EPID4070 Exam 3
Phenomenon of disease: what is disease?
"a type of internal state which is either an impairment of normal functional ability-that is, a reduction of one or more functional abilities below typical efficiency-or a limitation on functional ability caused by environmental agents" or "a state that places individuals at increased risk of adverse consequences"
Problem: How do we know when something makes a difference?
-Consider the following statement: If the rooster crows at the break of dawn, then the rooster caused the sun to rise. Association ≠ Causation
Collection of Exposure Data
-Employment records often are used and may include: --Personal identifiers to permit record linkage. --Demographic characteristics. --Work history. --Information about potential confounding variables, e.g., medical history, smoking habits.
Monitoring and Surveillance of Occupational Hazards
-Hazard surveillance -Health surveillance -Sentinel health event
Arrangement of DNA
DNA is organized into chromosomes -23 pairs of chromosomes -1 pair determines sex (X, Y) -22 pairs of autosomes --Contains roughly 750-2,800 genes
Selection Bias
Errors due to systematic differences in characteristics between those who are selected for study and those who are not.
Epidemiology as a Profession
More than 30 specialty areas: Reproductive epidemiology Nutrition epidemiology Health services research epidemiology Neuroepidemiology Injury epidemiology Coronary heart disease epidemiology Pharmacoepidemiology Renal epidemiology Oral epidemiology/dental epidemiology Veterinary epidemiology
Absolute Risk
Probability of an individual developing a disease or change in health status over a fixed time interval, conditional on the individual not dying during the same time period
Molds
Recent hurricanes have provided ideal conditions for mold growth.
Natural history is central to prevention
See slides 8-10 from Ch.11
Validity (Accuracy)
The degree to which a measurement measures what it purports to measure
Homozygous allele
locus and allele are the same ex: A-A
Reliability
• Also known as consistency • Ability to yield the same results with repeated measurements of same construct • Degree to which results are free from random error
Screening
• Objective: to reduce mortality and/or morbidity by early detection and treatment. • Secondaryprevention. • Asymptomatic individuals are classified as either unlikely or possibly having disease. • Important distinction between mass screening and selective screening.
Mass Health Examinations
• Population or epidemiologic surveys ‐‐purpose is to gain knowledge regarding the distribution and determinants of diseases in selected populations. • No benefit to the participant is implied. • Epidemiologic surveillance ‐‐aims at the protection of community health through case detection and intervention. • Case finding (opportunistic screening) ‐‐the utilization of screening tests for detection of conditions unrelated to the patient's chief complaint.
Predictive Values
• Positive predictive value • Negative predictive value
Combination Testing
• Serial Testing • Parallel Testing
Types of Coherence
• Theoretical • Factual • Biological • Statistical
In summary, Sir Bradford Hill's "guidelines" are useful guides for:
•Remembering distinctions between association and causation in epidemiologic research; •Critically reading epidemiologic studies; •Designing epidemiologic studies; •Interpreting the results of your own study.
10.2 Slides
**Go through beginning slides (almost all diagrams)**
Example of Confounding
*See diagram 10.2, slide 12 -Smoking is associated with pancreatic cancer -Smoking is associated with coffee drinking -Smoking is not a result of coffee drinking
Confounding
*See diagram 10.2, slide 9 -A mixing of effects - association between exposure and disease is distorted by the effect of a third variable that is associated with the disease -Alternate explanation for observed association between an exposure and disease
Random error vs. Systemic error
*see chart from 10.1 slide 10
Validity vs. Reliability
*see chart from 10.1 slide 9
Relationship between Bias and Chance
*see charts from 10.1 slides 6, 7
Attributable risk
- # of cases among the exposed that could be eliminated if the exposure were removed = Incidence in exposed ‐ Incidence in unexposed Risk Difference (RD)
Serial Testing
- A test is first applied to a group. All those with a positive result are retested. - AKA: sequential or two‐stage testing - Sensitivity reduced - Specificity improved
Understanding causal effects may lead to...
- New interventions for disease - Preventive strategies
Parallel Testing
- Two tests are applied together. All those with either or both tests are considered to be positive. - AKA: simultaneous testing - Sensitivity improved - Specificity reduced, more false positives Positive Test ‐ Either test A or test B are positive Negative Test ‐ Both test A and test B are negative
Multivariable regression
-Analyse the data in a statistical model that includes both the presumed cause and possible confounders -Measure the odds ratio OR for each of the exposures, independent from the others -Logistic regression is the most common model in epidemiology -But explore the data first with stratification!
