epi

what is epidemiology

the study of the distribution and determinants of health-related states and events in populations and the application of this study to control health problems

looking at how health varies over time

key aspects of epidemiology

population

disease distribution

disease frequency

disease determinants

disease control

basic assumptions of epidemiology

health and disability do not occur at random

causal factors can be identified through systematic investigation of human populations

identifying these factors can lead to preventative interventions

basic vs. clinical vs public health sciences

basic studies cells, tissues, animals

clinical studies patients

public health studies populations or communities

Epidemiologic Triad

there are factors associated with increased risks of human disease

host characteristics influence an individuals’ susceptibility to disease

• age

• sex

• race

• religion

• customs

• occupation

• genetic profile

• marital status

• family background

• previous disease

• immune status

agent characteristics

infectivity—ability to invade a host

pathogenicity—ability to cause disease

virulence—ability to cause death

types of agents and examples

biologic

• bacteria, viruses

chemical

• poison, alcohol, smoke

physical

• trauma, radiation, factor

nutritional

• lack, excess

environmental factors associated with increased risk of human disease

• temperature

• humidity

• crowding

• housing

• water

• food

• air pollution

• noise

• altitude

• racism

• income and educational disparities

• status and social class

factors involved in malaria transmission

direct modes of transmission

person-to-person contact (respiratory, orogenital, skin)

examples: HIV, measles, influenza

indirect modes of transmission

• common vehicle (contaminated air, food, water supply)

  • single, multiple, or continuous exposure

  • ex. salmonellosis

• vector (animals, insects)

  • ex. rabies, yellow fever, malaria

• fomites—inanimate objects (doorknobs, toothbrushes, etc.)

reservoir

the habitat (living or nonliving) of the infectious agent

the reservoir consists of:

• symptomatic cases

• carriers

• animals

• inanimate objects (water, food, soil, air)

types of carriers

• incubatory (capable of transmitting prior to onset of symptoms)

• inapparent (infected but have no symptoms)

• convalescent (after acute illness are still infectious)

• chronic (harbor infectious agents for an extended period of time, (e.g. typhoid)

chain of infection process—malaria

endemic

occurring at or near the usual rate of occurrence

epidemic

occurring in clear excess of normality

pandemic

epidemic affecting several countries or continents

common-vehicle exposure—where all cases are exposed to one agent either via

• single exposure—food served at a luncheon

• multiple exposure—same food served more than once

• periodic exposure—water supply contaminated with sewage bc of leaky pipes, contamination may be periodic depending on changing water pressure in the water supply

• continuous—a constant leak leads to persistent contamination

single exposure

typical exposure of a single-exposure, common vehicle outbreak-the most common form is foodborne outbreaks, in which a large population is exposed to the disease agent for a short period of time

continuous or multiple exposures

in the case of continuous or multiple exposures, there are several peaks, and the incubation cannot be easily expressed

R0—basic reproduction number

represents the average number of individuals infected per single case of a disease

used to describe contagiousness of an infectious agent

in a fully susceptible population, what % of cases that an infected person contacts will become infected?

influenced by host, agent, and environment

R0

the number of people that one sick person will infect (on average) is called R0

herd immunity

the resistance of a group to an attack by a disease to which a large portion of the members of the group are immune

• optimal level will depend on disease itself

• must be high (some estimates (95%) to prevent spread of highly contagious diseases (e.g. measles)

• must be even higher to prevent spread in highly dense populations (99%)

• if lower than optimal levels, can result in disease “targeting” those most susceptible (elderly, immune compromised, pregnant women) and many diseases are more virulent when contracted at older ages thereby raising the mortality rate

requirements for herd immunity

• the disease agent must be restricted to a single host species within which transmission occurs

  • if there is a reservoir in which the agent can exist outside the human host, herd immunity not possible

• infections must induce solid immunity

  • if immunity is only partial, then won’t build up a large population of immune people in the community

• depends on random mixing

  • the probability of an infected person encountering every other individual in the population is the same

incubation period

the interval from receipt of infection to the time of onset of clinical illness

incubation periods for selected infectious diseases

clinical and subclinical disease?

