Epidemiology: Review and Intro to Disease Frequency

Course policies and logistics (review from last class)

• Purpose of today: quick review of course policies, then main lecture on the approach and evolution of epidemiology.
• Textbook and access:
  • Three to four chapters posted; policy about how many chapters can be posted is not clearly defined.
  • First four chapters posted cover the material up to exam 1.
  • Textbook recommendations: Anne Ashegro (public health professor at BU) is a primary author; George Steech also referenced; you’ll be using Ashegro and Steech’s textbook as the core intro text.
  • Questions from end-of-chapter questions appear on the exam; answers posted in the back of the book.
• TAs and office hours:
  • Emily (in class liaison) available for logistics questions; can escalate to instructor if needed.
  • Kennedy O’Keefe handles EPI/class content questions and office hours; email for minor questions, but attending office hours is encouraged for better performance.
  • Zoom/office hours password issues noted; ensure access.
• Attendance and communications:
  • Attendance not formally taken; avoid sending excessive absence emails.
• Assignments and exams:
  • Nine assignments total; the lowest score is dropped; each assignment worth 5%; total assignment grade = 40% of final grade.
  • Four exams total; the lowest score is dropped; each exam worth 20%; total exam grade = 60% of final grade.
  • In-class Assignment #1 scheduled for Monday; extra space arranged to disperse students.
  • Assignment #2 posted Wednesday; individual assignment on prevalence and incidence; about one week to complete.
  • Late submissions: not accepted.
  • Submission window: Blackboard dropboxes close at 10:10 (start time of class); 10 extra minutes to submit to account for last-minute issues.
• Accommodations:
  • Accommodations processed electronically via the accommodations system.
  • If you plan to use accommodations (extra time, distraction-free environment, etc.) and take exams at the testing center, book those spots ASAP.
  • Exams may be taken on the scheduled exam date or up to two days after; early testing not allowed due to scheduling promises.
  • Space limitations at testing center due to pilot program; testing center is the recommended location for accommodated exams.
• General takeaways:
  • The instructor emphasizes using office hours and support resources to improve performance; there is a perceived strong correlation between attendance at office hours and exam performance.
  • The overall message: engage with the material, use resources, and plan accommodations early.

What is public health? Why epidemiology fits in

• Public health focus:
  • Emphasizes prevention of illness at the population/community level rather than treatment of illness at the individual level.
  • If you’re more interested in treating individuals (clinical route), consider medicine, PA, nursing, optometry, dental, etc.
  • Public health achievements contribute to longer life expectancy despite a small share of healthcare dollars. Specific points:
  • Roughly 2% of healthcare dollars go toward prevention, yet public health achievements have raised life expectancy by about $riangle LE \,=\, 40$ years overall, with about $35$ of those added years attributed to public health rather than medicine.
  • Major public health wins include vaccination, motor vehicle safety, workplace safety, control of infectious diseases via water/sanitation, safer food, healthier mothers and babies, family planning, fluoridation of water, and tobacco use recognition as a health hazard.
• Core epidemiology definition and purpose:
  • Epidemiology is the major science used in public health.
  • Derived from Greek: epi (upon), demos (the people), and logos (the study of).
  • Core concepts: distribution and determinants of disease; application to control health problems in human populations.
  • In practice, epidemiology often uses the term disease to refer to any health condition studied (pregnancy, autism, etc.), though this may feel imprecise outside epidemiology.
  • Goal: prevent, mitigate, or control health problems in populations.
• Four main objectives of epidemiology (as per the textbook):
  • Determine the extent of disease in a population.
  • Identify patterns and trends in disease.
  • Identify causes or risk factors.
  • Evaluate the effectiveness of prevention and treatment activities.
• Key premise:
  • Disease is not randomly distributed within a population; there are risk factors that influence who gets disease.
  • On the individual level, random variation can occur, but population-level analyses reveal systematic patterns that enable prevention and mitigation.
• Practical takeaway:
  • If there were no risk factors, we could not identify who is at greatest risk or how to prevent it.
  • The focus is on at-risk groups and modifiable determinants, enabling targeted public health actions.

