Transcript Environmental Health for Microbial Agents

Microbial Risk Assessment
Envr 133
Mark D. Sobsey
Spring, 2006
Definition of Quantitative Microbial
Risk Assessment
Applications of the principles of risk
assessment to the estimation of the
consequences from anticipated or
actual exposure to infectious
microorganisms
Relationship Between Exposure, Level
of Protection and Microbial Risk
= Confidence Region or Interval
Risk 
Exposure 
 Level of Technological Control
Some Differences Between
Chemical and Microbial Risks
• A single microbe (one unit) is infectious
• Microbes multiply:
– In a host
– In environmental media (some)
• Secondary spread
– Microbe infects a host from an environmental
route of exposure (water, food, etc.) can
– Then, it spreads to other hosts by person-toperson transmission
• Some microbes cause a wide range (spectrum)
of adverse effects
(Adapted from: National Academy of Sciences - National Research Council framework)
RISK ASSESSMENT FOR ENVIRONMENTALLY
TRANSMITTED MICROBIAL PATHOGENS:
ILSI/EPA PARADIGM
PROBLEM FORMULATION: HAZARD IDENTIFICATION
CHARACTERIZATION
OF EXPOSURE
EFFECTS
CHARACTERIZATION OF
HUMAN HEALTH
RISK CHARACTERIZATION
Risk Management
ILSI/EPA Risk Assessment Framework and Steps:
Analysis Phase
(Adapted from: National Academy of Sciences - National Research Council framework)
Conducting Hazard Identification
• Identify microbes as causative agent of disease
• Develop/identify diagnostic tools to identify symptoms, infection
and to isolate and identify causative microbe in host specimens
• Understand the disease process from exposure to infection,
illness (pathophysiology) and death
• Identify transmission routes
• Assess virulence factors and other properties of the microbe
responsible for disease, including life cycle
• Identify and apply diagnostic tools to determine incidence and
prevalence in populations and investigate disease outbreaks
• Develop models (usually animals to study disease process and
approaches to treatment
• Evaluate role of immunity in overcoming/preventing infection and
disease and possible vaccine development
• Study epidemiology associated with various exposures
(Adapted from: National Academy of Sciences - National Research Council framework)
Exposure Assessment
• Purpose: determine the dose
• Dose = number, quantity or amount
of microorganisms corresponding to
a single exposure (e.g., by ingestion)
• Average or typical dose
– a measure of central tendency
• mean or median
• Distribution of doses
Described as a probability or frequency
distribution; “probability density function”
CHARACTERIZATION OF EXPOSURE
ELEMENTS INCLUDED IN PATHOGEN CHARACTERIZATION:
OCCURRENCE (previous lecture)
• Temporal distribution, duration and frequency
• Concentration in food or environmental media
• Spatial distribution
– clumping, aggregation, particle-association, clustering
• Niche
– ecology and non-human reservoirs
– potential to multiply/survive in specific foods or media
•
•
•
•
Survival, persistence, and amplification
Seasonality
Meteorological and climatic events
Presence of control or treatment processes
– including their reliability and variability
• Indicators/surrogates for indirect evaluation
– predictive of pathogen
ELEMENTS CONSIDERED IN PATHOGEN
CHARACTERIZATION (previous lecture)
•
•
•
•
•
•
•
•
•
Virulence and pathogenicity of the microorganism
Pathologic characteristics and diseases caused
Survival and multiplication of the microorganism
Resistance to control or treatment processes
Host specificity
Infection mechanism and route; portal of entry
Potential for secondary spread
Taxonomy and strain variation
Ecology and natural history
Pathogen Characteristics or Properties Favoring
Environmental Transmission (previous lecture)
Multiple sources and high endemicity in
humans, animals and environment
– High concentrations released into or present in
environmental media (water, food, air)
– High carriage rate in human and animal hosts
– Asymptomatic carriage in non-human hosts
– Ability to proliferate in water and other media
– Ability to adapt to and persist in different media
or hosts
– Seasonality and climatic effects
– Natural and anthropogenic sources
Pathogen Characteristics or Properties Favoring
Environmental Transmission (previous lecture)
• Ability to Persist or Proliferate in Environment and Survive
or Penetrate Treatment Processes
• Stable environmental forms
– spores, cysts, oocysts, stable outer viral layer (protein coat),
capsule, etc.
