HLSC4P50 Final Exam PDF
Document Details
Uploaded by Deleted User
Tags
Summary
This document includes lecture notes on the topics of epidemiologic questions, analytic questions, experimental questions, and history, as well as landmark studies, famous trials, models, ecological studies, holistic models, nonclinical diseases, morbidity, and mortality rates.
Full Transcript
HLSC4P50 Final Exam Lecture 1 Epidemiologic Questions - What disease/ condition is present in excess? - Who is sick (Person) - Where do they live? (place) - When did they become sick? (time) - These are Descriptive questions Analytic Questions - Why did they become si...
HLSC4P50 Final Exam Lecture 1 Epidemiologic Questions - What disease/ condition is present in excess? - Who is sick (Person) - Where do they live? (place) - When did they become sick? (time) - These are Descriptive questions Analytic Questions - Why did they become sick? - How did they become sick? - More in depth to UNDERSTAND Experimental Questions - How can they be made well? - Which interventions are best - Inquire, propose or suggest History: Hippocrates - First to provide/ document rational vs supernatural explanations for disease based on OBSERVATION - Caused search of factors of disease like water, air, personal habits John Graunt - Quantitative approach to mortality; sex differences, infant rates - Numerical accounts of plague - Made the first life- table Edward Jenner - British physician - Cowpox material as vaccination John Snow - Cholera transmission through water pump Marc Lalonde - First modern government to propose to look beyond biomedical health care Landmark studies - Framingham study of CVD - British Doctors Study of tobacco and lung cancer - Nurses Health Studies of oral contraceptives, smoking etc. Famous Trials - Fluoride Supplementation - Salk Vaccine - HIV/ AIDS What is a model? - A systemic description of an object or phenomenon that shares important characteristics with the object or phenomenon Ecological Model - Triad of Agent, Host and Environment - Host Factors influence susceptibility - Agent factors induce disease - Environment influences both - You need these 3 for transmission of disease - Host; genetic susceptibility, resilience, status, behavior - Environment; conditions, social context, health care - Agent; virulence, infectivity, addictive qualities of a substance - Ecological Studies - Compare or describe characteristics of groups - Compare rates of exposures and outcome in populations rather than individuals - Looks for correlations Holistic Model of Health - Multiple factors that influence health - Other factors include SES, genetics, working conditions etc. Nonclinical/ Inapparent disease - Pre- clinical - Subclinical - Persistent - Latent Carrier Status - Someone who has the organism but is not infected, or evidence of clinical illness - Can infect others Morbidity - State of having an illness or condition - Can be acute or chronic - Diabetes, pneumonia, coronary disease - Can be measured with incidence or prevalence Attack Rate - Number of people at risk in whom a certain disease develops / total number of people at risk x 100% - Ex. Number of people who ate a specific type of food and became ill / total number of people who ate that food Mortality Rates Annual - Total number of deaths from all causes in a year / number of persons in the population in midyear X 100 000 Case Fatality - Number of deaths due to a disease / Number of people with the same disease X 100 Lecture 2 What is Descriptive Epidemiology - Describes the occurrence of an exposure or outcome in a population - Looks at patterns by person place or time tends to answer who, where, when - Used for developing hypotheses Epidemiologic Measures - Frequency - Potential impact - Effect/ association Counts - Counting the number of events in a population Beyond Counts - Including numerator and denominator - Ratio, Proportions and Rates Ratio - Relationship between 2 numbers - Numerator not always included in denominator - Ex. Sex ratio of 5:2 = 2.5:1 Proportion - Numerator included in denominator - Ranges between 0 and 1 - Ex. 2/4 = 0.5 = 50% Rate - Speed of occurrence of an event over time - Number of events observed for a given time / population in which the events occur (population at risk) - Has a period of time usually Crude Rates - # of events in a given time period / # of people in the population Confounding - A variable associated with both the exposure and outcome that does not lie on the causal pathway - Ex. Age with population and heart disease - Ex. Starting weight with exercise and blood pressure - Confounders are tied with the outcome a true risk factor - Non casually or casually tied with exposure - Can control by restriction criteria, matching, randomization, standardization, stratification - Confounders are tied with the outcome beyond the contribution of exposure What can we do for the Confounders? - Stratified rates: measure of a particular outcome that is calculated with specific subgroups based on age, sex, SES - So you could separate males and females Standardized Rates - Adjusted rate used