1.  Study hx of health and populations

2.  Dx health of community

3.  Study working of health services-operations research

4.  Estimate individual risks of dx or other conditions

5.  Identify syndromes

6.  Complete clinical picture of chronic disease

7.  Search for causes of health and disease

Secular trends

1.  Lack of comparability over timem d/t altered diagnostic criteria

2.  Aging of general population

3.  Changes in fatal course of condition

Conditions that were once common but no longer are present in epidemic form

Ex.  smallpox

Conditions for which key contributing factor is known but control methods have not been effectively implemented

Ex.  STDs, EtOH abuse, Tobacco use

Diseases for which there is no effective method of prevention or known cure

Ex.  Certain types of cancer

Diseases increasing in frequency

Ex.  Lung cancer and HIV

1.  Births

2.  Deaths

3.  Migration

1.  Adds no new members

2.  Therefore, only decreases in size d/t death

Dynamic population

1.  Adds through migration and birth

2.  Declines d/t emigration and death

1.  Provides a key to types of problems requiring attention

2.  Determines need for specific health services

1.  Age and Sex distribution

2.  Socieoeconomic status

3.  Family structure

4.  Racial, ethnic, and religious composition

1.  Availability of social and health services

2.  Quality of housing stock

3.  Social stability (residential mobility)

Study of the palcement and optimum utilization of health services in a community

Ex.  Coordination of programs for developmentally disabled; studies of health care utilization; minority access to health insurance

1.  Search for causes

2.  Individual risks

3.  Specific clnical concerns

1964 Surgeon General: 

5 Criteria for Causality

1.  Strength of association

2.  Time sequence

3.  Consistency upon repetition

4.  Specificity

5.  Coherence of explanation

1.  Biologic gradient

2.  Plausibility

3.  Experiment

4.  Analogy

Risk factor

Exposure that is associated with a disease

Ex.  smoking

3 Criteria for Risk Factors

1.  Frequency of disease varies by category or value of the factor (light smokers v. heavy smokers)

2.  Risk factor precedes onset of disease

3.  Observation must not be d/t error

1.  Compares pts with disease (cases) with those free from it (controls)

2.  Looks at exposures in both groups to determin risk factors

1.  Case control

2.  Cohort

Cohort Design

1.  Criteria of causality

2.  Relevance to each pt (size of risk)

3.  Public health implications (individual vs population)

1.  Prepathogenesis:  before agent-host interaction

2.  Pathogenesis:  after agent-host interaction

3.  Later stages include S/S and recover, disability, or death

Occurs during prepathogenesis

Designed to reduce disease occurance

Ex.  Education and Immunization

-Occurs during pathogenesis phase

-Designed to reduce progress of disease

-Ex:  screening programs cancer and diabetes

-Designed to limit diability from disease and restore optimal functioning

-Ex:  PT for a stroke patient

Determinants

Factors or events capable of brining about a change in health

Ex.  bacteria, carcinogens, stress, drinking, etc

Epidemiology examines occurance among population groups, not individuals

Variation by:

1.  Age

2.  Time

3.  Geographic location

4.  Other variables

1.  Morbidity:  designates illness

2.  Mortality:  refers to deaths that occur in a population or other group

Occurrence in a community or region of cases of an illness (or an outbreak) clearly in excess of expectancy

*can be as few as 2 cases associated 

**Does not have to be infectious (Brown lung, Love canal)

Seasonal variations and births and deaths

Known as "Columbus" of biostats

Cholera on Broad Street linked to contaminated water supplies

Lambeth Co. was safer water supplier

Epidemiologist does not manipulate a risk factor but rather observes the changes in an outcome as the result of a naturally occurring situation

Ex.  Seat belt law, tobacco tax

Compiler of Abstracts in England

Started classification of disease which we can trace to current ICD system

Examined links b/t mortality rates and population density

1.  All diseases caused by microorganisms

2.  If you isolate a pure culture of the microorganism and inoculate a susceptible animal, it will get the same disease

Refers to  the # of cases of a disease or other health phenomena being studied

Ex.  Traffic fatalities in Manhattan in a 24 hr period

A measure that states a count relative to the size of the group

Numerator ALWAYS a part of denominator

May be expressed as a %

A ratio that consists of a numerator and a denominator and in which time forms part of the denominator