stages in the natural history of disease

stage of susceptibility

exposure

subclinical disease stage

pathologic changes is before the onset of symptoms, there if you look hard enough for it

onset of symptoms

clinical disease stage

levels of prevention

primary prevention (activities that help prevent a health care problem)

secondary prevention (prevention of progression, e.g., from pre or subclinical to clinical)

tertiary (early treatment of disease and prevention of complications)

major factors contributing to emergence and re-emergence of infectious disease

• microbial adaptation and change

• human susceptibility to infection

• climate and weather

• changing ecosystems, human demographics and behavior

• international travel and commerce

• breakdown of public health measures

• poverty and social inequality

• war and famine

• lack of political will, vaccine hesitancy

• bioterrorism

• human habits—consuming using animal products

advanced epidemiology triangle

what is a “case”

someone with the outcome of interest

operational definition to validity (accurately) and reliably (consistently) identify who has the health problem

based on self-reported symptoms or behaviors, medical examination/clinical diagnoses (reported various ways)

what is the scope?

geographic area, population, time period

proportion—a part relative to a whole

numerator in denominator

e.g. proportion or % of children in my son’s daycare class with a runny nose

ratio—comparison of any two values

numerator and denominator can be unrelated

e.g. sex ratios

rate—estimate of characteristics of interest (i.e. disease or symptom) in relation to population or period of time

e.g. birth, fertility or morality rates

birth rate

birth rate=(number of live births in a population during a specified time period/population from which births occurred) x 1,000

fertility rate

fertility rate = (number of live births in a population during a specified time period/population of women aged 15-44 years) x 1,000

morbidity

morbidity is used to refer to the extent of disease or disease frequency within a population

two measures of morbidity: prevalence or incidence

prevalence

prevalence is the proportion of new and existing cases of a health condition in a study population at a point in time

describes the ‘burden’ of a disease or health condition in a population during a specified time period

prevalence formula

prevalence = (number of new and existing cases/number of persons in population at a specified time) x 100

types of prevalence

point prevalence

• number of existing cases at a point in time

• most commonly used

period prevalence

• number of existing cases during a defined period (i.e. a five year prevalence)

• less frequently used

incidence

incidence measures how frequently new cases of a health condition occur over a period of time (a measure of risk)

an incident case = when an individual changes from being susceptible to being diseased

two ways to express incidence

cumulative incidence: number of new cases per number of persons at risk during a specified period of time

incidence rate (or density): number of new cases in the population at risk per unit of time (person-time units at risk)

cumulative incidence

the proportion of disease-free individuals in a population who become ill during a defined time period

(number of new cases during a specific time period/number of persons at risk of developing disease during that time period) x 100

examples of who’s at risk

incidence rate (density)

number of new cases during the time time period/total person-time of observation

person-time = sum of periods of time at risk for each member of population (e.g., person-years)

calculating person time example

relationship between incidence and prevalence

in a stable population (where rates do not change and in and out migration are equal)

prevalence = incidence x duration of disease

duration of disease determined by mortality and recovery

crude mortality rate

mortality rate from all causes (per 1,000 population) is known as the crude or all-cause mortality rate:

crude mortality rate = (total # of deaths from all causes during a given time period/# of persons in population at that time period) x 1,000

a mortality rate is an incidence rate of deaths since it is the occurrence of new cases divided by the total population at risk over a specified period of time (usually during a year)

specific mortality rates

can specify mortality rates for specific age group

can specify for specific disease

• *denominator must reflect the population group studied*

can specify for multiple criteria (age group and specific disease)

cause-specific mortality rate

cause-specific mortality rate = (number of deaths from a specific cause occurring during a given time period/population from which death occurred) x 10^n

maternal mortality rate

maternal mortality rate = total number of maternal deaths during a specific time period/number of live births during that time period) x 100,000

case fatality rate

used when you want to know how many of those infected or inflicted with disease will die (percent)

case fatality rate = (number of deaths from a specific disease during a specific time period/number of cases of the disease during the same time period) x 100

factors that might impact morbidity and mortality rates include

• changes to medical and diagnostic standards (revisions in the ICD system, improved screening methods)

• place (environmental changes, regional differences in standard of care, etc)

• person (age, sex, ethnicity, social habits)

comparing crude mortality rates across populations

we can’t compare crude mortality rates across populations because of age distribution differences across populations

• age distribution—the most important factor that influences mortality AND that can differ between populations

must adjust crude death rate by age in order to make appropriate comparisons

public health surveillance

uses data to monitor health problems to prevent and control them

public health surveillance is the continued watchfulness over the distribution and trends of incidence [of a disease] through the systematic collection, consolidation, and evaluation of morbidity and mortality reports and other relevant data

why collect surveillance data?

• Assess population health status

• monitor temporal trends

• detect outbreaks

• define objectives, priorities, and strategies

• target interventions

• evaluate interventions, effectiveness of control efforts

• generate research hypotheses

questions that surveillance data can answer

• what is the infant mortality rate?