Metallic Compounds
-Arsenic -Mercury -Lead
Randomization
-Attempts to ensure equal distributions of the confounding variable in each exposure category. -Advantages: --Control for known and unknown confounders; permits straightforward data analysis. -Disadvantages: --Need control over the exposure and the ability to assign subjects to study groups. --Expensive
Controls for Bias
-Be purposeful in the study design to minimize the chance for bias --Example: use more than one control group -Define, a priori, who is a case or what constitutes exposure so that there is no overlap --Define categories within groups clearly (age groups, aggregates of person years) -Set up strict guidelines for data collection --Train observers or interviewers to obtain data in the same fashion --It is preferable to use more than one observer or interviewer, but not so many that they cannot be trained in an identical manner -Randomly allocate observers/interviewer data collection assignments -Institute a masking process if appropriate --Single masked study - subjects are unaware of whether they are in the experimental or control group --Double masked study - the subject and the observer are unaware of the subject's group allocation --Triple masked study - the subject, observer and data analyst are unaware of the subject's group allocation -Build in methods to minimize loss to follow-up
Environmental Hazards Found in the Work Setting
-Biologic hazards --Hospital employees, sewage workers, and agricultural workers may be exposed to hazardous biologic agents. For example, HIV may infect hospital workers through accidental needle sticks. -Mineral and organic dusts --Examples include coal dust (mining and black lung disease) and rubber dust (COPD). -Vapors and fumes are likely to become increasing hazards due to the growing use of chemical substances. -Vapors --Include organic solvents such as benzene, which may cause cancer and damage internal organs (particularly the liver).
Genetic code
-Blueprint of instructions for our body -Nucleic acids: A, C, G, and T --Adenine, cytosine, guanine, thymine -Sets of 3 make an amino acid (codons) -64 combinations, but only 20 amino acids --Multiple ways to "spell" an amino acid --Some codons signal the end of a gene -Codons are combined to make a gene --Instructions
Molecular Epidemiology
-Broader than genetic epidemiology -Greater precision in estimating exposure-disease associations --Improvement in measurement of exposures -Often use biomarkers --Serum levels of micronutrients --Intermediate marker of disease --Tumor markers --DNA fingerprinting
Health Effects Attributed to Environmental Exposures
-Cancer -Infertility -Reproductive impacts (malformations and low birth weight) -Infectious diseases such as malaria
Types of Agents
-Chemical agents -Metallic compounds -Electric and magnetic fields -Ionizing radiation -Allergens and molds -Dusts -Physical and mechanical energy
Human Exposures to Environmental Hazards
-Chemical agents (spills, pesticides, hazardous waste) -Electromagnetic radiation (high-tension wires) -Ionizing radiation -Heavy metals -Air pollution -Temperature increases from global warming
Drinking Water
-Chemical plants and nuclear facilities may contaminate ground water. -Chlorination of water supply has helped to decrease the incidence of gastroenteric diseases. -Lead and asbestos particles may be present in water and have potential for toxicity.
Case-Control Studies
-Compared with cross-sectional study designs, case-control studies can provide more complete exposure data. -However, precise quantitation of exposure and unobserved confounding may be difficult to achieve. --Example: bladder cancer among chemical factory workers (polycyclic aromatic hydrocarbons)
Ionizing Radiation
-Consists of either particle energy (e.g., highly energetic protons, neutrons, and α and β particles) or electromagnetic energy (e.g., γ-rays and X-rays). -Sources of ionizing radiation can be natural or synthetic. Natural sources --examples are radon and cosmic rays. --Radon is one of the largest sources of human exposure to ionizing radiation and may be the cause of nearly 20% of U.S. lung cancers, and is the leading cause of lung cancer among non-smokers. -Synthetic sources --examples are medical x-rays and nuclear generators.
Air Pollution
-Constituents of air pollution include sulfur oxides, particles, ozone, and lead and other heavy metals. -Lethal air pollution episodes have occurred worldwide. -Studies conducted in New York City, St. Louis, and Tennessee have shown a correlation between increases in daily mortality and increased air pollution.
Evidence for Genetic Factors
-Cross-sectional --Ask about disease status of individual --Disease status of relatives --Assess frequency of respondents with history -Case-control --Compare frequency of family history between both cases and controls -Cases more likely to be informed of disease -Cohort study
Organochloride Insecticides
-DDT (Dichlorodiphenyltrichloroethane) --Toxic to wildlife and persistent in the environment --Banned in 1972 (U.S.), but still used in other countries --Long half-life (2-15 years) --Biomagnification --May mimic estrogen
Study Designs Used in Environmental Epidemiology
-Descriptive studies --provide information for setting priorities, identifying hazards, and formulating hypotheses for new occupational risks. --Example: William Farr and Cornwall metal miners -Etiologic studies can be used to show exposure-effect relationships.
Toxicologic Concepts Related to Environmental Epidemiology
-Dose-response -Threshold -Latency -Synergism
Industrial Chemicals
-Exposure in occupational settings is up to 100 times higher than in the ambient environment. -Pesticides are of notable concern in the workplace due to the numbers of agriculture workers (up to 5 million). --Nausea, vomiting, vertigo, cancer, teratogenic effects, skin problems, and respiratory problems.