Has the following elements

1.  Disease frequency

2.  Unit size of population

3.  Time period during which event occurs

= (# deaths in a given year) / (Reference population during midpoint of the yr) 

Ans*100,000

Provides an indication of the extent of a health problem

1.  Point prevalence (at this point 33% of kids at summer camp have diarrhea)

2.  Period prevalence (how many women have breast cancer in a given unit of time)

1.  Describing the burden of a health problem in a population

2.  Estimating the frequency of an exposure

3.  Determining allocation of health resources such as facilities and personel

The rate of development of a disease in a group over a certain period of time

1.  Numerator = # new cases

2.  Denominator = population at risk

3.  Time = period during which cases occur

1.  Helps in research on the etiology/causality of disease

2.  Estimates the risk of developing a disease

3.  Used to estimate the effects of exposure to a hypothesized factor of interest 

= (# of new cases) / (population at risk)

Ans / (time period for followup)

Incidence rate uses frequency of new cases in numerator

Individuals who have hx of disease not included 

1.  Denominator for all incidence rate calculations

2.  Excludes those with disease

3.  In a case of chronic disease:  ppl that are susceptible

1.  Alternative form of incidence rate

2.  Used for a disease observed in a population for a short time period (ie, Salmonella gastroentereitis 

3.  AR = [(Sick) / (Sick + Well)] * 100 

= (# of new cases during time period) / (total persons * time of observation)

*Ex is a chart that has a bunch of different studies for different time periods...multiply # of ppl in each study by lenght of study and sum to get denominator...numerator is just sum of all adverse events (MI)

The prevalence (P) of a disease is proportional to the incidence rate (I) times the duration (D) of a disease

P = ID

*Short duration disease with high insidence, prevalence becomes similar to incidence

*  If duration of disease is long and incidence is low, prevalence increases greatly to incidence 

Used to project population changes; it is affected by the # and age composition of women of childbearing age

= [(# of live births per period) / (population size at middle time of period)] * 1000

Used for comparisions of fertility among age, racial, and socioeconomic groups

= [(# of live births w/in a yr) / (# of women aged 15-44 yrs during midpoint of yr)] * 1000

Used for international comparisions; a high rate indicates unmet health needs and poor environmental conditions

= [(# of living infant deaths aged 0-365 days during the yr) / (# of live births during yr)] * 1000 live births

Used to estimate the risk of death of the fetus associated with the stages of gestation

1.  Fedal death rate = [(# of fetal deaths after 20 wks or more gestation) / (# of live births + #of fetal deaths after 20 wks or more gestation)] * 1000

2.  Late Death Rate = same equation but after 28 wks gestation instead of 20 wks

Provides a measure of fetal wastage (loss) relative to the # of live births

= [(# of fetal deaths after 20 wks + gestation) / (# live births)] * 1000

Reflects events happening after birth (congenital malformations, prematurity, low birth wt)

= [(# of infant deaths under 28 days of age) / (number of live births)] * 1000

Reflects environmental events, control of infectious diseases, and improvements in nutrition

= [(# of infant deaths 28-365 d after birth) / (# live births - neonatal deaths)] *1000

Reflects environmental events that occur during pregnanc yand after birth; it combines mortality during the prenatal and postnatal periods

= [(# of late fetal deaths after 28 wks or more of gestation + infant deaths w/in 7 days birth) / (# of live births + # of late fetal deaths)] * 1000

Reflects health care access and socieoeconomic factors; it includes maternal deaths resulting from causes associated with pregnancy and puerperium (during and after child birth)

= [(# of deaths assigned to causes related to childbirth) / (# of live births)] * 100,000

Indicates relative importance of a specific cause of death; not a measure of the risk of dying of a particular cause

= [(mortality due to specific cause during time peroid) / (mortality due to all causes during same time period)] * 100