• are we advancing towards predefined health goals (ie, reducing cancer rates)?

• has there been an increase in unsafe sex among MSMs?

• are smoking rates declining in teens?

• what factors are associated with West Nile virus?

• is flu season here? Is it more severe than usual?

• what neighborhood has the highest rate of asthma?

steps in conducting surveillance

1. Identify, define and measure health problem of interest and objective of surveillance problem

2. collect and compile data about the problem

3. analyze and interpret these data

4. share data with those responsible for control

5. monitor and evaluate usefulness of surveillance problem

*does not include actions to control the problem

criteria for selecting and prioritizing health problems for surveillance (CDC)

• public health importance

  • incidence, prevalence, severity, consequences/mortality, socioeconomic impact, communicability, public perception, international requirements

• ability to prevent, control, treat the problem

• capacity of health system to implement control measures

  • speed of response, economics, availability of resources, ability to ensure people get the treatment they need

case definition slide?

International Classification of Disease (ICD)

• standard diagnostic tool for mortality and morbidity (WHO)

• used to:

  • classify disease and other health problems

  • identify health trends/statistics globally

  • reimbursement and resource-allocation decisions

• recorded in many different health and vital records

• most countries in world use it to report mortality data

• current version is ICD-11, national adaptations in many countries

sources of data for surveillance

• three sources: individuals, environment, healthcare providers/institutions

• information about disease course, size of population under surveillance and geography are helpful

• collected continuously, periodically, or defined period

types of data sources

Primary—environmental monitoring, syndromic surveillance, notifications, registries (including vital statistics)

Secondary—medical records, school records, other administrative data, surveys

active surveillance

go and collect health data

• health department contacts healthcare providers or labs about conditions to identify cases

• community-based surveys are conducted to measure particular conditions in a neighborhood

• health inspectors go to particular facilities to measure levels of infection

advantage—useful when need to identify all cases

disadvantage—requires more resources

passive surveillance

health departments rely on healthcare provider or labs to report cases of disease

advantage—efficient, requires limited resources

disadvantage—incomplete due to underreporting

sentinel surveillance

monitoring through a specific site or partner

what to do with surveillance data once it’s collected

• analyze data by person, time, and place

• interpret data

  • e.g., why might prevalence or incidence have increased or decreased over time

• disseminate data and results to key stakeholders

  • e.g., MMWRs, surveillance reports

• evaluate and improve surveillance systems

  • timely? useful? what resources are needed?

sources of epidemiologic data

• statistics from vital registration systems (birth, death)

• reportable disease statistics

• disease registries

• morbidity surveys of general population

• insurance data

• clinical data sources

• school health programs

• census data

• economic data

• peer-reviewed articles and systematic reviews

• reports from health organizations (e.g., CDC, WHO), advocacy groups

registries

• centralized database for documenting or tracking health events or persons over time

• certain registries are required by law (vital statistics)

• reported by health care providers, health care facilities, morticians/funeral directors, patients

examples:

• vital statistics

• immunizations/genetic conditions

• disease-specific or treatment-specific (e.g., specific cancers)

• exposure-specific (9/11)

vital statistics registration system in the US

• All deaths are recorded and reported to local health departments and to the state office of vital statistics

• Reports of vital event statistics including deaths are reported to the National Center for Health Statistics

death certificate data

• Date of birth and of death (for accurate age)

• Stated age

• Place of death

• Place of residence

• Occupation

• Sex

• Marital status

• Cause of death (ICD-11)

• Other information may include type of injury, place and time of injury, etc.

limitations of death certificates

• certification of cause of death (e.g., for an elderly person with a chronic illness, cause of death may be unclear)

• lack of standardization of diagnostic criteria

• stigma associated with certain diseases, e.g., AIDS, suicide, which may lead to inaccurate reporting

• coding errors and changes in coding (e.g., ICD codes)

surveys

• systematic, structured method of gathering information to quantitatively describe population

  • census (entire population) vs. sample

• typically cross-sectional, some conducted periodically

• many are population-based

Uses

• understand extent of health problem

• conduct surveillance and examine trends

• estimate and target needed resources

sampling methodology

strengths and limitations of surveys

Strengths:

• population-based/representative

• many conducted periodically, can be used for surveillance

• cover wide breadth of important public health topics and include many demographics

• often use validated measures

Limitations:

• not fully representative

• not always generalizable

• self-reported information

• could change over time reflect demographic change only?

