PBM12_lecture1.pdf

4/23/2024 Epidemiology 1 PBM12 Summer 2023...

4/23/2024 Epidemiology 1 PBM12 Summer 2023 Copyright © 2014 Oxford University Press All rights reserved. Lecture 1: This work is protected by U.S. copyright laws and is provided solely for the use of instructors in teaching their courses and as an aid for student learning. No part of this Introduction; What is a population? publication may be sold, reproduced, stored in a retrieval system, or transmitted in any form or by any means, electronic, mechanical, photocopying, recording, or otherwise, without the prior permission of Oxford University Press. Textbook: Keyes KM, Galea S (2014). Epidemiology matters: a new introduction to methodological foundations. New York NY: Oxford University Press 1 2 Examples of questions epidemiologists ask Epidemiology is the science of understanding the  What is the incidence of myocardial infarctions between causes and distribution of population health so that we 2010-2020 among women born in 1950 in the United States? may intervene to prevent disease and promote health.  What are the causes of myocardial infarctions in this population?  If we were to change population dietary habits, what improvement in myocardial infarction incidence could we affect? 3 4 3 4 Six Dimensions of Health 1. Evolution of epidemiology 2. Seven steps to conduct an epidemiologic study 3. Farrlandia 4. Summary Merrill 2020 5 6 5 6 1  4/23/2024 Evolution of epidemiology 1. Evolution of epidemiology  Epidemiology is a relatively new as a formal scientific discipline 2. Seven steps to conduct an epidemiologic study  Practice of conducting epidemiologic studies is not new; 3. Farrlandia ‘counting’ health and disease goes back centuries  Many of design and analytic techniques that we use 4. Summary today arose in response to health concerns during 19th and 20th century 7 8 7 8 Epidemiology, a beginning  John Graunt  John Graunt – 17th century - pioneered approaches to tabulating population health and mortality in rates, ratios, and proportions 1620-1674 9 10 9 10 Epidemiology, a beginning  William Farr  William Farr – 18th and 19th century - developed more sophisticated life table approaches to understanding the force and burden of mortality 1807-1883 11 12 11 12 2  4/23/2024  William Farr Epidemiology, a beginning  John Snow – 19th century - used epidemiologic approaches to understand London cholera epidemic; developed and applied basic measures of 15 March 1813 – 16 June 1858 Temperature and Mortality of London by William Farr in Report on the Mortality of Cholera in England, 1848–1849 (London, 1852) https://www.youtube.com/watch?v=lNjrAXGRda4 13 14 13 14 Epidemiology, a beginning Epidemiology, a beginning  John Snow, a London physician, traced a major outbreak of cholera in the 1850s to its source.  Using logic, statistics, and mapping, Snow rejected the idea that cholera was carried in a cloud of bad air. Instead, he believed contaminated water was responsible for spreading the disease among the local population. 15 16 15 16 Epidemiology, a beginning Epidemiology history, continued  19th century – focus on infectious disease  Snow’s surveillance and response tactics would become  20th century – high-income countries shifted a foundation of modern epidemiology—the science of public health that is built on a working knowledge of toward non-communicable diseases probability, statistics, and sound research methods.  Mid 20th century – methods formalized (1970s) disease frequency and occurrence.  Late 20th century – Miettinen, Rothman, and Greenland - modern epidemiology (1980s) formalized central disciplinary principles 17 18 17 18 3  4/23/2024 A lifecourse approach to health An ecosocial framework production Social and Economic Policies Institutions Neighborhoods and Communities Living Conditions Social Relationships Individual Risk Factors `` Genetic/Constitutional Factors `` Pathophysiologic pathways Individual/Population Health Kaplan, G. What’s wrong with social epidemiology, and how can we make it better? Epid Rev 2004; 26: 124-135 Uauy, R. et al. Diet, nutrition, and the life-course approach to cancer prevention. J Nutr 2005; 135: 2934S-2945S 19 20 Current conceptual movements 1. Ecosocial perspective on population health – suggests policies, institutions, and characteristics of context contribute to the shaping of 1. Evolution of epidemiology health 2. Life course perspective – determinants of health are distributed across 2. Seven steps to conduct an epidemiologic study the life course and even before conception 3. Farrlandia Therefore, epidemiology understands causes of population health across 4. Summary levels of influence - from cells to society - and across life course. Examples for life course and ecosocial perspectives? 