1.  Results in age-adjusted death rates that are much larger

2.  Affects trends in age-adjusted rates for certain causes of death

3.  Narrows race differentials in age-adjusted death rates

1.  Evaluate and compare trends in heatlh and disease

2.  Provde a basis for planning, provision, and evaluation of health services

3.  Identify problems for analytic studies (creation of hypothesis)

1.  Positive declaration (research)

2.  Negative declaration (null)

3.  Implicit question

1.  The method of difference:  all the factors in 2 or more places are the same except for a single factor

2.  The method of agreement:  a single factor is common to a variety of settings (air pollution)

3.  The method of concomitant variation:  the frequency of disease varies according to the potency of a factor

4.  The method of residues:  involves subtracting potential causal factors to determine which factors have greatest impact

1.  Case reports (counts):  simplest

2.  Case series:  summarize characteristics from major clinical settings

3.  Cross-sectional studies:  surveys of population to estimate prevalence of disease or exposure

Very important for any disease process

Linear (cancer) or multi-modal (TB)

**Incidence rates based on elderly often inaccurate

Females have higher morbidity than males for acute and chronic

Death rates for both sexes declining 

Married folks have lower morbidity and mortality rates

Could be protective or selective (healthier people tend to get married)

Race/Ethnicity Categories

*Since 2000, census has allowed multiple categories for a single person

1.  African Am

2.  Am Indian

3.  Asian

4.  Hispanic/Latino

Highest rate of mortality for all groups studied

Higher blood pressure levels than other populations

Pima Indians:  age and sex adjusted mortality much higher than for all races in US for:  accidents, cirrhosis, homicide, and diabetes

-ID 10th leading cause of death

-Males 25-34 yrs old, death rate 6.6 times that for all races in US

Modifications that individuals or groups undergo when they come in contact with another country

*Alters chronic disease states

Hispanic Health and Nutrition Examination Survey (HHANES):  examined health and nutritional statis of major Hispanic/Latino population centers

San Antonio Heart Studie:  high rates of obesity and diabetes amongst Mexican Americans

Place of origin of indivduals

Categories:  foreign born and native born

Seventh Day Adventists

1.  Vegitarians who don't drink or smoke

2.  Lower rate of CHD, cancer, and HTN

*Similar findings for Mormons

Low status correlates with increased mortality, morbidity, and disability rates d/t factors including:

Poor housing; crowded conditions; racial disadvantage; low income; poor education; unemployment

1.  Prestige of occupation or social position

2.  Educational attainment

3.  Income

4.  Combinded of 2 or more above

5 socioeconomic classes in New Haven, CT based on occupational prestige, education, and address

Inverse correlation b/t social class and likelihood of being mental pt under treatment (higher social class = less mental illness treatmetn)

*Type of Tx also varied by social class

1.  Social causation explanation (breeder):  conditions associated with lower social class produce metnal illness

2.  Downward drift hypothesis:  persons with severe metnal disorders move to impoverished areas

1.  Higher rate infectious disease

2.  Higher infant mortality rate and overall mortality

3.  Lower life expectancy

4.  Larger proportion of cancers with poor prognosis  (delay in seeking healthcare?)

1.  WHO tracks variations

2.  Infectious disease 44% of all deaths in less developed nations (4% developed)

3.  Variation d/t climate, cultural factors, dietary habits

D/t variations in climate, geology, latitude, pollution, ethnic and racial concentrations

*In us, you can compare regions or states

Disease patterns d/t unique environmental or social conditions found in particular areas of interest 

*Ex:  fluorosis:  associated with naturally occuring fluoride deposits in water

Goiter:  iodine deficiency in land-locked areas of US

A method to provide a spatial perspective on the geographic distribution of health conditions

*Produces choroplath map that shows variations in disease rates by different degrees of shading

1.  Gene/environment interaction (sickle-cell gene; Tay-Sachs)

2.  Influence of Climage (yaws, Hansen's)

3.  Environmental FActors:  Chemical agents linked to cancer

1.  Cyclic fluctuations

2.  Point epidemics

3.  Secular time trends

4.  Clustering (Temporal; Spatial)

-Periodic changes in frequency of disease and health conditions over time

-Related to changes in lifestyle of the host, seasonal climatic changes, and virulence of infectious agent (higher heard disease mortality in winter and more accidents in summer)

Response of a group of ppl circumscribed in palce and time to a common source of infection, contamination, or other etiologic factor to which they were exposed almost simultaneously