Examples
-Exposure to hepatitis B and aflotoxin --Induce genetic changes in p53 gene (suppress tumor formation) -Familiar adenomatous polyposis (FAP) --Mutations in APC, MLH1, MSH2, MSH6, or PMS2 genes --Elevated risk of colon, rectal, stomach, small intestine, liver, gallbladder duct, brain, skin, prostate, and endometrial cancers
Genetics and Public Health
-Genes and their interactions with modifiable disease risk factors on the health status of the population -Policies for genetic testing in prevention programs -Effective and high quality public health genetic programs
Noteworthy Community Environmental Health Hazards
-Hazardous waste sites -Air pollution -Nuclear facilities -Drinking water
Health Effects Associated with Work Environment
-Health risks for pregnant workers and the unborn fetus -Various lung diseases -Dermatologic problems -Bladder cancer among dye workers -Leukemia among workers exposed to benzene
Genetic Epidemiology
-Identification of inherited factors that influence disease and how variation in genetic material interacts with environmental factors -Focused on family groupings --Not individuals, specifically --Lack of independence -Techniques --Positional cloning or physical mapping --Functional cloning or functional mapping
Design of Family Studies
-Identify case proband and matched control proband -Enumerate relative of case and control probands -Prevalence of disease among case proband relatives are compared to prevalence of disease among relatives of control
Competencies
-Identify key sources of data for epidemiologic purposes -Identify principles and limitations of public health screening programs -Describe a public health problem (person, place, time, magnitude) -Explain the importance of epidemiology for informing scientific, ethical, economic, and political discussion of health issues -Comprehend basic ethical and legal principles pertaining to the collection, maintenance, use, and dissemination of epidemiologic data -Apply the basic terminology and definitions of epidemiology -Calculate basic epidemiology measures -Communicate information to lay people -Draw appropriate inferences from data -Evaluate strengths and limitations of epidemiologic reports
Odds Ratio continued
-In case-control study (CCS), we cannot calculate the cumulative incidence (CI) or incidence rate (IR), therefore, cannot calculate the RR "directly" -OR as a measure of association between exposure & disease is used when data are collected in case-control study -OR can be obtained however, from a cohort as well as a case-control study and can be used instead of RR
Physical and Mechanical Energy
-Include agents associated with accidental injury (e.g. automobile crashes). -Unintentional injuries are a leading cause of death within the age group 1-44 years in the U.S. -Also include such factors as noise, vibration, and extremes of temperature (global warming).
Nuclear Facilities
-Include weapons production plants, test sites, and nuclear power plants. -Studies of living in close proximity to nuclear installations have shown conflicting results regarding cancer rates. --Following the Chernobyl nuclear power plant accident, thyroid cancer rates increased near the reactor.
Pesticides: Used to Control Pests
-Insecticides -Herbicides -Rodenticides -4 Classes 1. Organophosphates 2. Organocarbamates 3. Pyrethroids 4. Organochlorides (organochlorines)
Hazardous Agents in the Work Environment
-Ionizing radiation -Infectious agents -Toxic substances -Drugs -Carcinogenic agents
Linkage analysis
-LOD score is given, based upon the probability of recombination event (proximity) -BRCA1 and BRCA2 genes on chromosome 17 were identified this way
LOD Score Linkage Analysis
-Logarithm of the Odds Score Linkage Analysis (linkage analysis) -Identify a DNA marker that co-segregates with the disease of interest and is considered strong evidence for the existence of a gene -Two genes that are in close physical proximity to one another on the same chromosome tend to be linked
Chemical Agents
-Many types used at home and at work --Household cleaning agents (soaps, air fresheners) --Automotive chemicals (volatile and corrosive solvents) --Paints --Pesticides -Potential effects on human health through acute toxicity, direct skin irritation, contact dermatitis, or long-term effects such as cancer.
Matching
-Matches subjects in the study groups according to the value of the suspected or known confounding variable to ensure equal distributions. -Frequency matching: the number of cases with particular match characteristics is tabulated. -Individual matching: the pairing of one or more controls to each case based on similarity in sex, race, or other variables. -Advantages: --Fewer subjects are required than in unmatched studies of the same hypothesis. --May enhance the validity of a follow-up study. -Disadvantages: --May be more costly, complicated to identify matches.
Restriction
-May prohibit variation of the confounder in the study groups. --For example, restricting participants to a narrow age category can eliminate age as a confounder. -Provides complete control of known confounders. -Advantages : straightforward, convenient, inexpensive -Drawback : limits generalizability , cannot control for unknown confounders.
Case-Control Family Studies
-Need to separate genetic and non-genetic risk factors in studies -Proband: the individual in a family who brings a disease of interest to the attention of the investigator -May include siblings or several generations of family members
Electric and Magnetic Fields
-Non-ionizing radiation -Sources include power lines, microwave ovens, stoves, clocks, cellular phones. -Los Angeles and Swedish studies found an association between residential proximity to power lines and childhood cancer risk. -U.S. and Norwegian studies found an increased risk for male breast cancer among male electrical workers. -Methodologic issues --Measurement of exposure -Cellular phones --Possibly brain tumors --accidents
Environmental Tobacco Smoke (ETS)
-Nonsmokers who work in a smoking environment have reduced pulmonary function compared to nonsmokers in a smoke-free work environment. -ETS causes 3,000 lung cancer deaths annually among nonsmokers. -ETS is associated with children's bronchitis, pneumonia, and asthma.