A concise look at the history and roots of epidemiology

• Early observations and ideas:
  • Traces back to ancient Greece and Hippocrates; early observers described disease patterns (e.g., cholera, malaria) in ancient texts.
  • Before modern epidemiology, disease theories often relied on the four humors (blood, phlegm, black bile, yellow bile) and practices like bloodletting to rebalance humors.
• Time horizon of modern epidemiology:
  • Roughly 400 years of modern epidemiology history, with slow, uneven progress over that period.
  • A major acceleration after World War II, leading to rapid growth and development of study designs and methods.
• four pivotal historical anecdotes and milestones (as highlighted in the lecture):
  • The Black Death (14th century):
  • Transition from nomadic hunter-gatherer lifestyles to settled towns increased disease transmission due to overcrowding and sanitation issues.
  • Early public health actions included quarantine, isolation, and hospitals for isolation of plague patients.
  • John Graunt (John Durant in the transcript) (mid-1600s):
  • Began mortality counts: recording who died, from what causes, and at what ages to identify patterns.
  • James Lind and the scurvy trial (mid-1700s):
  • Often cited as an early clinical trial; sailors with scurvy were given various treatments; citrus (oranges/lemon juice) produced major improvements.
  • Demonstrated the potential for controlled trials and the eventual policy response (British Admiralty mandated lemon juice for sailors roughly 42 years later).
  • Smallpox vaccination (Edward Jenner, around 1800):
  • Inoculation with cowpox provided protection against smallpox; a landmark in vaccination history.
  • John Snow and cholera transmission (mid-1800s):
  • Father of modern epidemiology by many accounts; used data collection and mapping to investigate cholera outbreaks in London.
  • His Broad Street pump analysis led to removing the pump handle, supporting the waterborne transmission hypothesis.
  • Snow’s approach contrasted with miasma theory; he mapped cases, surveyed households, and examined water sources to strengthen the argument.
  • Koch and identification of disease agents (late 19th century):
  • Used microscopes to link diseases to specific agents (e.g., TB, anthrax, cholera).
• post-World War II study designs and landmark studies (emphasizing the development of evidence):
  • Streptomycin TB trial (late 1940s):
  • First randomized controlled trial: TB patients randomly assigned to bed rest vs. streptomycin + bed rest; the streptomycin group improved.
  • Doll and Hill studies on smoking and lung cancer (1950s–1960s):
  • Pioneered classic study designs: case-control studies followed by prospective cohort studies to test smoking-disease hypotheses.
  • Framingham Heart Study (started 1947):
  • A landmark cohort study tracking residents of Framingham, MA; now in a third generation; foundational for identifying risk factors for coronary heart disease; originally NIH-led, now co-run by NIH and Boston University.
• takeaway on study designs:
  • The modern era introduced robust study designs (randomized trials, case-control, cohort) that enabled stronger causal inference about disease determinants.
  • These designs paved the way for understanding risk factors and prevention strategies.