21 22 21 22 Epidemiology of consequence, seven steps Epidemiology of consequence, seven steps 1. Define the population of interest Descriptive epidemiology 2. Conceptualize and create measures of exposures and health  Step 1. Define the population of interest indicators  Step 2. Conceptualize and create measures of exposures and health indicators 3. Take a sample of the population  Step 3. Take a sample of the population  Step 4. Estimate measures of association between exposures and health indicators of 4. Estimate measures of association between exposures and health interest indicators of interest Assessing for causal effect 5. Rigorously evaluate whether the association observed suggests a  Step 5. Rigorously evaluate whether the association observed suggests a causal association causal association Conceptualizing and testing for interactions 6. Assess the evidence for causes working together  Step 6. Assess the evidence for causes working together 7. Assess the extent to which the result matters, is externally valid,  Step 7. Assess the extent to which the result matters (is externally valid) to other populations to other populations 23 24 23 24 4  4/23/2024 Farrlandia 1. Evolution of epidemiology  Examples often based on hypothetical geographic area, 2. Seven steps to conduct an epidemiologic study Farrlandia  Inspired by William Farr, pioneering epidemiologist and 3. Farrlandia statistician  Through use of Farrlandia examples, students will focus 4. Summary on applying foundational concepts to populations 25 26 25 26 What is a population? What is population health? A brief introduction to the graphical notation used in Epidemiology Matters 28 27 28 Set of individuals 100 individuals 20 diseased individuals Diseased Non-diseased 29 30 29 30 5  4/23/2024 Multiple sets of individuals 1. Defining a population of interest 2. Dynamic versus stationary populations 3. Summary 31 32 31 32 Seven steps 1. Define the population of interest 2. Conceptualize and create measures of exposures and health 1. Defining a population of interest indicators 3. Take a sample of the population 2. Dynamic versus stationary populations 4. Estimate measures of association between exposures and health indicators of interest 3. Summary 5. Rigorously evaluate whether the association observed suggests a causal association 6. Assess the evidence for causes working together 7. Assess the extent to which the result matters, is externally valid, to other populations 33 34 33 34 Defining a population of interest Defining a population of interest  A population is a collection of individuals, at moments in time, defined by at least one organizing  The definition of a population has implications for characteristic analysis, interpretation, and generalizability of results from epidemiologic studies – Social – Economic – Family (marriage and divorce). – Work and labor force – Geographic factors 35 36 35 36 6  4/23/2024 Defining a population of interest Defining a population of interest Epidemiologists: asking what causes we can intervene Thus, a central principle in epidemiology, and the first upon to produce the best health for the greatest number of people in the specific population. step in our epidemiologic rubric, is to define a population or group of scientific or public health To do so requires studying not individuals with good or interest. poor health, but rather the overall incidence and prevalence of health indicators across groups, asking why poor health is more common in some groups than in others.. 37 38 37 38 Two axes of population definitions What could be eligibility criteria? 1. Geographic area and time period of interest 1. Populations are defined by eligibility criteria 2. Characteristics of persons, events, or exposures for 2. Populations are defined by whether individuals which health-related factors are of interest move in and out of eligibility; i.e., populations can 3. Factors that promote successful study completion be dynamic or stationary. 39 40 39 40 Key terms What eligibility criteria define the following populations?  Health-related states or events a. Baby boomers – Disease states b. Inner-city teens  cholera, influenza, pneumonia, mental illness – Conditions associated with health c. People whose homes were flooded during a flood/hurricane  physical activity, nutrition, environmental poisoning, seat belt use, and provision and use d. Children with asthma whose parents will follow instructions to of health services administer a new medication – Events e. Adults with mental disorders  injury, drug abuse, and suicide f. Elderly people who physically inactive Merrill 2020 41 42 41 42 7  4/23/2024 Dynamic vs stationary populations Dynamic eligibility criteria for the population: allow movement in and/or out of the population. For example, people may move out of 1. Defining a population of interest geographic region and no longer be considered part of the population 2. Dynamic versus stationary populations of interest. Stationary eligibility criteria for the population: do not allow 3. Summary movement in and/or out of the population. For example, population may include all those born on a particular date only and follow them forward in time, never losing any nor adding others. 