DNA to Protein
-Not all DNA contained in our cells is transcribed into protein -Within a region of DNA on a particular chromosome that codes for a gene, only certain segments are transcribed --Introns (not read) and exons (read) -Individuals differ from one another in terms of their DNA --Roughly the same number of genes but not the same sequence, and amount (base pairs) can vary
Hazardous Waste Sites
-Notorious sites in the U.S. include: Love Canal, NY; Valley of the Drums, KY; Times Beach, MO; Stringfellow acid pits, CA; Casmalia Waste Disposal Facility, CA. -Of great concern is the contamination of water supplies by toxic wastes. -Some possible adverse effects of hazardous waste exposure include birth defects, neurologic disease, and cancer. -Mixture of different substances
The Healthy Worker Effect Bias
-Observation that employed populations tend to have a lower mortality experience than the general population. -The healthy worker effect may reduce the measure of effect for an exposure that increases morbidity or mortality. --Sick people are less likely to work than healthy people
Lead
-Once widely used in paint and gasoline. -Associated with serious central nervous system effects even at low levels. -Has adverse effects on intelligence, behavior, and development. -Between 1988 and 2002, percentage of children with elevated blood lead levels declined steeply. -Still a risk factor for some
Ecologic Study Designs
-One use is the study of the health effects of air pollution. -Researchers measure the association between average exposure to air pollution within census tracts and the average mortality in those census tracts. -Unable to control for individual factors, e.g., smoking habits.
Selection bias with 'Survival Cohorts'
-Patients are included in study because they are available, and currently have the disease -For lethal diseases patients in survival cohort are the ones who are fortunate to have survived, and so are available for observation -For remitting diseases patients are those who are unfortunate enough to have persistent disease -Also known as 'Available patient cohorts'
Impact of Confounding
-Pulls the observed association away from the true association -Positive confounding --Exaggerates the true association --True relative risk (RR) = 1.0 and confounded RR = 2.0 -Negative confounding --Hides the true association --True RR = 2.0 and confounded RR = 1.0
Synergism
-Refers to a situation in which the combined effect of several exposures is greater than the sum of the individual effects. -Example: Study conducted among asbestos insulation workers clearly demonstrated a synergistic relationship between asbestos and smoking in causing lung cancer.
Psychosocial Aspects of Employment and Health
-Research topics include work overload and coronary heart disease; job stresses and absenteeism; and health effects of physical activity at work. -For example, one study found an association between work stress and periodontal disease.
Applications
-Screening for genetic susceptibility -Early detection of disease -Targeting of high-risk population subgroups -Recognition of underlying differences in host susceptibility -New therapies and treatments for disease
Bad Luck
-Several cases occur in same family --Not due to genetics --Obesity, mental illness, heart disease, cancer -Age --Older person with older relatives vs. younger person with younger relatives -Family size --2 vs. 21 family members
Bad environment
-Shared lifestyle among family members --Environment --Habits --Diets --Exercise practices --Alcohol /smoking
Examples of Selection bias
-Subjects: hospital cases under the care of a physician -Excluded: --Die before admission - acute/severe disease. --Not sick enough to require hospital care --Do not have access due to cost, distance etc. -Result: conclusions cannot be generalized Also known as 'Ascertainment Bias'
Necessary and Sufficient Causes
-Sufficient but not necessary causes --Cigarette smoking, asbestos, and radon are sufficient but not necessary causes of lung cancer -Neither necessary nor sufficient Example of sufficient cause: Sufficient causes are very rare in medicine, because it is exceptional that one exposure is by itself enough to cause disease. Usually exposures are much more common than the diseases they cause. Only about 5% of people who smoke get lung cancer. The measles virus virtually always causes people to get clinical measles, and rabies infection is always fatal.
Information bias
-Systematic error arising from inaccurate measurements (or classification) of subjects or study variables -Occurs when individual measurements or classifications of disease or exposure are inaccurate (i.e. they do not measure correctly what they are supposed to measure) -If patients in one group stand a better chance of having their outcomes detected than those in another group.
Stratification
-The process of or the result of separating a sample into several sub-samples according to specified criteria such as age groups, socio-economic status etc. -The effect of confounding variables may be controlled by stratifying the analysis of results. -After data are collected, they can be analyzed and results presented according to subgroups of patients, or strata, of similar characteristics.
Latency
-The time period between initial exposure and a measurable response. -Latency can range from seconds (acute toxic agents) to years (mesothelioma). -The long latency of health events in environmental research makes the detection of hazards difficult.
Arsenic
-Ubiquitous in nature -Varies in toxicity, depending on chemical form -Associated with skin, lung, bladder, kidney, and liver cancers
Mercury
-Used for the treatment of syphilis, as an agricultural fungicide, and in dental amalgams. -Responsible for Minamata disease, which occurred in the mid-1950s in Minamata Bay, Japan. --A neurological condition linked to the consumption of fish contaminated with mercury. -Fish advisory for swordfish and canned albacore
When can a RR be calculated from case control study?