to compare outcomes across population with different demographic structures. - Adjusts for the difference to have the same distribution of key characteristics - 2 Methods - Direct: applies the age specific rates of the study population to a standard population distribution to produce the standardized rate - You would need rate and population size Direct Method - Multiple rate and population size; this equal expected number of cases in each stratum for the standard population - Sum the expected number of cases in each stratum - Standardized rates = total number of expected cases in standard population / total population size - Direct = ∑(Rate ×Standard Population) / Total Standard Population - Use direct when you have age specific rates in the study population and age distribution of standard population Indirect Method (Standardized Mortality Ratio) - Uses rates from standard population and applies tgem to age distribution of the study population - You would need standard population rate and study population size of each stratum and study population total cases - SMR = Observed events in study population / expected events based on standard population rates - Use indirect method when you have: age specific rates in standard population and age distribution of study population and total number of cases or deaths in study population Incidence - Frequency of disease ONSET - What's NEW - Number of new events or cases of disease that develop in a population of people at risk during time interval - Cumulative and incidence density - New cases in a defined population / total population at risk during a certain time period (X time frame?) Cumulative - Subjects average risk of developing an events - # of new events occurring during a specific time period / population at risk during that time period - Underestimate of disease risk Incidence Density/ Rate or Person- Time Incidence - # of new events in a specified time period / total person time at risk - Numerator counts of disease events - Denominator is time at risk and free from disease - Can account for loss of subjects Prevalence - Measures population disease status - What exists - Proportion of people who have the disease Point Prevalence - Numerator: # of existing cases at a specific point in time / Total population at the same point in time - Rate base per 100, 1000, 10000 Period Prevalence - Numerator: # of existing cases during a specified time period / total population during the specified time, average or mid- interval population Person Years - How much time an individual contributes to observation - Allows us to account for people entering and leaving the study - 100 people for 5 years is 500 PY Two Basic Designs Case Report - Lowest level of evidence, but first line - New issues and ideas - Base of the pyramid - If multiple case reports show something similar, next step is case control to find a relationship - They are useful for new trends or diseases, side effects of drugs, rare manifestations - But they may not be generalizable, not based on systematic studies, can be seen as misleading or bizarre Case Series Design - Description of characteristics and outcomes among a group of people with either a siease or exposure over a period of time without a control - Retro or prospective, no randomization - Tend to describe population and outcomes rather than risks across groups Specialized Measure – Infant Mortality Rate - Infant mortality rate/ # of infant deaths during a time period X 1000 or 100000 Case Fatality Ratio - # of deaths from a specific disease during a time period / # of cases of the disease during the same time period Years of Potential Life Lost - YPLL or PYLL - Impact of premature death on society - Sum (average longevity – age at death) / # in population below average age Observational Studies - Formal comparison group - Analyze associations between exposure and outcome - Casual Evidence Relative Risk - Incidence in people exposed / incidence in non- exposed - RR = 1 No evidence of association - RR > 1 risk in harmful direction - RR < 1 risk in protective direction Odds Ratio - A/c|b/d=axd|bxc - Used in case control - OR estimates RR - Measure of association - OR = 1 no association - OR > 1 positive association - OR < 1 Negative association 3 Study Designs Cohort - Relationship between exposures - Assesses risk factors - Participants selected based on exposure status - Smoker vs Nonsmoker - Prospective and Retro - RR and incidence rates used here - Multiple and rare exposures - Not for rare outcomes Case Control - Selected based on outcome status whether they have disease or not - Those with disease are cases those without are controls - OR used - Rare outcomes - Can look for multiple exposures - Recall bias - Fewer resources - No incidence or risk since you start with cases Cross Sectional - Collect data at a single point in time - Assess exposure and outcome at the same time - Inclusion and exclusion criteria - Quick - Selection and information bias - Describes features of a population but not dynamics of a disease - Prevalence of outcomes Lecture 3 Validity - How well the