• Resistance to biodegradation, heat, cold (freezing), drying,
dessication, UV light, ionizing radiation, pH extremes, etc.
• Resists proteases, amylases, lipases and nucleases
– Posses DNA repair mechanisms and other injury repair
processes
• Colonization, biofilm formation, resting stages, protective
stages, parasitism
– Spatial distribution
– Aggregation, particle association, etc.
Virulence Properties of Pathogenic Bacteria Favoring
Environmental Transmission (previous lecture)
Virulence properties: structures or chemical constituents
that contribute to pathophysiology:
– Outer cell membrane of Gram negative bacteria:
endotoxin (fever producer)
– Exotoxins
– Pili: for attachment and effacement to cells and
tissues
– Invasins: to facilitate cell invasion
– Effacement factors
Spores:
– highly to physical and chemical agents and
– very persistent in the environment
Others:
– plasmids, lysogenic bacteriophages, etc.
Pathogen Characteristics or Properties Favoring
Environmental Transmission (previous lecture)
Genetic properties favoring survival and pathogenicity
• Double-stranded DNA or RNA
• DNA repair
• Ability for genetic exchange, mutation and selection
– recombination
– plasmid exchange, transposition, conjugation, etc.
– point mutation
– reassortment
– gene expression control
• Virulence properties: expression, acquisition,
exchange
• Antibiotic resistance
Role of Selection of New Microbial Strains in
Susceptibility to Infection and Illness (previous lecture)
• Antigenic changes in microbes overcome immunity, increasing
risks of re-infection or illness
– Antigenically different strains of microbes appear and are
selected for over time and space
– Constant selection of new strains (by antigenic shift and drift)
– Partly driven by “herd” immunity and genetic recombination,
reassortment , bacterial conjugation, bacteriophage infection and
point mutations
• Antigenic Shift:
– Major change in virus genetic composition by gene substitution or
replacement (e.g., reassortment)
• Antigenic Drift:
– Minor changes in virus genetic composition, often by mutation
involving specific codons in existing genes (point mutations)
• A single point mutation can greatly alter microbial
virulence
Pathogen Characteristics or Properties Favoring
Environmental Transmission (previous lecture)
Ability to Cause Infection and Illness
• Low infectious dose
• Infects by multiple routes
– ingestion (GI)
– inhalation (respiratory)
– cutaneous (skin)
– eye
– etc.
Microbe Levels in Environmental Media Vary Over Time Occurrence of Giardia Cysts in Water: Cumulative Frequency Distribution
Previous lecture
CHARACTERIZATION OF EXPOSURE
ELEMENTS CONSIDERED IN EXPOSURE ANALYSIS
(previous lecture)
• Identification of water, food or other
media/vehicles of exposure
• Units of exposure
• Routes of exposure and transmission potential
• Size of exposed population
• Demographics of exposed population
• Spatial and temporal nature of exposure
(single or multiple; intervals)
• Behavior of exposed population
• Treatment, processing, and recontamination
(Adapted from: National Academy of Sciences - National Research Council framework)
CHARACTERIZATION OF HUMAN HEALTH EFFECTS
ELEMENTS CONSIDERED IN HOST CHARACTERIZATION
(previous lecture)
•
•
•
•
•
•
•
Age
Immune status
Concurrent illness or infirmity
Genetic background
Pregnancy
Nutritional status
Demographics of the exposed population
(density, etc.)