-When the cases are representative, with regard to history of exposure, of all people with the disease in the population from which cases are drawn -When controls are representative, with regard to history of exposure, of all people without disease in the population from which controls were drawn -When the disease being studied does not occur frequently (rare disease)
Issues in interpreting an association
-You have a data set and it appears that there are some trends, relationships, etc. -What are the possible explanations? (Formulation of hypothesis) --Artifactual effect (Bias) --Chance occurrence (random errors) --Confounding --Causal explanation
Asbestos
-chemical agent -Strictly speaking, a mineral fiber. -Was used commonly for ship building, construction, insulation, and automobiles. -Associated with asbestosis, mesothelioma, and lung cancer.
Causes of morbidity and mortality include:
-exposure to toxic chemicals, -indoor and outdoor air pollution, -dangerous conditions found in the work place and -land use policies that encourage survival of pathogens and disease vectors.
Molecular Epidemiology
-helps to further identify groups that may be at risk of health outcomes -Will facilitate the ability of scientists to conduct etiologic research -Will increase our knowledge about the determinants of disease -Will contribute to the development of approaches for disease prevention -Will improve public health
Allergens
-substances that provoke an allergic reaction in susceptible individuals. -Allergic reactions range from dermatitis, asthma, and itchy eyes to anaphylactic shock.
Screening tests
-tests used for persons who are asymptomatic but who may have early disease or disease precursors -Identifies asymptomatic people who may have a disease
Diagnostic tests
-tests used for persons who have a specific indication of possible illness -Determines presence or absence of disease when patient shows signs or symptoms
Criteria for confounding continued
1. Associated with disease 2. Associated with exposure 3. Not in the causal pathway from exposure to disease
Attributes of the causal pie
1. Completion of a sufficient cause is synonymous with occurrence (although not necessarily diagnosis) of disease. 2. Component causes can act far apart in time. 3. A component cause can involve the presence of a causative exposure or the lack of a preventive exposure. 4. Blocking the action of any component cause prevents the completion of the sufficient cause and therefore prevents the disease by that pathway.
Genetic Epidemiology Asks 4 questions:
1. Does the disease of interest cluster in families? 2. Is the clustering a reflection of shared lifestyle, common environment, or similar risk factor profiles? 3. Is the pattern of disease within families consistent with the expectations under Mendelian transmission of a major gene? 4. Where is the chromosomal location of the putative gene?
Types of Information Bias
1. Interviewer Bias 2. Recall Bias 3. Observer Bias 4. Loss to follow-up 5. Hawthorne effect 6. Surveillance bias 7. Misclassification bias Two major types: --Non-differential --Differential
Common Types of Reliability
1. Intra‐subject 2. Intra‐rater 3. Inter‐rater 4. Instrument
Characteristics of a cause
1. Must precede the effect (proximate vs.distant) 2. Can be either host or environmental factors (e.g., characteristics, conditions, actions of individuals, events, natural, social or economic phenomena) 3. Positive or negative (lack of a preventive exposure)
Types of error
1. Random error (Chance) 2. Systematic error (Bias)
Selection Bias Examples
1. Volunteer Bias -Systematic error due to differences in characteristics b/w those who choose or volunteer to take part in a study and those who do not -Volunteer bias may be associated with outcome under study 2. Healthy worker effect -A phenomenon observed initially in studies of occupational diseases: workers usually exhibit lower overall death rates than the general population, because the severely ill and chronically disabled are ordinarily excluded from employment. Death rates in the general population may be inappropriate for comparison if this effect is not taken into account. --Question: association b/w formaldehyde exposure and eye irritation --Subjects: factory workers exposed to formaldehyde --Bias: those who suffer most from eye irritation are likely to leave the job at their own request or on medical advice --Result: remaining workers are less affected; association effect is diluted
Environmental exposures responsible for ___% of all cancers globally
19
WHO estimates that environmental factors are linked to as much as ___% of the global burden of disease and ___% of all deaths.
24, 23
Definition of Error
A false or mistaken result obtained in a study or experiment
Occupations
Academic worker Research workers Public health nurse epidemiologist Biostatisticians who focus on epidemiology Health care planner Pharmaceutical and biotech industry Consultants in epidemiology Program evaluation/community needs assessment specialists Epidemiology surveillance workers
Epidemiologic Definition of Causation
An event, condition, or characteristic that preceded the disease event and without which the disease event either would not have occurred at all or would not have occurred until some later time. - Rothman and Greenland
Systemic Error (Bias)
Any systematic error in the design, conduct, or analysis of a study that results in a mistaken estimate of an exposure's effect on the risk of disease Systematic error which often has a recognizable source, e.g., a faulty measuring instrument, or pattern, e.g., it is consistently wrong in a particular direction
Stratification Example
Bedsores and Mortality Study -Objective: The association between bedsores and death among elderly hip fracture patients. -Sample: 9,400 patients aged 60 and over, admitted with hip fracture to one of 20 study hospitals. -Methods: Medical charts were reviewed by research nurses in order to identify exposure and outcome.