• not timely

• not easy to add or change survey questions

objective of epidemiological studies

1. quantify the burden/risk of disease in specific populations

2. identify causes of disease to inform prevention

3. test strategies for prevention and control

types of study and approach

descriptive epidemiology questions

person, time, and place

When was the population affected?

Where was the population affected?

Who was affected?

person, place and time characteristics

types of descriptive and experimental studies

cross-sectional study

a snapshot of the population at a certain point in time

• exposure and disease measured simultaneously at the individual level

• also known as a prevalence study

• purpose

  • estimation of the magnitude and distribution of a health problem

  • planning health services and administering medical care facilities and other intervention planning

  • if conducted serially (ie, annually) can help compare trends over time

  • hypothesis generation-seeing what factors are associated with outcome of interest

cross section study process

Cross sectional studies advantages

Inexpensive

Easy to conduct

Quantity health problems

Useful for planning (e.g., needs assessments) and resource allocation

examine many associations at once

hypothesis generating

intervention planning

disadvantages of cross-sectional studies

identifying prevalent cases, not incident cases → cases with a longer duration are more likely to be detected

non-response bias—people who respond to questionnaires are systematically different than those who do not

not possible to establish temporality (and thus causality)

threat of reverse causality*

*beware-looking at linear trends over time → the same people aren’t surveyed each time

describing disease occurrence, by exposure status:

prevalence of disease among the exposed: a/(a+b)

prevalence of disease among the unexposed: c/(c+d)

odds of disease in the exposed: a/b

odds of disease in the unexposed: c/d

odds formulas (cross-sectional study)

odds of disease in the exposed: a/b

odds of disease in the unexposed: c/d

comparing disease occurrence, by exposure status

prevalence (risk) ratio: compares prevalence in exposed to prevalence in the unexposed

= (a/(a+b))/(c/(c+d))

odds ratio: compares odds of having disease if you’re exposed to odds of having disease if you’re unexposed

= (a/b)/(c/d) = ad/bc

prevalence ratio and risk ratio

prevalence ratio in cross-sectional studies is comparable (same formula) to the risk ratio in cohort studies

ecologic studies

• units of analysis are “groups” or “aggregates”

  • countries provinces, states, counties, regions, neighborhoods

  • organizations: hospitals, school districts, schools, classrooms, etc

• both exposure and outcome are measured for each group, usually at the same time

  • in this sense similar to cross sectional studies

types of group-level variables

• summary of individual-level data

  • e.g., median. household income, % with high school diploma, HIV seroprevalence, etc.

• property of the aggregate

  • e.g., neighborhood grocery stores, seat belt legislation, community norms, etc

ecologic fallacy

drawing inferences about associations between exposure and outcome at the individual level based on an observed association at the group level

remember: with ecologic studies, we only have data at the group level not the the individual level

strengths of ecologic studies

• some exposures may vary very little, or not at all, within a given group

  • by comparing groups (e.g. countries) ecologic studies allow us to study a greater range of exposures

    • ex. if diet varies very little within countries, it may not appear to be an important risk factor for CHD. The true relationship between diet and CHD can only be seen by comparing countries with very different diets

• some variables are only defined (and can only be measured) for groups

  • the influence of these exposures can only be studied if we have similar information for different groups

    • exposures such as income inequality, level of industrialization, physicians per capita, policies, are all things that characterize countries and are relevant to health

why conduct outbreak investigations?

• control and prevention

• severity and risk to others

• research opportunities to gain additional knowledge

• training opportunities

• program considerations

• public, political, or legal concerns

outbreak investigation steps

1. define the epidemic

2. describe your data

3. develop and test hypothesis

recommend control measures and communicate findings

further outbreak investigation steps

• recommend control measures

  • control of present outbreak

  • prevention of future similar outbreaks

• communicate findings

  • brief local health officials

  • the community

outbreak investigation step 1: define the epidemic

• verify the existence of an outbreak

  • do cases in a certain time and place exceed what is expected?

• verify the diagnosis

  • clinical features: is the disease known

  • what are its serologic/cultural aspects? Clinical symptoms?

  • are the causes partially understood?

• establish a case definition (define the numerator)

  • pathology (microscopic examination)

  • clinical characteristics (symptoms)

  • specific disease agent (e.g., virus)

  • international classification of disease (ICD)

• also define the denominator: what population is at risk of developing disease

case definition examples

case definition should be practical, reliable, and applied without bias

outbreak investigation step 2: describe your data

• identify and count cases

• examine the distribution of cases by:

  • Time

  • Place

  • look for time-place interactions

• look for combinations of relevant variables