• Social and behavioral traits
CHARACTERIZATION OF HUMAN HEALTH EFFECTS
ELEMENTS CONSIDERED IN HEALTH EFFECTS
(previous lecture)
•
•
•
•
•
•
Duration of illness
Severity of illness
Infectivity
Morbidity, mortality, sequelae of illness
Extent or amount of secondary spread
Quality of life
• Chronicity or recurrence
Characteristics or Properties of Pathogens
Interactions with Hosts (previous lecture)
• Disease characteristics and spectrum
• Persistence in hosts:
– Chronicity
– Persistence
– Recrudescence
– Sequelae and other post-infection
health effects
• cancer, heart disease, arthritis,
neurological effects
• Secondary spread
Elements That May be Included in Dose-Response Analysis
• Statistical model(s) to analyze of quantify doseresponse relationships
• Human dose-response data
• Animal dose-response data
• Utilization of outbreak or intervention data
• Route of exposure or administration
• Source and preparation of challenge material or
inoculum
• Organism type and strain
– including virulence factors or other measures of pathogenicity
• Characteristics of the exposed population
– age, immune status, etc.
• Duration and multiplicity of exposure
Dose-Response Data
and Probability of Infection for Human Rotavirus
Dose
90,000
9,000
900
90
9
0.9
0.09
# Dosed
3
7
8
7
7
7
5
# Infected
3
5
7
6
1
0
0
Dose-Response Models and
Extrapolation to Low Dose Range
• Most dose-response data for microbes are for high
doses of microbes and few hosts
– due practicalities and cost limits
• Real world exposures to microbes from water, food
and air are often to much lower microbial doses
• It becomes necessary to extrapolate the doseresponse relationship to the low dose range where
there are no experimental data points
– a best-fit modelling approach is employed
Models Typically Applied in Microbial
Dose-Response Analyses
• Exponential model: Pinfection = 1 - e-r
where r = probability of infection and  = mean concentration/dose
– assumes organisms are distributed randomly (Poisson) and
– probability of infection = r
– approaches a linear model at low doses
• Exponential (linear) model; two populations:
–
–
one-hit kinetics, but
two classes of human susceptibility to microbe
• Beta-Poisson: a distributed threshold model
– assumes Poisson distribution of microbes and a Betadistributed probability of infection
• r is not a constant but a probability distribution (Beta-distribution)
– two variables in the model
Probabilities of Exposure and Infection
• Pexp (j Dose) = Probability of having j
pathogenic microbes in an ingested dose
• Pinf (j Inf) = Conditional probability of
infection from j pathogens ingested
Probability of Exposure
Exponential Dose-Response Model
Beta-Poisson Dose-Response Model
Rotavirus Dose-Response Relationships:
Experimental Data, Exponential Model and Beta-Poisson Model
Daily and Annual Risks of Various
Outcomes from Exposure to Water
Containing 4 Rotaviruses per 1000 Liters
Volunteer Dose-Response Data
for Norwalk Virus*
Dose (ml)
4
1
0.01
0.0001
No. Dosed
16
21
4
4
No Ill
11
14
2
0
% Ill
69
67
50
0
*"1st passage NV": Dolin et al. 1972; Wyatt et al., 1974.
Norwalk Virus Dose-Response
Analysis Using Alternative Models
1
0.9
Measured
P(D) Fraction withEffect
0.8
Linear (exp)
0.7
Lin(2pop)
0.6
b-Poisson
0.5
0.4
0.3
0.2
0.1
0
0
0.0001
0.01
Dose (ml)
1
4
Dose-Response Relationships for Various
Waterborne Pathogens: Downward Extrapolation
to Low-Dose Range
Comparing Risks of Disease Agents
• Comparing chemical to microbial risks as well
as among agents of each type
• Effects vary widely in severity, mortality rates
and time scale of exposure
• Need to protect both quality and quantity of life
• Drinking water policy needs to be linked to
overall public health policy
• Decision making process needs to take social
and economic factors into account
Desirable attributes of an
integrated measure of risk
• Address probability, nature and
magnitude of adverse health
consequences
• Incorporate age and health status of
those affected
DALYs as unit measures
for health
• Conceptually simple:
– health loss = N x D x S
• N = number of affected persons
• D = duration of adverse health effect
• S = measure for severity of the effect
• Disability Adjusted Life Years
– mortality: years of life lost (YLL)
– morbidity: years lived with disability (YLD)
– DALY = YLL + YLD
Hypothetical example
Disability weight
1
0.8
0.6
Acute
(infectious)
disease
0.4
0.2
Premature
death
0
0
20
40
60
Age Residual
disability
80
Key Question: define health?