Controlling confounding
Before data collection: -In the design --Randomization --Restriction --Matching After data collection: -In the analysis --Stratification --Multivariable regression
Bias
Can create spurious association when there really is not one (bias away from the null) Can mask an association when there really is one (bias towards the null) Primarily introduced by the investigators or study participants
Survival time must increase > lead time
Diagram from Ch.11
Reliability (Reproducibility or Precision)
Extent to which repeated measurements of a stable phenomenon - by different people and instruments, at different times and places - get similar results
Absolute Risk Example
Framingham heart study (cohort study): Coronary Heart Disease (CHD) Risk among Smokers -1-year risk of CHD among smokers (CIE+)* -Assess the cumulative incidence (CIE+) of disease in an exposed group (absolute Risk) -Assess the cumulative incidence (CIE-) of disease in unexposed group (absolute Risk)
Molecular vs. Genetic
Genetic:concerned with inherited factors that influence the risk of disease Molecular: uses molecular markers (in addition to genes) to establish exposure-disease associations
Dose-Response Curve
Graph that is used to assess the effect of exposure to a chemical or toxic substance upon an organism.
Criteria for Confounding
In order for a factor (X) to be a confounder, all of the following must be TRUE: --Factor X is associated with Disease B (risk factor or preventive factor) --Factor X is associated with Factor A (exposure) --Factor X is not a result of Factor A (not on causal pathway)
GWAS
International HapMap Project -Started in 2002 -Goal was to develop a haplotype map of the human genome -Generate 200,000 to millions of SNPs --14 million human SNPs reported, as of 2012
Genetic Variation
Mutation -Change in DNA that may adversely affect the host --Frameshift mutations: deletions or insertions of DNA bases --Changes to the chemical structures of nucleic acid -Not all changes result in mutations (AAA to AAG) --Mutations within introns -Even non-coding regions can have effect on protein production --Copy number variations (CNV) -Number of copies of a gene differ between individuals -May also play role in disease
Cause-Effect
One of the main objectives of Epidemiology is to identify the cause, or causes, of disease in humans
Survival time after diagnosis - lead time
Pre-detectable--> dectable, preclinical--> clinical--> disability or death Lead time: possible detection via screening
Applications in Epidemiology
Predictive value of biomarkers for disease -Improvements in the classification of exposures -More accurate definition of the risk groups -Increased specificity in the classification of disease -Greater understanding of etiologic mechanisms for disease
PPV vs. NPV
Probability (proportion) of those tested who are correctly classified PPV = Cases identified / all positive tests NPV = Noncases identified / all negative tests
Attributable fraction (AF) among exposed
Proportion of disease that could be prevented if the risk factor was suppressed from a population where 100% are exposed Equation: AF= [incidence (exposed) - incidence (unexposed)] / incidence (exposed) AF= (RR-1)/ RR Example conclusion: 80% of driving‐related deaths in population who drive fast were presumably due to fast driving
Population Attributable Risk Percent (PAR%)
Proportion of disease that could be prevented if the risk factor was suppressed in the population Equation: PAR%= [incidence (population) - incidence (exposed)] / incidence (population) x 100 Example conclusion: 44% of driving‐related deaths in population were presumably due to fast driving
Threshold
Refers to the lowest dose at which a particular response may occur.
Retrospective Cohort Studies
Retrospective: looking at exposed vs. unexposed -Various end points are used to study the effects of occupational exposures. --Morbidity: self-reports of symptoms and results of clinical examinations. --Mortality: comparison of mortality rates of exposed workers with nonexposed workers in the same industry.
Reliability (Repeatability) of Tests
See bell-curve graphs slide 26, ch.11
Factors That Contribute to Systematic Errors
Selection bias Information bias (Measurement bias) Confounding
The 2x2 Table
Slide 31, ch.11
Criterion-Based Causal Assessment Hill's Criteria (1897-1991)
The first complete statement of the epidemiologic criteria of a causality is attributed to Austin Hill (1897 - 1991). They are: 1. Consistency (on replication) 2. Strength (of association) 3. Specificity 4. Dose response relationship 5. Temporal relationship (directionality) 6. Biological plausibility (evidence) 7. Coherence 8. Experiment 9. Analogy
Non-differential Misclassification
The proportion of erroneous classifications is equal between the two groups Common because frequently measures of exposure or disease are inexact Because this always increases the similarity between the two groups, the bias is "toward the null" or the true effect is diluted
Odds Ratio
The ratio of the odds of exposure in the cases compared with the odds of exposure in the non‐disease (controls) Equation: Odds Ratio = odds of exposure (disease)/ odds of exposure (no disease)
Relative Risk
The ratio of the risk of disease in persons exposed compared to the risk of disease in persons unexposed Equation: Relative Risk = incidence (exposed)/ incidence (unexposed)
False: Measures of association are not measures of impact: It is not only the strength of association that matters but also the proportion of the population exposed to the risk factor.
True or False? -The relative risk of lung cancer among smokers is 9 -Therefore, if nobody was smoking, the incidence of lung cancer would be divided by nine
Differential Misclassification
When proportion of measurement error are different among the two groups being studied Can either under or over estimate the real relationship depending on where the error is made
Phenomenon of health: what is health?
World Health Organization: "a state of complete physical, mental, [and] social well-being and not merely the absence of disease or infirmity"
Sentinel health event
a case of unnecessary disease, unnecessary disability, or untimely death whose occurrence is a warning signal that the quality of preventive or medical care may need to be improved.
