality Rate
• Number of people who die from a disease
divided by the number of people who have a
disease
• Caveats:
– Usually applied to acute diseases/outbreaks
– There is typically no statement of time
– Different from risk of death or mortality rate
 Person-Years
• Number of deaths divided by the person-
years over which the group was observed
• Caveat: Not all person-years are equal
Figure 6-3 Two examples of 10 person-years: five
people, each observed for 2 years, and two people,
each observed for 5 years.
 Survival or Survival Rate
• Five-Year Survival Rate
– The percentage of people diagnosed who are alive five
years after diagnosis or 5 years after treatment has
begun
– Often used in cancer prognosis
• Disease-free survival rate
• Progression-free survival rate

Source: Cancer Research UK

 Survival Time

Figure 6-9 The problem of 5-year survival in a screened population:

I. No screening test available.
II. Earlier disease detection by screening.
 Survival Time
• Screening tests may actually lead to longer
survival because of earlier treatment
Or…
• Survival may appear longer simple because
diagnoses were made earlier, not because
actually lived longer, i.e., lead-time bias
 Lead Time Bias

Figure 18-8 Possible outcomes of a screening program.

 Length-Bias Sampling

Figure 18-6 Short and long natural histories of disease: relationship of length of clinical phase to length of preclinical
phase.
 Length-Bias Sampling
• Cases with a long preclinical course are
more likely to have a long clinical course and
better survival
• Screening tends to identify these cases,
making the program appear to have a
beneficial effect on survival
 Mortality
• Overall Mortality Rate
– Pros: Not affected by lead-time bias or length-
bias sampling
– Cons: Insensitive if the disease accounts for only
a small proportion of deaths. Ex. prostate cancer
• Disease-Specific Mortality Rate
– Best for evaluating a screening program
– Cons: Not appropriate for diseases that do not
lead to death
 Volunteer Bias
• This type of bias relates to the people who
are screened
– Are the people who are screened similar to
those who are not screened?
• Some people who are generally healthier volunteer
for screening and this can affect the outcome
• Some people who are at higher risk of disease
volunteer for screening and this can also affect the
outcome
 What is Screening?
• Allows early detection of disease
• Followed by clinical diagnosis and treatment
• Useful if early detection will improve
treatment outcomes
 NU

Q&A

4
NU

Disease Transmission
Session 6

Hussain Alqhtani
BS.Pharm., M.Sc., MPH, Ph.D.
5
 Modes of Transmission
• Key factors needed for the spread of disease
– Source
– Host (person who is able to get the disease)
– Methods of transmission
• Contact transmission
• Vector transmission
• Airborne transmission
• Vehicle transmission
"Water Source" from Culture, Politics, and Community:
Living Public Health in Nigeria. Available at:
http://ocw.jhsph.edu. Copyright © Johns Hopkins
Bloomberg School of Public Health. Creative Commons BY-
NC-SA. Photo by Bill Brieger.
Modes of Transmission
Transmission

Direct Indirect

Person-to-
Air
person

Fomite

Food/Water
 Direct Contact: Examples
• Sexually transmitted diseases
• Touching, kissing
• Vertical transmission

https://aidsinfo.nih.gov/understanding-hiv-
aids/glossary/738/vertical-transmission
 Indirect Contact: Examples
• Air (tuberculosis, influenza)
• Food, water (Salmonella)
• Objects (MRSA or methicillin-resistant
Staphylococcus aureus)
• Animal or insect (West Nile virus, malaria,
Lyme disease)
CDC Public Health Image Library

Anopheles mosquito (Malaria)

 Modes of Transmission
• All organisms have different characteristics
that affect their spread
– Growth rate
– Survival in the environment
– Mode of entry/transmission
– Mode of exit/transmission
The Epidemiology Triad

Source
 Modes of Transmission

"Portals of Exit and Entry" from Fundamentals of

Epidemiology I . Available at: http://ocw.jhsph.edu.
Copyright © Johns Hopkins Bloomberg School of Public
Health. Creative Commons BY-NC-SA.
 Disease Severity

Preclinical Clinical Disease Outcome

Disease • Signs/symptoms • Cure, control,
• No signs or are present disability, death
symptoms yet

Subclinical
Disease
 Disease Severity
Carrier Status

Actively
Colonized
Infected

Has signs and No sings or

symptoms symptoms

Can transmit Can Transmit

 Disease Outbreaks

Levels of
Disease

Endemic Epidemic Pandemic

Downloaded from: StudentConsult (on 24 August 2010 09:05 PM)
© 2005 Elsevier
Determinants of Disease Outbreaks

Types of
Susceptibility to
Disease

Immune Susceptible
Determinants of Disease Outbreaks
• Immune status applies to individuals as well
as communities
– Low vaccination rate – many susceptible people
and higher likelihood of disease
– High vaccination rate – few susceptible people
and lower likelihood of disease
Determinants of Disease Outbreaks
• Key points to Community Immunity (i.e., Herd
Immunity)
– Works for direct transmission
– Must be a disease that only has one host
– Disease or immunization must confer full
protection
– Works best when communities randomly interact
Determinants of Disease Outbreaks
• Measles Outbreak
– Spring of 2007
– Alachua County, FL
– Student traveled to India and contracted
measles (not vaccinated)
– Was part of a religious community, many
members were not vaccinated
– Four other cases occurred in that community
Effect of Herd Immunity
 Incubation Period
• Period of time between infection and
development of disease signs and symptoms
• Differs for every disease
• Variable
– e.g., Chickenpox is usually 14-16 days
Incubation periods of viral diseases. (From Evans AS, Kaslow RA [eds]: Viral Infections of Humans: Epidemiology
and Control, 4th ed. New York, Plenum, 1997.)

Downloaded from: StudentConsult (on 24 August 2010 09:05 PM)

© 2005 Elsevier
 NU

Q&A

6
 NU

T h a n k Yo u !