Epidemiology of Infectious Diseases

1
Epidemiology of Infectious Diseases

• Catherine Diamond MD MPH
• UCI Infectious Diseases Epidemiology Divisions
• [phone]
• diamondc_at_uci.edu

2
Surveillance

• Routine collection, analysis reporting of data
  monitoring morbidity mortality trends for
  public health purposes
• Active agency actively contacts hospitals/MDs on
  a regular basis to find cases
• Passive agency receives reports from
  hospitals/MDS
• Sentinel systems certain MDs or hospitals report
  designated cases in order to catch trends early

3
Trends

• Secular trend a change in the prevalence of
  infection over years, usually due to better
  living conditions, better hygiene vaccination,
  e.g. decrease in TB in US
• Seasonal trend refers to changes in the
  prevalence of infection occurring over the year,
  e.g. influenza outbreaks. Changes of
  temperature, crowding, humidity may play a role

4
Endemic vs. Epidemic vs. Pandemic

• Endemic infection-infection or disease that
  occurs regularly at low to moderate frequency
• Epidemics occur when there are sudden increases
  in frequency in frequency above endemic levels
• Pandemics are global epidemics, an epidemic
  occurring over a wide area usually affecting a
  large portion of the population. The size of
  outbreaks is dependent on factors such as the
  ratio of susceptible to immune subjects, period
  of infectivity, population density etc

5
What is an Outbreak?

• The occurrence of cases of an illness clearly in
  excess of expectancy
• Usually compare current number of cases with the
  number over a comparable period sometime in the
  past an increase in the number of cases over past
  experience for a given population, time place
• Must take into account seasonal variations in
  disease incidence for diseases such as influenza

6
Three Common Conditions for the Occurrence of an
Epidemic or Outbreak

• The introduction a new pathogen, or the increased
  amount of or change in virulence of a known
  pathogen from an infected human, animal, bird or
  arthropod vector or from air, water, food, soil,
  drug or other environmental source
• An adequate number of exposed susceptible
  persons
• An effective means of transmission between the
  source of the pathogen the susceptible persons

7
Why Should We Study Outbreak Investigations?

• Outbreaks are important public health events
• Outbreaks are experiments of nature allow the
  opportunity to learn more about the natural
  history of disease
• Outbreaks represent breakdowns in public health
• You might end up investigating an outbreak
  someday
• Outbreaks are real life examples of practical
  applications of epidemiologic methods
• Outbreaks are interesting

8
What Are the Steps in a Outbreak Investigation?

• Verify the diagnosis
• Confirm the existence of an outbreak
• Identify count cases (establish case
  definition)
• Orient data in terms of time, place person
• Formulate test hypotheses
• Identify implement control measures

9
Steps in an Outbreak Investigation

• Different investigators may have slightly
  different lists of steps
• The logistics of preparing for an outbreak
  investigation could be considered as one of the
  steps
• The steps are not necessarily carried out exactly
  in the order listed

10
Step 1 Verify the Diagnosis

• Is it an epidemic? Be certain it is real not a
  false alarm ( pseudoepidemic)
• In hospital setting, pseudoepidemics may be due
  to false positive diagnosis resulting from
  contamination e.g. environmental contamination of
  specimens during laboratory processing
• In contrast, there may also be artifactual
  clustering of real cases, e.g. change in
  reporting due to change in diagnostic methods,
  new physician/clinic in town, changes in
  local/national awareness. HIV/AIDS examples

11
Step 2 Confirm the Existence of an Outbreak

• Compare the magnitude of the current problem with
  baseline
• Problems with determining baselines
• Lack of data
• Varying or no case definition
• Incomplete reporting, lack of surveillance

12
Step 3 Identify Count Cases

• Case definition usually specifies a person with
• Some set of symptoms or signs or laboratory
  diagnosis
• Occurring in some time period
• In some specific setting
• Remember spectrum of disease (you may want to
  include asymptomatic/subclinical cases since
  information about them may be crucial to
  investigation)