Genotype
a person's genetic makeup
Measure of potential impact
a statistic that quantifies the potential impact of removing a hazardous exposure
Measure of association
a statistic that quantifies the relationship between an exposure and a disease
Hawthorne effect
an effect first documented at a Hawthorne manufacturing plant; people act differently if they know they are being watched
Interviewer Bias
an interviewer's knowledge may influence the structure of questions and the manner of presentation, which may influence responses
Disease (D)
any health‐related outcome; the response; the dependent variable
Exposure (E)
any potential health determinant; an explanatory factor; the independent variable
Gene
basic unit of heredity
Hazard surveillance
characterization of known chemical, physical, and biologic agents in the workplace.
Statistical coherence
compatible with a reasonable statistical model (e.g. dose-response)
Biological coherence
compatible with current biological knowledge from other species or other levels of organization
Factual coherence
compatible with pre-existing knowledge
Theoretical coherence
compatible with pre-existing theory
Health surveillance
concerned with the health of individual workers or groups of workers.
Misclassification bias
errors are made in classifying either disease or exposure status
Phenotype
expression of genotype
Observer Bias
observers may have preconceived expectations of what they should find in an examination
Allele
one of two or more alternative forms of a gene that occurs at the same locus
Mass Screening
screening on a large scale of total population groups regardless of risk status.
Selective screening
screens subsets of the population at high risk for disease. - More economical, and likely to yield more true cases. - Example: Screening high‐risk persons for Tay‐ Sachs disease.
Inherited trait
segment of DNA molecule on a chromosome that determines the nature of an inherited trait
Locus
site or location on a chromosome occupied by a gene
Surveillance bias
the group with the known exposure or outcome may be followed more closely or longer than the comparison group
Random error (Chance)
the portion of variation in measurement that has no apparent connection to any other measurement or variable, generally regarded as due to chance, or because of sampling variability.
Definition of Epidemiology
the scientific study of the distribution and determinants of health‐ related states or events in human populations, and the application of this study to the prevention and control health problems.
Environmental epidemiology
the study of diseases and conditions (occurring in population) that are linked to the environmental factors (involuntary)
Loss to follow-up
those that are lost to follow-up or who withdraw from the study may be different from those who are followed for the entire study
Recall Bias
those with a particular outcome or exposure may remember events more clearly or amplify their recollections (e.g. patients suffering from MI are more likely to recall and report 'lack of exercise' in the past than controls)
Specificity
• A cause leads to a single effect, not multiple effects • Weak criterion - Often one cause-->multiple effects Cigarette smoking causes lung cancer, atherosclerosis, emphysema That it is not a specific cause does not nullify its role as a causal exposure for all of these diseases. • When found, it may be a useful criterion
Measures of Association
• Absolute risk • Relative Ratio - Relative risk (RR) - Odds ratios (OR) • Risk Difference (RD)
ROC Curves
• An ROC curve is a graphical representation of the trade off between the false negative and false positive rates for every possible cut off. Equivalently, the ROC curve is the representation of the tradeoffs between sensitivity (Sn) and specificity (Sp). • By tradition, the plot shows 1‐Sp on the X axis and Sn on the Y axis. • We are usually happy when the ROC curve climbs rapidly towards upper left hand corner of the graph. This means that Sensitivity and specificity is high. • We are less happy when the ROC curve follows a diagonal path from the lower left hand corner to the upper right hand corner. This means that every improvement in false positive rate is matched by a corresponding decline in the false negative rate 1 = Perfect diagnostic test 0.5 = Useless diagnostic test -If the area is 1.0, you have an ideal test, because it achieves both 100% sensitivity and 100% specificity. -If the area is 0.5, then you have a test which has effectively 50% sensitivity and 50% specificity. This is a test that is no better than flipping a coin.
Biological Plausibility
• Biologic mechanism proposed to explain the observed association(s) • Animal models and laboratory experiments contribute to our understanding of mechanisms and enhance causal relations Example: Presence of a serological marker of hepatitis B infection is associated (in Asia at least) with greatly elevated rates of liver cancer. That hepatitis B infection is a true cause of liver cancer is also supported by the finding of the viral genome in many liver cancers
Assessing the feasibility of screening
• Burden of disease - Effectiveness of treatment without screening • Acceptability - Convenience, comfort, safety, costs(=compliance) • Efficacy of screening - Test characteristics (Se, Sp) - Potential to reduce mortality • Efficiency - Low PVP - Risks and costs of follow‐up of test positives - Cost‐effectiveness • Annual Mammary screening (50‐70 yrs) = $30 - 50,000 /YLS • Annual Pap screening (20‐75 yrs) = $1,300,000 YLS • Balance of risks (harms) vs. benefits
What's a good value for the area under the curve?