‘a state of complete physical, mental
and social well-being, and not merely
the absence of disease or infirmity’
(WHO charter, 1946)
‘the ability to cope with the demands of
daily life’ (the Dunning Committee on
Medical Cure and Care, 1991)
the absence of disease and other
physical or psychological complaints
(NSCGP, 1999)
Deriving severity weights
• Global Burden of Disease Project
– Define 22 indicator conditions
– Use Person Trade Off method to elicit
severity weights
– Panel of physicians and public health
scientists
– Use scale of indicator conditions to
attribute severity weights to other
conditions
– Methodology also applied in other studies
Using Epidemiology for Microbial Risk Analysis
• Problem Formulation: What’s the problem? Determine what
infectious disease is posing a risk, its clinical features, causative
agent, routes of exposure/infection and health effects
• Exposure Assessment: How, how much, when, where and why
exposure occurs; vehicles, vectors, doses, loads, etc.
• Health Effects Assessment:
– Human clinical trials for dose-response
– field studies of endemic and epidemic disease in populations
• Risk characterization: Epidemiologic measurements and
analyses of risk: relative risk, risk ratios, odds ratios; regression
models of disease risk; dynamic model of disease risk
– other disease burden characterizations: relative contribution to
overall disease burdens; effects of prevention and control
measures; economic considerations (monetary cost of the disease
and cost effectiveness of prevention and control measures
Types of Epidemiological Studies that Have Been Used in
Risk Assessment for Waterborne Disease
Some More Epidemiological Terms and Concepts
• Outbreaks: two or more cases of
disease associated with a specific
agent, source, exposure and time
period
• Epidemic Curve (Epi-curve): Number
of cases or other measure of the
amount of illness in a population over
time during an epidemic
– Describes nature and time course of
outbreak
– Can estimate incubation time if
exposure time is known
– Can give clues to modes of
transmission: point source, common
source, and secondary transmission
Point
Source
Time
Common Source
Time
Databases for Quantification and
Statistical Assessment of Disease
• National Notifiable Disease Surveillance
System
• National Ambulatory Medical Care Survey
• International Classification of Disease
(ICD) Codes
• Other Databases
– Special surveys
– Sentinel surveillance efforts
Infectious Disease Transmission (SIR) Model:
Host States in Relation to Pathogen Transmission
Pathogen
Exposure
Susceptible

Infected

Resistant

 = the rate or probability of movement from one state to
another
“Dynamic State” Epidemiological Model of
Microbial Risk - Modeling Infectious Disease
Dynamics and Transmission in Populations
• Members of population move between states
– States describe status with respect to a pathogen
• Movement from state-to-state is modeled with ordinary
differential equations;
– define rates of movement between states: rate terms
• Each transmission process is assumed to be independent
• Change in fraction of population in any state from one time
period to another can be described and quantified
• Different sources of pathogen exposure can be identified
and included in the model
“Dynamic State” Epidemiological Model of
Microbial Risk - State Variables
“SIR” Model of Infectious Disease
State Variables: track no. people in each state at a point in time
• S = susceptible = not infectious; not symptomatic
• I = Infected
– C = carrier = infectious; not symptomatic
– D = disease = infectious; symptomatic
• R = Resistant; same as P = post infection (or) not infectious;
not symptomatic; short-term or partial immunity
• In epidemiology these states are called SIR
Host States in Relation to Pathogen Transmission
Pathogen
Exposure
Susceptible

Infected

Resistant

 = the rate or probability of movement from one state to
another
Infectious Disease Transmission Model at
the Population Level: Dynamic Model
• Risk estimation depends on transmission dynamics
and exposure pathways. Example: Water
Model Development: Household-level Model
of Pathogen Transmission from Water
“Dynamic State” Epidemiological Model
of Microbial Transmission and Disease Risk
Susceptible
Carrier I
Diseased I
Post-infection
Impact of Waterborne Outbreaks of Cryptosporidiosis
on AIDS Patients (previous lecture)
Outbreak
Oxford/
Swindon,
UK, 1989
Attack Rate Mortal.