results among the study participants true findings among similar people outside the study - All types of studies - Fewer or avoiding errors (Sources of error = chance; random sampling error, bias; systematic error in participants selection and confounding Internal - The extent to which the observed results represent the truth in the population we are studying not due to methodological errors Once you have calculated a measure of association you need to determine if the association is valid or if its casual External / Generalizability - Will be achieved by a sample that represents the target population - Dont confuse with selection bias - Can you apply it to a broader context? Sir Bradford Hill Criteria - Temporal Sequence - Strength of Association - Consistency of Association - Dose Response - Biological Plausibility - Experimental Evidence Temporality - For an exposure to cause an outcome it must precede the outcome - Cigarette smoking Strength of Association - Stronger the tie more likely to be casual - Strong ties less likely to be influenced by confounders Consistency - Same findings across various setting, population, places, times Dose Response - Increase exposure leads to increase outcome - Absence may occur if there is a threshold effect Biological Plausibility - Is the tie credible based on natural history Experimental Evidence - RCT reduces bias Random Sampling Error - Don't confuse with information bias and randomization - Role of chance - P Value only shows how much the observed results may be only due to chance in random sampling - Wide CI shows high probability of random sampling error; increase sample size Random Sampling - Selection technique where in the probability of selecting each sampling unit is known Bias - Systematic error in the design - Difference between observed association between exposure and outcome with the true association - Selection who and information where Selection - Error in the way people are selected and retained - Occurs when people have different probability of being included or retained in the study according to the exposure or outcome - Participation differs on exposure and disease - Can happen in case control and retrospectives - Also volunteer, non-response and membership bias - You can't control after done, must be avoided, but you can use a defined target population and maximize follow up Information - Case control and retro - Different techniques used to collect information High precision is reliable; repeated High accuracy is valid; proportion of true results Measurement Error - Interviewer of Observer - From participants (random or recall) Lecture 4 Prospective Cohort Studies - Must be free of outcome at outset - Classified on exposure status - Analytic - Not good when disease is rare - High evidence level Retrospective Cohort - Exposure status based on based records - Good when records exist Case Control - Based on outcome status - Disease is rare - Can control for confounding using matching, restricting logistic regression Cross Sectional - Compares characteristics of individuals - Data on exposure and outcome at one point in time - Based on membership for subjects Lecture 5 - Case control can answer the same question as a prospective cohort Matched VS Unmatched OR Calculation - Use unmatched when cases and controls are not paired, independent A x D | B x C - Tells is the odds of exposure among cases relative to odds of exposure among controls - Matched: # of pairs where the case is exposed, and the control is unexposed | # of pairs where the control is exposed and the case in unexposed Effect Modification - Effect of an exposure and outcome differs across levels of a 3rd variable - Relationship between exposure and outcome change depending on the value of another variable - Ex. Age with drug effectiveness - Does not distort effect unlike confounders - Mediators also called intermediate factors Information Bias - Non-Differential Misclassification: error in measuring or classifying exposure or outcome that affects all groups equally - Often Random - Ex. Alcohol and liver failure (people recall wrong evenly) - Association weak - OR towards null is weak association (1.0) - Time window is missed more likely Differential Misclassification - Different between study and comparison group - Errors in classifying exposure and outcome differ amongst groups - Ex. People with disease will recall differently - Also called recall bias - Overestimate or under - OR in either direction Interviewer Bias - Investigators are aware of persons outcome status Ecological Study - National differences in injury rates to public health measures - Get a hypothesis - Define ecological units - Get summary - Plot data on graph - Analyze - Units can be city, state, school Lecture 6 Cross Sectional Studies - Has a target and sample population - Exposure and outcome status at the same time - Different groups at the same time - Prevalence study - Snapshot of population Strengths - Quick and economical - One-time surveys no follow up - Samples easier to get - Useful when prevalence is the only measure of frequency Weaknesses - Timing