Ex:  foodborne illness

Refer to gradual changes in the frequency of a disease over long time periods

Ex:  decline of mortality d/t heart disease in US

1.  Case:  refers to unusual aggregation of health events grouped together in space and time

2.  Temporal:  eg, post-vaccination reactions, post-partum depression

3.  Spatial:  concentration in a specific geographic area (Hodgkin's)

A study is said to have internal validity when there have been proper selection of study groups and a lack of error in measurement

Concerned with the appropriate measurement of exposure, outcome, and association b/t exposure and disease

Reflect fluctuations around a true value of a parameter b/c of sampling variabilty

1.  Poor precision

2.  Sampling error

3.  Variability in measurement

Occurs when factor being measured is not being measured sharply 

Ex:  aiming rifle at target not in focus

Increase precision by increasing sample size or number of measurements

Occurs when sample selected is not representative of target population

Reduced by increasing sample size

Deviation of results or inferences fromt he truth, or processes leading to such deviation.  Any trend in the collection, analysis, interpretation, publication, or review of data that can lead to conclusions that are systemictically different from the truth

1.  Selection bias

2.  Information bias

3.  Confounding

Arises when relation b/t exposure and disease is dfferent for those who particcipate and those who theoretically would be eligible for study but do not participate

Ex:  younger people may be more likely to respond to health surveys

Can be introduced as a result of measurement error in assessment of both exposure and disease 

1.  Recall:  better recall among cases than namong controls (family recall bias)

2.  Interviewer/abstractor bias:  occurs when interviewers probe more thoroughly for an exposure in a case than in a control

3.  Prevaricaiton (lying) bias:  occurs when participants have ulterior motives for answering questions and may exaggerate exposure

Distortion of the estimate of the effect of an exposure of interest b/c it is mixed with the effect of an extraneous factor

*Occurs when crude and adjusted measures of effect are nto equal (at least 10%)

Can be controlled for in data analysis

To be a confounder, an extraneous factor must satisfy the following criteria (3):

1.  Be a risk factor for the disease

2.  Be associated with the exposure

3.  Not be an intermediate step in the causal path b/t exposure and disease

1.  Air pollution and bronchitis are associated, but also influenced by crowding which is a confounder

2.  High altitude and lower heart disease mortality also may be linked to ethinc composition of ppl in these regions

1.  Develop an explicit (objective) case definition

2.  Enroll all cases in a defined time and region

3.  Strive for high participation rates

4.  Take precautions to ensure representativeness

1.  Ensure all medical facilities are thoroughly canvassed

2.  Develop an effective system for case ascertainment

3.  Consider whether all cases require medical attention; consider possible strategies to identify where else the cases might be ascertained

1.  Compare the prevalence of the exposure with other sources to evaluate credibility

2.  Attempt to draw controls from a variety of sources

1.  Use memory aids; validate exposures

2.  Blind interviewrs as to subjects' study status

3.  Provide standardized training sessions and protocols

4.  Use standardized data collection forms

5.  Blind participants as to study goals and classification status

Prevention strategies:  attempt to control confounding through the study design itself

1.  Randomization

2.  Restriction

3.  Matching

Attempts to ensure equal distributions of the confounding variable in each exposure catetory

*Advantages:  convenient, inexpensive; straightforward data analysis

*Disadvantages:  Need control over exposure and ability to assign subjects to study groups (large sample sizes needed)

May prohibit variation of the confounder in study groups (restricting age can narrow age as a confounder)

*Provides complete control of known confounders, BUT NOT unknown confounders-->randomization takes care of unknown as well

Matches subjects in the study groups according to the value of the suspected or known confounding variable to ensure equal distributions

1.  Frequency matching:  # of cases with particular match characteristics is tabulated

2.  Individual matching:  pairing of one or more controls to each case based on similarity in sex, race, or other variables

*Good:  fever subjects required than unmatched; Bad: costly

Two Analysis Strategies to Control Confounding

1.  Stratification:  analyses performed to evaluate the effect of an exposure wihin strata (levels) of the confounder

2.  Multivariate techniques:  use computers to construct mathematical models that describe simultaneously the influence of exposure and other factors that may be confounding the effect

