L1: Introduction and epidemiology Flashcards
What is epidemiology?
According to the World Health Organization:
“Epidemiology is the study of the distribution and determinants of health-related states or events (including disease), and the application of this study to the control of diseases and other health problems
Epidemiologists use statistics to quantify the effect of exposure on the outcome. Trying to prevent disease by eliminating or reducing exposure to these determinants.
Clinical epidemiology- might look at cancer patients so in this case not trying to prevent disease but focus on what can be done after so an outcome might not just be disease but progression, death, quality of life.
what do epidemiologists look at?
Outcome of interest → This refers to the disease or health condition being studied (e.g., cancer, diabetes, infections). Researchers want to understand how often it occurs and what factors influence it.
Exposure of interest → This refers to potential risk factors or causes linked to the disease, such as smoking, pollution, genetics, or diet. Researchers analyze whether certain exposures increase or decrease the risk of developing the outcome of interest.
Biostatistical methods → These are the statistical tools and techniques used to analyze data in epidemiology. They help researchers determine associations between exposures and outcomes, measure risks, and ensure results are statistically significant.
the scope of epidemiology?
illustrating how different factors influence disease progression and outcomes. Let’s break it down step by step:
Etiogenesis (E) → Exposure, Risk Factor, Cause
This refers to the origins of disease.
Example: Smoking (exposure) is a risk factor for lung cancer (cause of disease).
Diagnosis → Disease, Event, Effect (E)
After exposure to a risk factor, a person may develop a disease or condition (event/effect).
Example: A smoker may develop lung cancer (disease).
Prognosis (P) → Outcome, Death, Effect
Once a disease is diagnosed, it has different possible outcomes (e.g., recovery, chronic illness, or death).
Example: A lung cancer patient might recover, experience long-term complications, or die.
Intervention (P) → Outcome, Death, Effect
Interventions (like treatment, lifestyle changes, or public health measures) can change the course of the disease and influence outcomes.
Example: Chemotherapy (intervention) may prolong life or lead to remission (better outcome).
Summary:
Epidemiology studies how diseases develop (etiogenesis), how they are diagnosed, how they progress (prognosis), and how interventions impact outcomes.
The arrows represent causal relationships—for example, risk factors lead to disease, and interventions can alter outcomes.
claiming causality?
To claim causality many observational studies must prove associations and association should be biologically explained and then causality can be claimed between risk factor to disease/ cause and effect. Only type of study design where you can claim causality is randomised controlled trials. The difference is that it is not observed. Rather than observational that require multiple studies. Then hypothesis can be generated and experiments can occur.
clinical epidemiology?
Study of illness outcomes in persons seen by providers of health care
what is a confounder?
Confounding occurs when a third variable (the confounder) distorts the true relationship between an exposure and an outcome because it is independently associated with both. There is a particularly great potential for confounding in the absence of randomisation. In many observational studies of therapy, it is uncertain whether it was the treatment or the type of patient for whom that treatment was selected that effected disease progression, complications, or mortality.
Example of Confounding (Parity & Down Syndrome Risk)
Exposure: Number of children (parity)
Outcome: Risk of having a child with Down syndrome
Confounder: Maternal age
Why is maternal age a confounder?
Associated with exposure (parity) – Older women might have had more pregnancies, but this is not always true (a young woman could have multiple children).
Associated with outcome (Down syndrome risk) – Older maternal age is a well-established risk factor for Down syndrome.
Not an intermediate in the causal pathway – If maternal age directly caused higher parity (e.g., more children because they are older), then it would not be a confounder. However, age does not necessarily determine how many children a woman has.
Confounding in Observational Studies
In non-randomized studies, confounding can create uncertainty about whether the observed effect is due to the exposure or due to differences in patient characteristics (like age, health status, or socioeconomic factors). This is why randomization in clinical trials is crucial—it helps distribute confounders evenly between treatment and control groups, reducing bias.
study base?
Also known as Reference population
Is the Source of the study population
Population giving rise to the cases
Defined before cases appear by a geographical area or some other entity like a cohort study
person time?
Estimate of the actual time-at-risk in years, months, or days that all persons contributed to a study
Time will differ by people depending on endpoint: disease, death, end of follow-up
Time that everyone in your study contributes to the study. Reported usually as the mean total. Sum of everybody’s follow up.
Product of number of people in the study x average follow up time.
A case-control study does not have person-time because it does not follow individuals over time to measure the incidence (new cases) of disease. Instead, it starts with people who already have the disease (cases) and compares them to people who do not have the disease (controls) to look back at past exposures.
In a cohort study, person-time is used to track how long each person is at risk before they develop the disease.
Key Difference:
Cohort Study: “We followed 1,000 people for 5 years and counted how many developed cancer.” → Tracks person-time
Case-Control Study: “We took 500 cancer patients and 500 healthy people and looked at their smoking history.” → No person-time needed
types of outcomes?
Outcome
Disease
Disease progression
Death
Comorbidity
Questionnaire data
Biological endpoints – expression levels
prevelance vs incidence?
Term Definition How It’s Reported Example
Prevalence The proportion of a population that has a disease at a specific point in time % (proportion) or number of cases “10% of the population has diabetes right now”
Incidence The number of new cases that occur over a period of time Incidence rate: e.g., 55 per 100,000 people per year “Every year, 55 out of 100,000 people develop lung cancer”
Key Differences
Prevalence = Existing cases at one point in time (“snapshot”)
Incidence = New cases over time (“flow of new cases”)
How Are They Used?
Prevalence is useful for understanding the burden of disease in a population (e.g., planning healthcare services).
Incidence is used for studying risk factors and causes (e.g., how many new cases appear over time).
Incidence often reported as incidence rates e.g: 55 per 100,00 people. Ideally get age standardised incidence rates so you can make comparisons with different geographical areas.
Prevalence reported as number of people or proportion (%).