Ratio (%)
36
Not
reported
Milwaukee, 45
WI, 1993
68
Las Vegas, Not known; 52.6
NV, 1994 incr.
Crypto-pos.
stools
Comments
3 of 28 renal
transplant pts.
shedding oocysts
asymptomatically
17% biliary disease;
CD4 counts <50
assoc. with high risks
CD4 counts <100 at
high risk; bottled
water case-controls
protective
(Previous Lecture)
Predicted Waterborne Cryptosporidiosis in NYC in AIDS Patients
Compared to the General Population
Adults
Adults
Pediatric AIDS
with AIDS
6,080,000 1,360,000 30,000
1,200
40
30
390
10
Total NYC population
Reported cases
(1995)
Predicted tapwater2 (5%)
related reported cases (%
of total actually reported)
Predicted annual risk
5,400
from tapwater unreported (0.03%)
(% of those predicted to
be reported)
Children
3 (10%)
33 (8.5%) 1(10%)
940
(0.3%)
56 (59%)
Perz et al., 1998, Am. J. Epid., 147(3):289-301
1 (100%)
Impacts of Household Water Quality on Gastrointestinal
Illness - Payment Study #1 (An Intervention Study)
Percent of Study Subjects Reporting HCGI Symptoms and Mean Number of
Episodes per Unit of Observation in Both Periods Combined
Group
Filtered Water (n=272)
Tap Water (n=262)
Unit of
% with
Mean Number % with
Mean Number
Observation Episodesa
of Episodesb Episodes
of Episodes
Family
62.0
3.82
67.7
4.81
Informant
20.0
1.70
23.1
2.10
Youngest
42.3
1.83
46.3
2.37
child
aDerived by logistic regression with covariables age, sex, geographic subregion.
bMean number of episodes among those subjects who reported at least one
episode.
Additional Analyses of Health Effects:
Health Effects Assessments
(previous lecture)
• Health Outcomes of Microbial Infection
• Identification and diagnosis of disease caused by
the microbe
–
–
–
–
disease (symptom complex and signs)
Acute and chronic disease outcomes
mortality
diagnostic tests
• Sensitive populations and effects on them
• Disease Databases and Epidemiological Data
Methods to Diagnose Infectious Disease
(previous lecture)
• Symptoms (subjective: headache, pain) and
Signs (objective: fever, rash, diarrhea)
• Clinical diagnosis: lab tests
– Detect causative organism in clinical specimens
– Detect other specific factors associated with
infection
• Immune response
– Detect and assay antibodies
– Detect and assay other specific immune
responses
Health Outcomes of Microbial Infection
(previous lecture)
• Acute Outcomes
– Diarrhea, vomiting, rash, fever, etc.
• Chronic Outcomes
– Paralysis, hemorrhagic uremia, reactive
arthritis, etc.
• Hospitalizations
• Deaths
Morbidity Ratios for Salmonella (Non-typhi)
(previous lecture)
Study
1
2
3
4
5
6
7
8
9
10
11
12
Avg.
Population/Situation
Children/food handlers
Restaurant outbreak
College residence outbreak
Nursing home employees
Hospital dietary personnel
"
Nosocomial outbreak
Summer camp outbreak
Nursing home outbreak
Nosocomial outbreak
Foodborne outbreak
Foodborne outbreak
Morb. (%)
50
55
69
7
8
6
27
80
23
43
54
66
41
Acute and Chronic Outcomes Associated with
Microbial Infections
(previous lecture)
Microbe
Campylobacter
E. coli O157:H7
Helicobacter
Sal., Shig., Yer.