13
Secondary Cases

• Consider whether you want to include secondary
  cases.
• Secondary cases are persons who were infected as
  a result of exposure to a primary case
• E.g. in a food borne outbreak of E coli 0157 H7
  primary cases were infected by consumption of
  contaminated hamburger secondary cases would be
  infected through exposure to a primary case (e.g.
  in a day care)
• These secondary cases were not exposed to the
  source of the outbreak their inclusion in the
  risk factor analysis would tend to bias toward
  the null
• In many cases it is desirable to look for
  secondary cases collect information on them but
  not include them in the primary analysis

14
Evolving Case Definitions

• Patients may not have laboratory tests because
  the tests are expensive, difficult to obtain or
  clinically not necessary
• Early on, investigators use a loose case
  definition which includes confirmed, probable
  even possible cases
• Later a tighter case definition may increase the
  ability to detect a true association
• Ideally your case definition will include most if
  not all of the actual cases very few or no
  false positive cases

15
Definite vs. Probable vs. Possible Cases

• Definite case laboratory confirmation
• E.g. E coli 0157H7 isolated from stool culture
  in a resident of the county with onset of
  symptoms during a specified time period
• Probable case typical clinical features without
  laboratory confirmation
• Bloody diarrhea with same person, place time
  restrictions
• Possible case fewer of the typical clinical
  features
• Abdominal cramps diarrhea with the same person,
  place time restrictions

16
Identify Cases

• 1. Conduct a systematic search
• Cast a wide net regarding geography population
• The original cases may or may not be
  representative of the true extent of the problem
• 2. Use multiple sources which may include
• Medical systems hospitals, laboratories,
  physician's office, clinics
• Surveillance data
• Media/press announcements
• Special surveys e.g. if outbreak involved a
  defined population such as persons on a cruise
  ship you could survey that entire population

17
Step 4 Orient Data in Terms of Time, Place
Person

• Characterize the cases in terms of time, place
  person
• Time draw epidemic curves
• Place construct spot maps
• Person compare groups

18
Time Outbreaks The Epidemic Curve

• An epidemic curve is a graph of the distribution
  of cases according to time of onset. From the
  curve you can tell
• Where you are in the time course of an epidemic
  (e.g. beginning, middle, end)
• From the pattern of the curve, you may be able to
  draw inferences about the mode of spread of the
  causative agent ( e.g. person-to-person, common
  source)

19
Common Source or Point Outbreaks

• Common source or point outbreaks refer to the
  exposure of a susceptible population to a common
  source of a pathogen often at the same time, such
  as at a church picnic or neighborhood restaurant
• These most frequently result in a short, sharp
  epidemic curve with cases clustered around single
  defined peak value. If the agent is
  transmissible to other by person to person
  contact then secondary peaks may occur

20
(No Transcript)
21
(No Transcript)
22
Common Source or Point Outbreaks

• For a common source exposure, if you have
  identified the disease know its usual
  incubation period (the time interval between
  exposure to an infectious agent the appearance
  of the first signs or symptoms of disease) , you
  can estimate a probable time period of exposure
  use that information to focus your investigation
• The minimum incubation time should correspond to
  the interval between exposure the first case
• The average incubation period should correspond
  to the interval between exposure the peak of
  the outbreak or the time occurrence of the median
  case

23
(No Transcript)
24
Types of Epidemic Curves

• The commercial distribution of food through many
  states or a chain of stores has created a new
  form of common source epidemic in which the time
  pattern of delivery, purchase consumption in
  a local area define the nature of the epidemic
• If exposure occurs over different times, the
  epidemic curve can spread out continuously

25
Figure 1. Outbreak-Associated Confirmed Cases of
S. enteritidis Infection in Minnesota in
September and October 1994, According to the Date
of Onset. One hundred fifty cases were reported.
26
Propagative Epidemic Curves

• Propagative or progressive epidemic curves result
  from epidemics involving the spread of a pathogen
  from one susceptible individual to another e.g.
  measles, influenza-frequently occur in
  propagative epidemics
• Curve with some clusters of irregular peaks
  somewhat spread out is consistent with person to
  person spread
• Mixed epidemics involve both a single, common
  exposure to an infectious agent secondary
  propagative spread to other individuals e.g. many
  food borne pathogens (Salmonella, Hepatitis A)
  airborne organisms (TB)
• Mixed type of curves such as a single large peak
  followed by successive smaller peaks may be seen
  when a common source outbreak occurs followed by
  secondary person to person spread