1.  Performing analyses within strata is a direct and logical strategy

2.  Minimum assumptions must be satisfied for the analysis to be appropriate

3.  The computational procedure is straightforward

1.  Small #s of observations in some strata

2.  Variety of ways to form strata with continuous variables

3.  Difficulty in interpretation when several confounding factors must be evaluated

4.  Categorization produces loss of information

1.  Good:  continuous variables do not need to be converted to categorical variables; allow for simultaneous control of several exposure variables in a single analysis

2.  Bad:  potential for misuse

Occurs b/c of the influence of study results on the chance of publication (+ studies more likely to be published than - studies)

*May result in preponderance of false-positive results in literature

Bias is compounded when published studies are subjected to meta-analysis

Presumptive identification of unrecognized disease or defects by the application of tests, examinations, or other procedures that can be applied rapidly

*positive screens followed by diagnostic to confirm

1.  Defined as use of 2 or more screening tests together among large groups of people

2.  Information obtained on risk factor status, hx of illness, health measurements

3. Commonly used by employers and HMOs

1.  Mass Screening:  screening on a large scale of total population groups regardless of risk status

2.  Selective:  screens subset of high risk population for disease (more economical and likely to yield true cases)

1.  Population/Epidemiologic Surveys:  purpose is to gain knowledge regarding distribution and determinants of diseases in selected populations (no benefit to participant implied)

2.  Epidemiologic surveillance:  aims at the protection of community health through case detection and intervention

3.  Case finding (opportunistic screening):  the utilization of screening tests for detection of conditions unrelated to pts chief complaint

1.  Social

2.  Scientific

3.  Ethical

Health problem should be important for individual and community

*Diagnostic follow-up and intervention shoudl be available to all who require them

*Favorable cost-benefit

*High public acceptance

1.  Natural hx of condition should be adequately understood (permits early stage diagnosis and appropriate biological markers of progression)

2.  Prevalence of disease or condition is high

1.  Program can alter natural Hx of condition in a significant proportion of those screened

2.  Suitable, acceptable tets for screenign and Dx of condition as well as acceptable, effective methods of prevention are available

1.  Simple

2.  Rapid

3. Inexpensive

4.  Safe

5.  Acceptable to target group

Evaluation of Screening Tests

1.  Reliability types (3)

2.  Validity Types (3)

1.  Repeated measurements; internal consistency; interjudge

2.  Content; criterion-referenced (predictive or concurrent); construct

The abilty of a measuring instrument to give consistent results on repeated trials

1.  Repeated measurement reliabilty:  the degree of consistency among repeated measurements of the same individual on more than one occassion

2.  Internal consitency reliability:  evaluates the degree of agreement or homogenetity within a questionnaire meaure of attitude, personal characteristic, or psychologic attribute

3.  Interjudge reliability:  reliability assessments derived form agreement among trained experts

The ability of a measuring instrument to give a true measure

*Can be evaluated only if an accepted and independent method for confirming the test measurement exists

1.  Predictive validity:  denotes ability of a measure to predict some attribute or characteristic in the future

2.  Concurrent validity:  obtained by correlating a measure with an alternative measure of the same phenomenon taken at the same point in time

It is possible for a measure to be higly reliable but invalid

It is NOT possibe for a meausre to be valid but unreliable

Measurement Bias (Source of Unreliability and Invalidity)

1.  To improve sensitivity, the cut point used to classify individuals as diseased should be moved farther in the range of the nondiseased (normals)

2.  To improve specificity, the cut poitn should be moved farther in the range typically associated with the disease

1.  Retrain screeners:  reduces misclassification in tests requiring human assessment

2.  Recalibrate screening instruments

3.  Utilize a different test

4.  Utilize more than one test

1.  Randomized control trials:  subjects either get new screening test or usual care

2.  Ecologic time trend studies:  compare geographic regions with screening programs to those without

3.  Case-control studies:  cases (fatal) and controls (nonfatal)

1.  Lead time bias:  perception that the screen-detected case has longer survival b/c disease was identified early

2.  Length bias:  cancer screening; tumors identified by screening grow slower and have better prognosis

3.  Selection bias:  motivated participants have different probability of disease than do those who refuse to participate

Nomenclature and classification of disease are central to reliable measurement of the outcome variable in epidemiologic research

Nomenclature:  highly specific set of terms for describing and recording clinical or pathologic dx to classify ill persons into groups