• Deciding what a good value is for area under the curve is tricky and it depends a lot on the context of your individual problem. • What are the costs associated with misclassifying someone as non‐diseased when in fact they were? (False Negative) • What are the costs associated with misclassifying someone as diseased when in fact they weren't? (False Positive)
Phenomenon of disease: natural history
• Disease is a process that unfolds over time • Natural history - sequence of developments from earliest pathological change to resolution of disease or death
Importance of Screening
• Earlydetection - Leads to early treatment - Can lead to a decrease in morbidity and mortality - Can break the chain of transmission and development of new cases - Is often cost‐effective
Population Attributable Risk (PAR)
• Excess risk of disease in total population attributable to exposure • Reduction in risk which would be achieved if population entirely unexposed • Helps determining which exposures relevant to public health in community Equation: PAR= incidence (population) - incidence (exposed)
Approaches to Causality
• General model of causation (Causal pies) • Criterion-based Assessments (Hill's Criteria )
Analogy
• Has a similar relationship been observed with another exposure and/ or disease? • Example: Effects of Thalidomide and Rubella on the fetus provide analogy for effects of similar substances on the fetus.
Diagnostic vs. Screening tests
• Important part of medical decision making • In practice, many tests are used to obtain diagnoses
Disease natural history and prevention
• Knowledge of the natural history of disease is fundamental for effective prevention • Levels of prevention: - Primary: prevent the disease --[Primordial - prevent the risk factors] - Secondary: early detection and Rx - Tertiary: treat and minimize disability
Sensitivity
• Measures validity of screening tests • Ability to identify those with disease correctly - Minimizes false negatives... if test highly sensitive - Sensitive test with Negative result rules OUT disease Probability (proportion) of correct classification of cases Cases found / all cases Sensitivity= true positives/all cases = a/a+c
Specificity
• Measures validity of screening tests • Ability to identify those without disease correctly - Minimizes false positives...if test highly specific - Specific test with Positive result rules IN disease Probability (proportion) of correct classification of noncases Noncases identified / all noncases Specificity= true negatives/all non-cases = d/c+d
Negative Predictive Value (NPV)
• NOT inherent characteristic of a screening test • Percent of negative tests that are truly negative - If test result is negative, what is the probability that patient does not have the disease? NPV= true negatives/all negatives = d/c+d
Positive Predictive Value (PPV)
• NOT inherent characteristic of a screening test • Percent of positive tests that are truly positive - If test result is positive, what is probability that the patient has the disease? • Is affected by several factors - Specificity & specificity of the screening test - Prevalence of disease PPV= true positives/all positives = a/a+c
Conclude
• None of viewpoints can bring indisputable evidence for or against the cause-and-effect hypothesis and none can be required as a sine qua non except temporality.
Biases in Screening
• Referral Bias (volunteer bias) • Length Bias - Screening selectively identifies those with a long preclinical and clinical phase (i.e., those who would have a better prognosis regardless of the screening program) • Lead Time Bias - The apparently better survival that is observed for those screened is not because these patients are actually living longer, but instead because diagnosis is being made at an earlier point in the natural history of the disease • Overdiagnosis Bias (a misclassification bias) - Enthusiasm for a new screening program may result in a higher rate of false positives and give false impression of increased rates of diagnosis and detection - Also, false positives would result in unrealistically favorable outcomes in persons thought to have the disease
Evaluation of Screening Test
• Reliability and validity are central concepts in evaluating tests • Distinction between reliability and validity - Reliability: consistency of test at different times or under differing conditions - Validity: how well test distinguishes between who has disease and who does not
Test Characteristics and Population Tested
• Sensitivity and specificity are constant for a particular test • PPV and NPV vary dramatically, depending on prevalence of target condition in population tested Low prevalence: low PPV, high NPV High prevalence: high PPV, low NPV
GENERAL MODEL OF CAUSATION (CAUSAL PIES)
• Sufficient cause: -A set of conditions without any one of which the disease would not have occurred. (This is one whole pie.) • Component cause: -Any one of the set of conditions which are necessary for the completion of a sufficient cause. (This is a piece of the pie.) • Necessary cause: -A component cause that is a member of every sufficient cause.
Dose-Response Relationship
• The greater the exposure, the greater the observed effect • Absence of dose‐response does not rule out causal relationship
Strength of Association
• The larger the relative risk or odds ratio, the more likely the relation is to be causal • Examples RR = 10 (95% CI 8.2 - 11.9) One would need a large unrecognized bias or confounder to nullify the effect of the exposure RR = 1.2 (95% CI 0.99 - 1.9) It is more likely that an unrecognized bias of confounder could nullify the result
Temporality
• The only absolute criterion on the list • A cause must precede its effect - Case‐Control studies cannot establish this criterion with certainty - Cohort studies and clinical trials establish exposure before disease BY DESIGN
Consistency
• The replication of findings in independent studies enhances argument for causality • If the factor is a cause, we expect consistency of findings - Different designs - Different populations - Among sub‐groups
Measures of Potential Impact
• These measures predicted impact of removing a hazardous exposure from the population • Two types - Attributable risk among exposed - Attributable risk in the population as a whole *Does not make sense in the absence of causality!
Experiment
•Investigator-initiated intervention that modifies the exposure through prevention, treatment, or removal should result in less disease. •Example: Smoking cessation programs result in lower lung cancer rates. •Provides strong evidence for causation, but most epidemiologic studies are observational.