Coxsackie B3
Giardia
Toxoplasma
Acute Outcomes
Diarrhea
Diarrhea
Gastritis
Diarrhea
Encephalitis, etc.
Diarrhea
Newborn
Syndrome
Chronic Outcomes
Guillain-Barre Syndrome
Hemolytic Uremic Syn.
Ulcers & Stomach Cancer
Reactive arthritis
Myocarditis & diabetes
Failure to thrive; joint pain
Mental retardation,
dementia, seizures
Outcomes of Infection Process to be Quantified
(previous lecture)
Exposure
Advanced
Illness,
Chronic
Infections
and
Sequelae
Infection
Disease
Asymptomatic Infection
Acute Symptomatic Illness:
Severity and Debilitation
Sensitive Populations
Mortality
Hospitalization
Health Effects Outcomes: E. coli O157:H7
Health Effects Outcomes: Campylobacter
Sensitive Populations
(previous lecture)
• Infants and young children
• Elderly
• Immunocompromized
– Persons with AIDs
– Cancer patients
– Transplant patients
• Pregnant
• Malnourished
Mortality Ratios for Enteric Pathogens in Nursing
Homes Versus General Population
(previous lecture)
Microbe
Mortality Ratio (%) in:
General Pop.
Nursing Home Pop.
Campylobacter
jejuni
E. coli O157:H7
0.1
1.1
0.2
11.8
Salmonella
0.01
3.8
Rotavirus
0.01
1.0
Snow Mtn. Agent
0.01
1.3
Impact of Waterborne Outbreaks of Cryptosporidiosis
on AIDS Patients
Outbreak
Attack Rate Mortal. Comments
Ratio
(%)
Oxford/
Swindon,
UK, 1989
36
Milwaukee, 45
WI, 1993
Not
reporTed
3 of 28 renal transplants
pts. Shedding oocysts
asymptomatically
68
17% biliary disease; CD4
counts <50 associated
with high risks
Las Vegas, Not known; 52.6
NV, 1994 incr.
Crypto-+
stools
CD4 counts <100 at high
risk; bottled water casecontrols protective
Mortality Ratios Among Specific Immunocompromised
Patient Groups with Adenovirus Infection
(previous lecture)
Patient Group
% Mortality
(Case-Fatality Ratio)
Overall Mean Age of
Patient Group (Yrs.)
Bone marow
transplants
Liver transplant
recipients
Renal transplant
recipients
Cancer patients
60
15.6
53
2.0
18
35.6
53
25
AIDS patients
45
31.1
Databases for Quantification and
Statistical Assessment of Disease
• National Notifiable Disease
Surveillance System
• National Ambulatory Medical Care
Survey
• International Classification of
Disease (ICD) Codes
• Other Databases
– Special surveys
– Sentinel surveillance efforts
Waterborne Outbreak Attack Rates
Waterborne Outbreak Hospitalizations
Predictied Waterborne Cryptosporidiosis in NYC in AIDS Patients Compared to the General Population
Adults
Adults with
AIDS
6,080,000 1,360,000 30,000
40
30
390
Total NYC population
Reported cases
(1995)
Predicted tapwater-related reported 2 (5%)
cases (% of total actually reported)
Predicted annual risk from tapwater 5,400
unreported (% of those predicted to (0.03%)
be reported)
Children
Pediatric AIDS
1,200
10
3 (10%)
33 (8.5%)
1(10%)
940
(0.3%)
56 (59%)
1 (100%)
Perz et al., 1998, Am. J. Epid., 147(3):289-301
Elements That May Be Considered in Risk
Characterization
• Evaluate health consequences of exposure scenario
– Risk description (event)
– Risk estimation (magnitude, probability)
• Characterize uncertainty/variability/confidence in
estimates
• Conduct sensitivity analysis
– evaluate most important variables and information needs
• Address items in problem formulation (reality check)
• Evaluate various control measures and their effects
on risk magnitude and profile
• Conduct decision analysis
– evaluate alternative risk management strategies