Measuring disease frequency: core concepts introduced today

• Core idea:
  • Epidemiology quantifies how often disease arises in populations to guide prevention and control.
• Key steps in measuring disease frequency (as outlined in the book and professor):
  • Define what constitutes a disease (case definition) to decide who has the disease vs. who does not.
  • Count the number of people affected (case counts).
  • Determine the size of the population from which cases arise (denominator) and consider the passage of time when relevant.
• Population concepts to distinguish between fixed and dynamic populations:
  • Fixed population: membership is permanent (e.g., Hiroshima survivors, graduates of a university).
  • Dynamic population: membership changes over time (e.g., residents of Boston, current BU students).
• Disease definition issues:
  • A careful, explicit case definition is crucial; for some diseases (e.g., autism), diagnostic criteria can be debated.
  • Consistent case definitions across time and studies are essential for valid comparisons.
• Frequency measures and the role of counts vs denominators:
  • Counting (counts) is the simplest frequency measure: the number of cases, often written as "counts" or "case counts".
  • Counts alone lack context without a denominator, which limits interpretation and resource allocation.
• When counts are most useful:
  • For very rare conditions, a simple count can trigger a public health response (e.g., meningitis outbreaks on university campuses).
  • In such cases, counts can prompt action even without denominators, but they do not reveal the relative burden across populations.
• Important caveats with counts:
  • Without a population denominator and follow-up period, it’s hard to compare across locations or time or to assess relative importance within a population.
  • A cited example from the textbook compares breast cancer frequency in two counties using only counts; without population size and follow-up duration, it would be misleading to allocate resources based purely on counts.
• Preview of next topics (to transition from frequency to incidence and prevalence):
  • Next class will cover prevalence and incidence, which refine the measurement of disease frequency by considering existing cases (prevalence) vs new cases (incidence) over time.
• Definitions and mathematical context (to be elaborated in upcoming classes but introduced here):
  • Case counts: $N_{ ext{cases}}$, the raw number of individuals with the disease in a population.
  • Population at risk and follow-up time: used to interpret counts meaningfully across populations and periods.
  • Preliminary formulas (to be refined in next class):
  • Prevalence: P = rac{N{ ext{cases (existing at a point in time)}}}{N{ ext{population at that time}}}
  • Incidence rate (incidence density): I = rac{N_{ ext{new cases}}}{ ext{person-time at risk}}
  • Cumulative incidence (risk over a defined period): CI = rac{N{ ext{new cases}}}{N{ ext{at risk at start}}}
• Practical example to reinforce concepts:
  • A county reports 5,000 influenza cases in a population; interpretation depends on the population size (denominator) and the observation period (time frame).
  • For meningitis outbreaks on campus, even a single case can trigger a public health response because the counts indicate a potential outbreak in a small, defined population (students in dormitories).
• Closing note:
  • The next lecture will deepen understanding of prevalence and incidence and illustrate how to compute and interpret these measures with real data.

Key terms recap and a few practical implications

• Public health vs epidemiology:
  • Public health aims to prevent disease at population level; epidemiology provides the data and methods to understand distribution and determinants to guide prevention.
• Core premise: disease is not randomly distributed across a population; there are detectable patterns and risk factors that enable prevention strategies.
• Historical landmarks to remember for exams:
  • Hippocrates and early observations; four humors theory; bloodletting.
  • Graunt’s mortality counts and pattern recognition.
  • Lind’s scurvy trial; citrus remedy; delayed policy change.
  • Jenner’s smallpox vaccination; early vaccination history.
  • Snow’s cholera map; waterborne transmission evidence; intervention by removing pump handle.
  • Koch’s identification of disease agents with microscopy.
  • Post-WWII study designs: randomized trials (streptomycin TB trial), case-control and cohort studies (Doll and Hill), and the Framingham Heart Study as a cornerstone cohort.
• Practical exam preparation tips:
  • Be ready to discuss how disease frequency is defined, counted, and interpreted.
  • Understand why counts alone can be misleading without a denominator or time component.
  • Be able to articulate the difference between fixed and dynamic populations with examples.
  • Anticipate questions about foundational studies and what study design they exemplify (e.g., randomized trial, case-control, cohort).

Connections to broader themes

• The evolution of epidemiology mirrors broader shifts in science: from descriptive observations to analytical methods and experimental designs.
• The emphasis on prevention and population-level thinking is central to public health, shaping policy decisions and resource allocation.
• Real-world relevance: historic outbreaks and landmark studies illustrate how data, measurement, and study design translate into public health action and policy changes.

Quick reference: sample counts vs current measures (glossary)

• Counts: raw number of cases, N_cases; useful for rare events or initial signal but lacks context without denominators.
• Population at risk: the subset of the population that could develop the disease during the study period.
• Denominator: population size used to contextualize counts (e.g., per 100,000 people).
• Fixed population: membership cannot change (e.g., graduates of a program).
• Dynamic population: membership changes over time (e.g., residents of a city).
• Prevalence: proportion of the population with the disease at a specific time.
• Incidence: rate at which new cases occur in a population over a period of time (risk over time).
• Incidence density vs cumulative incidence: incidence density uses person-time; cumulative incidence uses a fixed at-risk population over a time interval.

Summary takeaway

• Epidemiology is the study of how diseases are distributed and what determinants influence them, with the ultimate aim of preventing and reducing disease burden in populations.
• The field has a rich history of methodological advances that transformed health data into actionable public health measures.
• Early emphasis on counting and descriptive methods laid the groundwork for more sophisticated measures (prevalence, incidence) and robust study designs that inform policy and practice today.