27
(No Transcript)
28
Place of Outbreak Spot Maps

• Demonstrate geographic extent of the problem
• Demonstrate cluster or pattern illustrating where
  cases live, work or may have been exposed
• Most famous example of a spot map is John Snows
  spot map of the distribution of cholera cases
  around the Broad Street pump

29
Distribution of Cholera Deaths in Golden Square
Area of London August-September 1848
30
(No Transcript)
31
Characterize Cases

• Characterize the cases completely
  systematically by developing a questionnaire
  before the patients are contacted. Collect
• Identifying information Name, address, telephone
  number
• Demographic information Age, sex, race,
  occupation
• Clinical information symptoms, date of onset,
  medical evaluation
• Risk factor information depends on the disease
  being investigated

32
Personal Characteristics Outbreaks

• The case group must be thoroughly described in
  terms of age, race, sex, occupation, diagnosis
  other factors
• Rates are then calculated to identify high risk
  groups
• E.g. age, sex (HIV), occupation (HCW)

33
Step 5 Formulate Test Hypotheses

• Generate hypotheses to explain outbreak
• Usually conduct a case control or cohort study
• Consider evidence for causation

34
Cohort vs. Case Control Study Design

• In cohort study, you have knowledge of entire
  population (e.g. can count how many individuals
  were exposed and how many were infected)
• Can compare attack rate in the exposed and
  unexposed
• Can calculate RR (incidence in exposed/incidence
  in unexposed)
• But often only have information regarding some of
  the population and then need to use a case
  control design
• Can compare proportion of cases and proportion of
  controls eating food item
• Can calculate OR (AD/ BC)

35
(No Transcript)
36
Coccidiodomycosis Outbreak Following Northridge
Earthquake

• 39 of cases vs. 17 of controls reported
  physically being in a dust cloud OR 3.0 (95 CI
  1.6-54). Duration of time spent in dust cloud
  correlates with OR
• 0 minutes OR 1.0
• 1-15 minutes OR 1.7
• 16-30 minutes OR 3.0
• gt30 minutes OR 5.2

37
Causation

• Strength of the association. The stronger the
  association, the more likely it is real
• Consistency with other studies. A consistent
  finding is more plausible
• Exposure precedes disease
• Dose-response effect (not mandatory but adds
  credibility)
• Biologic plausibility
• Removal of agent decreases or eliminates disease

38
Attack Rate

• Attack rate is an incidence rate
• Used when occurrence of disease among a
  population at risk increases greatly over short
  period of time, often related to a specific
  exposure
• The disease rapidly follows the exposure during a
  fixed time period
• Often used for food borne illness
• Attack rate ill X 100 during a time
  period
• (ill well)

39
(No Transcript)
40
Secondary Attack Rate

• Secondary attack rate yields an index of the
  spread of disease within a household or similar
  circumscribed unit
• The secondary attack rate is used to measure
  infectivity the capacity of the agent to enter
  multiply in a susceptible host thus produce
  infection or disease

41
Case Fatality Rate

• Case fatality rate the number of deaths caused
  by a disease among people who have the disease
• An index of the deadliness of a particular
  disease
• CFR of deaths due to disease X X 100
• of cases due to disease X

42
Step 6 Implementing Control Prevention Measures

• Eliminating or treating the source-e.g. removing
  an infected foodhandler from work treating
  appropriately
• Cohorting patients- a common approach in
  hospitals, childcare other institutional
  setting
• Preventing further exposures-e.g. as in the
  HIV/AIDS epidemic through educational efforts to
  change knowledge behaviors
• Protecting the population at risk- e.g. through
  vaccination
• System changes. For example, changing the method
  by which meat inspection is conducted to decrease
  the risk of contamination with E coli 0157H7

43
Use of Molecular Subtyping in Infectious Disease
in Infectious Disease Outbreak

• Molecular subtyping of patient source outbreaks
  has become an increasingly important part of
  outbreak investigations
• Many methods of subtyping including pulsed field
  gel electrophoresis (PFGE) restriction fragment
  length polymorphism (RFLP)
• Within a species of infectious agents, there are
  multiple strains with different genetic
  compositions. Subtyping techniques attempt to
  determine the degree of genetic relatedness of
  different isolates
• Outbreaks are nearly always caused by a single
  strain of the causative organism thus termed
  clonal
• Subtyping of isolates can be used to determine
  whether the isolated involved are closely related
  are therefore likely to be associated with a
  common source
• The results of subtyping can be applied to the
  case definition
• Subtyping is usually not immediately available
  thus is not used in the initial case definition
  but may be incorporated in later analysis