is hard to establish - Not good for rare disease - Associations or lack of them seen with prevalent cases may not apply to incident cases Selection Bias: Prevalence- Incidence Bias - Type of selection bias that occurs when mild, fatal, resolved, asymptomatic cases are excluded - Not good cause this study measures what exists - Minimize selection bias by target population being representative, good sample size and being aware or non-responders Response Bias - This happens when those who respond to the surveys and studies are different from those who do not respond - Methods may attract those who are more health conscious - CI has 0 means not significant - CI no 0 means statistically significant Prognostic Studies - Prediction of outcomes among groups of people who have the disease - Outcomes can have death, survival, recovery - Use multiple variables to predict risk of future outcomes - Causality - Prognostic factors are predictors of disease outcomes like demographic, behavior, comorbidities - Ex. Breast cancer; age, smoking, obesity - Can resemble a cohort - Disadvantages are external validity, expense, loss to follow up, patients groups can have specific characteristics and experiences with time Lecture 7 Nested Case Control - Cases of a disease that occurred in a defined cohort are identified and for each a specified number of matched controls is selected from among those in the cohort who have not developed the disease - Case control within a prospective cohort - Population of people without the disease than branches off into disease to cases and no disease to samples and controls - Follow up over years - Its efficient to only test cases and controls that need to be tested - Best works for expensive or time-consuming exposures Examples of expensive exposure - Genetic tests - Repetitive tests - Large populations So... - You take a big cohort - Followed over time, documenting outcomes - Identify the cases as they occur (Incidence density cases) and then you sample controls at that same time to match the cases - So the cases and controls are matched on time - You can compare cases and matched controls for past exposures, but you have to account for matching in the analysis Steps in that Analysis - Analytic Steps - Contingency table analyses: cases and controls matched on time, creates on OR, ratio of discordant pairs - Regression Analysis: logistic regression, creates an OR, conditions on matching with adjusting for confounders Concordant and Discordant Pairs - To calculate this, data are treated as ordinal (In order) - To calculate the procedure compares the classifications for 2 variables on the same two items, if the direction is the same the pairs are concordant Strengths - Efficient, not all members of the original cohort require expensive testing - Flexible, allows testing of hypotheses not anticipated when the cohort developed - Reduces selection bias, cases and controls sampled from the same population Weaknesses - If the hypothesis is new, might not have measured all relevant confounders and effect modifiers (covariates) Case Cross over studies - One of a family self- controlled study designs - Each participant serves as their own control, and the analysis tests whether exposure times are tied with outcome times within individuals - Standard observational studies make comparisons between individuals - So in this design, comparisons are made within people by comparing exposures just before an event to exposures at another control time, eliminating confounding problems - Non cases are excluded - Smaller number of subjects - Longer duration though - Potential for unblinding and carry over effects Lecture 8 Experimental Studies RCT (Good example questions in this powerpoint) What is a controlled Clinical Trial? - Trial: an experiment where investigators manipulate the environment - Clinical: Subjects are people, and we are studying what happens to them - Controlled: have 2 groups (experimental and control) and conclusions will be based on a comparison between them Basic RCT Design - Population to eligibility to baseline assessment to randomization branhes into standard treatment to outcome over time and experimental to outcome over time, and those are compared Types of Bias - Selection - Performance: systematic differences between groups in the care that is provided or in exposure to factors other than interventions of interest. Blinding of study participants may reduce risk that knowledge of which intervention was received rather than the intervention itself, affects outcomes - Reporting: systematic differences between reported and unreported findings. - Non-significant differences (within study publication bias) Detection Bias - Systematic differences between groups in how outcomes are determined Attrition Bias - Systematic differences between groups in withdrawals from a study - Incomplete data - Attritions refers to situations in which outcome data are not available Phases of RCT Phase I Clinical trials of toxicity - Goal is to determine the best way to give the new treatment - Group of patients receive the same treatment at various