Lecture 1 Epidemiology Flashcards
What is epidemiology?
The study of patterns, causes and effects of health and disease, defined in certain populations
Plays a major role in evidence based medicine
Epidemiology has identified risk factors, targets for prevention, statistics for prognosis and survival, and evidence for treatments
Epidemiology also informs health policy decisions - where the government puts its health dollar
Define Incidence
The number of cases ocurring over a specified period of time
Excludes already existing cases, as people with the disease no longer have the capacity to get the disease
Also excludes those who cannot get the disease - i.e. would exclude men from incidence of ovarian cancer
Prevalence
The number of cases in existence at one moment in time
How might a study involve incidence and prevalence?
1) The prevalence of a disease may be taken at the base-lin beginning of the study.
2) A follow-up is immediately undertaken of all those who do not have the disease
3) They are visited later on, after a certain period of time, and the incidence of new cases is taken
What are the two measures of incidence density?
1) Cumulative Density
The number of new cases occuring in a CLOSED population
2) May be measured in person time
Used in dynamic populations - with people constantly entering/leaving the population. Each person contributes a certain number of ‘person years’ to the study. The incidence density = (number of new cases) / (total number of person years). Incidence density is expressed as E.g. 16 cases per 300 person years
Things that affect frequency studies
Numerator = the number of cases
This is affected by how you define a ‘case’. Different classification systems may vary, and thus affect the value of the numerator.
Can also depend on where you choose to place your ‘cut off’ with measures that use a continuum (e.g. at what point cholesterol is considered ‘high’)
Denominator = the population
This is affected by who you involve in the study. Selection biases exist here. There must also always be the potential for the population in question to get the disease. I.e. do not include men in a study of incidence of ovarian cancer
What are the types of participant samplings?
Random Sampling = everyone in he population has equal chance of being selected
Probability Sampling = Investigators adjust sampling so that there is ‘equal representation’ of all minority groups in the sample (e.g. ensures equal representations of all ethnicities and SES)
Non-random Sampling:
Convenience sampling = e.g. patients visiting a certain medical clinic who are willing to participate
Grab Sampling = grabbing participants wherever you can find them
**Non-random sampling is open to more selection bias
Define Epidemic
High concentration of cases in an area
Endemic
Cases are confined to a certain area. E.g. this type of flu is endemic to the Phillipines
Epidemic Curve
Shows number of cases over a period of time, in a single specified population
Pandemic
When a disease is widespread
What are Prevalence Studies good for?
Planning health services to the ‘health landscape’ at hand
Highlighting areas that may need more attention for further research
Provide insight for future study
What are prevalence studies NOT good for
Inferring cause and effect
Prevalence can be affected by any treatments that prolong life. If you introduce a treatment that proplongs survival post-diagnosis, your study could look like the treatment has increased prevalence, when in reality it has increased survival