44
Case Scenario

• You are the county epidemiologist in a county on
  the Pennsylvania/Ohio border
• Between November 13 December 3, 26 cases of
  hepatitis are reported to your county health
  department
• Although you just started working there, this
  seems like a lot to you! You decide to
  investigate

45
Step 1 Verify the Diagnosis

• How would you verify the diagnosis of Hepatitis A?

46
Step 1 Verify the Diagnosis

• Review clinical laboratory features of the
  cases to determine if they are consistent with
  the diagnosis of hepatitis A

47

• HEPATITIS A CLINICAL FEATURES

• Jaundice by lt6 yrs lt10
  age group 6-14 yrs
  40-50
  gt14 yrs 70-80
• Rare complications Fulminant hepatitis
  Cholestatic hepatitis
  
  Relapsing hepatitis
• Incubation period Average 30 days
  Range 15-50
  days
• Chronic sequelae None

48
EVENTS IN HEPATITIS A VIRUS INFECTION
Clinical illness
Infection
ALT
IgM
IgG
Viremia
Response
HAV in stool
0
1
2
3
4
5
6
7
8
9
10
11
12
13
Week
49
Step 2 Confirm the Existence of an Outbreak

• How would you confirm the existence of an
  outbreak?

50
Step 2 Confirm the Existence of an Outbreak

• Compare the number of cases against historical
  numbers
• Between November 13 December 3, 26 cases of
  Hepatitis A were reported to the county health
  department compared with 1 case in the previous 4
  months
• FYI in Orange County in 2002, there were 91
  reported cases of hepatitis A (3.1 cases per
  100,000 population)

51
Step 3 Identify Count Cases

• What would be your case definition?
• How would you identify cases?
• Would you include secondary cases?

52
ACUTE HEPATITIS A CASE DEFINITION FOR
SURVEILLANCE

• Clinical criteriaAn acute illness with
• discrete onset of symptoms (e.g. fatigue,
  abdominal pain, loss of appetite, intermittent
  nausea, vomiting),
• jaundice or elevated serum aminotransferase
  levels
• Laboratory criteria
• IgM antibody to hepatitis A virus (anti-HAV)
  positive
• Case Classification
• Confirmed. A case that meets the clinical case
  definition is laboratory confirmed or a case
  that meets the clinical case definition occurs
  in a person who has an epidemiologic link with a
  person who has laboratory-confirmed hepatitis A
  (i.e., household or sexual contact with an
  infected person during the 15-50 days before the
  onset of symptoms)

53
Step 3 Identify Count Cases

• Case definition defined as a person with
  discrete symptom onset between November 13
  December 4 in association with the presence of
  IgM antibody to HAV in your county

54
Step 3 Identify Count Cases

• Identify cases
• State health department requests counties to
  immediately report all HAV cases
• The county contacts all county physicians,
  hospitals, laboratories neighboring county
  state health departments to rapidly report cases
• Advertisements/media/signs
• Secondary cases might bias toward the null could
  collect early on but might not include in analysis

55
Step 4 Orient Data in Terms of Time, Place
Person

• What would the epidemic curve look like?
• What would a spot map look like?
• Are there specific demographic or other
  characteristics of the cases?

56
(No Transcript)
57
Step 4 Orient Data in Terms of Time, Place
Person

• A spot map would likely show more cases in the
  geographic area nearest the restaurant
• Presumably cases would reflect the demographics
  of restaurant goers, patients with symptomatic
  hepatitis A residents of the county
• 22 (51) were male
• Median age 34 years (range 5-66)
• All were white
• 14 (33) hospitalized
• No travel, injection drug use or male-male sex

58
Step 5 Formulate Test Hypotheses

• What do you hypothesize is the cause of the
  outbreak?
• How would you collect data to prove this
  hypothesis?
• How would you analyze data to validate your
  hypothesis?