doses - Does not allow for treatment effectiveness Phase II Trials - Determine if the treatment is active against the illness - To determine estimates of the effectiveness and toxicity of the new treatment - Same treatment Phase III Trials - Effectiveness of a new treatment in comparison to the standard treatment - Efficacy (ideal conditions) Phase IV - Helps establish long term effectiveness of a therapy after clinical implementation - Effectiveness in real world conditions Randomization - Random allocation - Purpose of this is get balance or equality of baseline characteristics - This will also balance unmeasured or unknown factors with no confounding - Goal is for the intervention to be the only thing different between the groups - Theres Random Allocations, Stratified Randomization, Blocked Randomization Types of Blinding - Single Blind: participants are unaware of the group they are in - Double Blind: participants and caregivers are unaware of participants group - Triple Blind, gold standard: participants, caregivers and analysts - Blinding is after randomization, allocation concealment before and the randomization sequence Intention to Treat Analysis ITT - All randomized subjects analyzed according to allocated group even if subjects never get the treatment they were supposed to - Analyze everyone according to treatment assignment Per Protocol Analysis - Only analyze those who completed treatment as assigned Loss to Follow up - 5 and 20 rule of thumb is too much loss - >20% poses threats to validity Strengths of RCT - Lowest susceptibility to bias - Best design to address questions of therapy efficacy - Clinically significant and statistically significant association = casual evidence Limitations - Expensive and time consuming - Not reflective of routine practice for it has highly selected populations and has controlled conditions - Unethical for studying hypothesis of harm Lecture 9 Systematic Reviews Synthesis - Integration of existing knowledge from previous research All review articles need - An extensive search of the literature - Extraction of key information from relevant articles - Clear and concise presentation of the information Clinical Tradition: The Narrative Review - A commentary consisting of a unique perspective about a topic using evidence - Must be organized by theme, methods, chronology by an expert - Becoming less common as science pushes for systematic methods Narrative Reviews - Only authors POV - No transparency - No guidelines for narrative reviews - The quality of narrative review may be improved by borrowing from systematic reviewed methods that are aimed at reducing bias in the selection of articles for review. - The dynamics of writing, organization, analysis and synthesis process are discussed Systematic Review Process - Unbiased, identifies all articles ever published on a narrow well-defined topic - Includes article screening; set inclusion criteria - Has Quality appraisals, abstraction and summary - Can identify areas of consensus and areas of disagreement - Have population, exposure and outcomes - Mentions interventions and etiology - Uses Pubmed/ Medline Cochrane reviews Inclusion Criteria's - Studies selected based on pre- defined criteria - Depending on the research question may restrict by study type, sample size, participant features, specific outcomes and exposures, duration of follow-up Publication Bias - Studies with positive or statistically significant findings are more likely to be published due to selective submission by authors or acceptance by journals Be Aware of - Multiple papers on the same data set - Papers in other languages other than English Quality of Evidence is based on - Study design - Methods used to minimize bias and chance - Cochrane = risk of bias Meta Analysis - Calculation of pooled statistic combining results of similar studies identified during a systematic review - High quality quantitative studies - Use of similar methods to analyze and collect data - But combing dissimilar studies could hide differences in study populations Process - State question and publish protocol - Comprehensive systematic strategy to identify every eligible article - Use predefined inclusion ad exclusion criteria - Extract statistics from each study - Assess quality and comparability - Conduct systematic review - Pool comparable statistical results into one summary statistics Comparability of Results - Only Homogenous (comparable) studies can be pooled - Use caution if heterogenous - Comparable results use similar measures Heterogeneity - Larger = more heterogeneity (Cochrans Q statistics I2) - If studies are not similar; random effects models - If studies are similar; fixed effects model - Both have a similar point estimate, but random models will have wider 95% CI for summary statistic Measures of Heterogeneity - What extent the results are consistent - If CI are given for the results of individual studies have poor overlap, this has statistical heterogeneity. - Use a chi squared test in a forest plot cochrane review - Assesses whether observed differences in results are compatible with