43 44 43 44 Population, defined by Dynamic population of Farrlandia geographic space and time over 1 year 45 46 45 46 Dynamic population of Farrlandia Dynamic population of Farrlandia over 1 year over 1 year 47 48 47 48 8  4/23/2024 Examples of populations, defined by characteristic, event, or exposure  If we are interested in understanding determinants of neonatal health, the population of interest is all newborn infants. Therefore, the defining event for inclusion is date of birth, and eligible individuals for population can be those within a year of their birth.  Pregnant women  Individuals over the age of 65  Individuals exposed to a natural or human-made disaster  Individuals of certain races or ethnicities  Individuals subject to particular laws 49 50 49 50 An example Populations, defined by study considerations  After the World Trade Center 2001 attack, a study considered Population of interest may be select group collected neonatal outcomes among offspring of women in California who specifically to achieve successful study gave birth six months after the attack comparing them to women  Good responders to our surveys who gave birth six months before the attack  Likely to continue to attend the health facility for follow- up visits  This study therefore defined population of interest defined by  Individuals who are healthy so that they are unlikely to a. Event or exposure = pregnancy die during the course of the study outside scope of b. Geographic location = California inquiry c. Time period = six months before and after September 11, 2001 51 52 51 52 An example Arrange the study populations according to whether entry is dynamic The Women’s Health Initiative is a study that assesses or stationary health benefits and risks of hormone replacement therapy after menopause. Eligibility is limited to women who had a. Baby boomers undergone menopause. b. Inner-city teens The study excludes a. Women with a history of certain medical conditions c. People whose homes were flooded during a flood/hurricane b. Women who were unlikely to be adherent to study d. Children with asthma whose parents will follow instructions to protocol administer a new medication c. Women who failed to meet other criteria e. Adults with mental disorders 53 f. Elderly people who physically inactive 54 53 54 9  4/23/2024 Questions that Need Epidemiology Questions that Need Epidemiology Diagnosis Health promotion and protection Is prostate-specific antigen a good test for prostate cancer? Do current school meals harm children's future health? Causes Health and disease surveillance Why did this patient suffer a stroke? Why are there 10-fold international differences in Is obesity the cause of metabolic syndrome? suicide rates? Treatment When will the next influenza pandemic occur? Is this the best treatment for Parkinson's disease? Health inequalities Is my surgery as good as that of everyone else? Why should life expectancy be nearly 5 years lower in Prognosis unskilled manual workers? What are the chances of a recurrent heart attack? Do health services reduce or increase health How long will this knee joint prosthesis last? inequalities? Merrill 2020 55 Merrill 2020 56 55 56 Epidemic, Endemic, and Pandemic Epidemic – Health-related state or event in a defined population above the expected over a given period of time Endemic – Persistent, usual, expected health-related state or 1. Defining a population of interest event in a defined population over a given period of time 2. Dynamic versus stationary populations Pandemic – Epidemic affecting a large number of people in many countries, continents, or regions 3. Summary Merrill 2020 57 58 57 58 What is a population? Seven steps What is population health? 1. Define the population of interest  Epidemiology is concerned with understanding the health of 2. Conceptualize and create measures of exposures and health populations indicators  Populations of interest can be defined by geography, space, time or 3. Take a sample of the population by characteristics of participants or of events of interest 4. Estimate measures of association between exposures and health  Regardless of eligibility criteria, population may be dynamic or indicators of interest stationary 5. Rigorously evaluate whether the association observed suggests a  Careful definition of the population base from which we conduct causal association an epidemiologic study underlies many of the core methods in 6. Assess the evidence for causes working together epidemiology 7. Assess the extent to which the result matters, is externally valid, to other populations 59 60 59 60 10