59
Step 5 Formulate Test Hypotheses

• Infected food product vs. infected employee
• The CDC Viral Hepatitis Surveillance Program
  Questionnaire
• Case control study
• Genetic relatedness of hepatitis A sequence.

60
(No Transcript)
61
(No Transcript)
62
CONCENTRATION OF HEPATITIS A VIRUS IN VARIOUS
BODY FLUIDS
Feces
Serum
Body Fluids
Saliva
Urine
102
104
100
106
108
1010
Infectious Doses per mL
Source Viral Hepatitis and Liver Disease
19849-22 J Infect Dis 1989160887-890
63
HEPATITIS A VIRUS TRANSMISSION

• Close personal contact(e.g., household contact,
  sex contact, child day-care centers)
• Contaminated food, water(e.g., infected food
  handlers)
• Blood exposure (rare)(e.g., injection drug use,
  rarely by transfusion)
• No risk factor identified for 40-50 of cases

64
GLOBAL PATTERNS OF HEPATITIS A VIRUS TRANSMISSION
Disease Rate
Peak Age of Infection
Transmission Patterns
Endemicity
Low to high
High
Early childhood
Person to person
outbreaks uncommon
Late childhood/ young adults
High
Moderate
Person to person
food waterborne
outbreaks
Low
Young adults
Low
Person to person
food waterborne
outbreaks
Very low
Very low
Adults
65
RISK FACTORS ASSOCIATED WITH REPORTED HEPATITIS
A, 1990-2000, UNITED STATES
Source NNDSS/VHSP
66
Cause of the HAV Epidemic

• Green onions grown processed in Mexico then
  shipped on ice to US restaurant where they are
  chopped and placed raw in giant vats of mild
  salsa
• Possible contact with HAV-infected workers
  especially children working in the field during
  green onion harvesting/preparation
• Possible contact with HAV-contaminated water
  during irrigation, rinsing, icing

67
Step 6 Implementing Control Prevention Measures

• What control measures would you implement?
• How would you prevent future cases?

68
Step 6 Implementing Control Prevention Measures

• Control
• Immunoglobulin
• Close restaurant
• Prevention
• Vaccination
• Public health announcements to avoid raw green
  onions
• Agricultural quality control
• Water quality for irrigation
• Provide sanitary facilities for field workers
• Child-care for field workers
• Prevent HAV transmission

69
PREVENTING HEPATITIS A

• Hygiene (e.g., hand washing)
• Sanitation (e.g., clean water sources)
• Hepatitis A vaccine (pre-exposure)
• Immune globulin (pre- post-exposure)

70
HEPATITIS A PREVENTION IMMUNE GLOBULIN

• Pre-exposure
• travelers to intermediate high HAV-endemic
  regions
• Post-exposure (within 14 days)
• Routine
• household other intimate contacts
• Selected situations
• institutions (e.g., day-care centers)
• common source exposure (e.g.,
• food prepared by infected food handler)

71
HEPATITIS A VACCINES

• Highly immunogenic
• 97-100 of children, adolescents, adults have
  protective levels of antibody within 1 month of
  receiving first dose essentially 100 have
  protective levels after second dose
• Highly efficacious
• In published studies, 94-100 of children
  protected against clinical hepatitis A after
  equivalent of one dose

72
Hepatitis A Incidence, United States, 1980-2002
2002 rate provisional
73
1987-97 average incidence
Hepatitis A Incidence
2002 incidence
74
Conclusion

• Step 1 Verify the diagnosis
• Step 2 Confirm the existence of an outbreak
• Step 3 Identify count cases
• Step 4 Orient data in terms of time, place
  person
• Step 5 Formulate test hypothesis
• Step 6 Implementing control prevention measures

75
Acknowledgements

• Lisa Jackson, MD MPH University of Washington for
  lecture outline
• HAV slides from CDC website www.cdc.gov
• Dentinger et. al. An Outbreak of Hepatitis A
  Associated with Green Onions. J. Infect Dis 2001
  183 1273-6.
• CDC. Hepatitis A Outbreak Associated with Green
  Onions at a Restaurant-Monaca, PA, 2003. MMWR
  2003 52 1155-1157.