chance alone, low p value gives heterogeneity of intervention effects - To quantify inconsistency: I^2 = (Q- df | Q) x 100 % Random Effects Model - Assumes that explanatory variables have fixed relationships with the response variable across all observations, but that those with fixed may vary from one observation to another. - Ex in fixed, stress and BP may very across people with different circumstances Fixed Effects Model - Random variables are treated as though they were in non random or fixed. - Some limitations, cant control for variables that vary over time Pooled Effect Estimates - Should follow the type of estimated effect, ex if studies are control, pooled estimate is OR Forest Plot - Summarize data in multiple papers in a single image - Takes all relevant studies asking the same question and finds a common statistic in the papers and displays them on a single set of axis - Compare directly - Large square means large sample size Funnel Plot - Assess potential of publication bias - Assumes small studies are more likely to have this bias than large studies and it is this difference with is detectable - Large papers will want the paper released anyway cause of effort - Small studies will not publish due to those negative results - Graphical representation of the size of trials plotted against the effect size they report. - As size of trial increases trials are likely to converge around the true underlying effect size Lecture 10 Infectious Diseases Epidemiology What is an Infectious Disease - Disorders caused by organisms such as bacteria, fungi, viruses, or parasites that can be passed from person to person - These diseases are grouped into 3 categories; high level of mortality, heavy disease burden on population, global repercussions due to spread - Etiology is cause or origin of an infection/ disease Handwashing - Semmelweis founded that its the most important tool in healthcare - Prevent spread of disease 4 Stages of natural course of disease - Susceptibility; lack of ability to resist an extreme agent - Presymptomatic disease; no symptoms between infection moment and onset of clinical disease - Clinical disease; manifesting symptoms - Outcome; recovery, disability or death Incubation Period - Time between exposure to infectious agent and the onset of symptoms Outbreak - Occurrence of disease more than normal expectancy - The number of cases varies according to the agent, and the size and type of previous existing exposure to the agent - Usually caused by an infection transmitted from person-to-person contact, animal to person or from the environment. Can also be from chemicals Outbreak Investigation - Establish existence - Confirm diagnosis - Establish case definition - Count cases - Describe cases by certain characteristics - Hypothesis - Test hypothesis - Epidemiological investigation - Prepare reports - Control measures - Surveillance is kept - With case control; get cases and controls, get likely incubation periods, question cases and controls (how they got the exposure), compare cases and controls with OR - With retrospective; get population for study, incubation period, categorize by past exposures, and then groups by disease experiences (attack rates by exposure) get RR Types of Outbreaks - Point source and propagative Types of cases - Clinical case; fits case definition - Possible case; exposed to agent - Probable case; exposed to agent and some lab evidence - Confirmed case; all 3 above with confirmed lab evidence - Case definition; needed to establish whether there is a link to a person and an outbreak, set criteria for this (person, place, time, clinical features) Epidemiological Curves - Distribution of cases overtime - Outliers - Magnitude of outbreak - Inferences about outbreaks pattern on spread - Time period of exposure - Intermittent (broken up), and then there continuous (which is continuous) Propagative Transmission (serial) - Onset of illness time (almost like a hill) Point Source - Duration – incubation time with number of cases (triangle shape curve) Endemic: area or community Epidemic: rapid of a community Pandemic: international Types of transmission - Vertical: mother to baby - Horizontal: person to person through bodily excretion - Types of contact; direct, indirect, vector borne (mosquito), food and water borne, droplets Infectivity - Ability of an agent to cause infection in a susceptible host Basic measures - Minimum, number of infectious particles required to establish infection - Proportion of susceptible people who are exposed who develop an infection - Case fatality - Number of clinical cases that get ill - Virulence; severity of disease after infection - Pathogenicity; ability of a microbe to induce disease (attack rate) Bioterrorism - Biological attack is the intentional release of bacteria that can sicken anything - Bacillus anthracis or anthrax is one of the most likely agents to be used in a biological attack Public health Response Health Surveillance - Passive; reportable disease list - Active; contact tracing - Types of prevention; primordial social distance, primary isolating sick, secondary contact tracing, tertiary physio after injury, quaternary vaccinations Types of Immunity - Natural and artificial - Acquired passive and innate - Herd; level of immunity in a population which prevents epidemics, the higher the R0 is, the higher proportion of the population will have to be vaccinated to get herd immunity. In order to prevent epidemics, proportion of the population that must be immunized is higher than 1 minus the inverse of the basic reproductive rate (1 – ¼ = 0.75, 75%) Types of Acquired Immunity - Active; natural with infection - Artificial with vaccine - Passive; natural with maternal antibodies - Artificial with antibody transfer Reproductive Rate - R0 average number of persons directly infected by an infectious case during their entire infectious period when they enter a totally susceptible population - R0 < 1 disease will go away - R0 = 1 disease will be endemic - R0 > 1 epidemic Variolation - Procedure of infecting healthy people with smallpox by taking material from smallpox patients Lecture 11 Future of Epidemiology - Epidemiology is the basic science of public health - Applied science - Systematic meta-analyses are the best of studies - Theres different fields; social of SDOH, clinical of patients, bioinformatics, molecular, pharmaco, human genomic (genetic variations between human health and disease, gene of breast cancer) - Epidemiology used to be biological based now its more diverse Epidemiological Surveillance What is it? Epidemiological surveillance is the continuous, systematic collection, analysis, interpretation, and dissemination of health-related data. It is essential for planning, implementing, and evaluating public health interventions. Roles in Daily Healthcare Work Monitoring Health Trends: a. Tracks the occurrence of diseases or health events, such as flu outbreaks or chronic disease prevalence. Detecting and Controlling Outbreaks: b. Identifies early signs of disease outbreaks to enable timely interventions. Evaluating Interventions: c. Measures the effectiveness of vaccines, treatments, or public health campaigns. Policy Formation: d. Informs decisions on resource allocation, healthcare delivery, and prevention strategies. Risk Communication: e. Provides healthcare workers with data to educate patients and the community about disease risks. Benefits Early Detection and Response: a. Prevents widespread outbreaks by enabling rapid containment measures. Data-Driven Decision-Making: b. Improves public health strategies through evidence-based insights. Resource Allocation: c. Directs resources to areas or populations with the highest need. Trend Identification: d. Tracks the progress of diseases or health conditions over time. Improved Health Outcomes: e. Reduces morbidity and mortality through timely interventions. Steps to Implement Epidemiological Surveillance Define Objectives: a. Establish what diseases or conditions will be monitored and the goals of the surveillance program. Develop a System: b. Design data collection methods, reporting formats, and feedback mechanisms. Identify Data Sources: c. Utilize hospitals, laboratories, community health centers, and registries as data contributors. Data Collection and Recording: d. Ensure standardized collection of data (e.g., cases, demographics, and risk factors). Analysis and Interpretation: e. Use statistical tools to identify trends, outbreaks, and anomalies. Dissemination of Findings: f. Share insights with stakeholders, healthcare providers, and policymakers to inform actions. Take Action: g. Implement prevention and control measures based on findings. Evaluation and Adaptation: h. Regularly assess the effectiveness of the surveillance system and make improvements as needed. Direct and Indirect Standardization Applications Differences Procedures to adjust two populations Interpretation of results (you need to write 2 3 lines reporting the results and its implication Standardization Practice Crude Rate: # of events in a given time period / # of people in the population (can multiply by 100,00 to get a better understanding number) Crude rates cannot be compared without age adjustment because age is major determinant of death, especially CHD since older population more likely to die. So crude rate will be higher if the population has older people. Crude rates are good for the overall burden of disease. Age specific rates To calculate the age-specific rates per 1,000, the formula is: Rate per 1,000= (Deaths | Population) ×1,000 Using the formula for each age group: 35-44: Rate= (4 |14, 531) × 1,000= 0.28 45-54: Rate= (3 | 8,010) × 1,000= 0.37 55-64: Rate= (25 | 6,208) × 1,000= 4.03 65-74: Rate= (25 | 5,077) × 1,000= 4.92 Age Adjusted Mortality Rate aR= ∑(wi ⋅ ri ) wi = weight from the reference population ri = age-specific rate in the study population Age Group Weight (wi ) Rate (ri ) Wi⋅ ri 35-44 0.361 0.28 0.361⋅ 0.28=0.1011 45-54 0.299 0.37 0.299⋅ 0.37=0.1106 55-64 0.194 4.03 0.194⋅ 4.03=0.7818 65-74 0.146 4.92 0.146⋅ 4.92=0.7183 AR =0.1011 + 0.1106 + 0.7818 + 0.7183=1.7118 Ri (rate was from earlier) age specific rate formula of deaths | population x 1000