What is primary intervention?
This is intervention before the onset of disease, by altering or reducing risk factors.
What is secondary intervention?
This is intervention into the progression of a disease, through early diagnosis and effective treatment.
What is tertiary intervention?
This is intervention into an established disease through effective care.
What are the 6 parts of the 'chain of infection'?
• Infectious agent
• Exit Portal
• Entry portal
What is the hierarchy of evidence?
• Systematic review with at least one RCT.
• Experimental studies;
- Non-randomised clinical trial.
• Observational studies;
- 'Nested' case control study
- Cohort studies
- Case control studies
- Case-control studies
- Self-controlled case series
- Case series and reports
What is an Odds Ratio?
This is the ratio between the odds of the case group having a condition and the control group. If it is greater than one it suggests causation.
What is a cohort study?
• The group is chosen on the basis of their exposure.
• Follow up to determine who develops disease.
• Goor for rare causes and the effect of multiple exposures.
• Less prone to bias and time sequence easier to judge than case-control studies.
• Expensive and slow.
What is a case study?
• Subjects on chosen on the basis of whether they have the disease or not.
• Good for rare diseases.
• Quick and cheap
• Can be prone to bias and time sequence is difficult to judge.
What is a confounding factor?
The effect of a proposed risk factor is mixed up with other risk factors.
What is Incidence?
Incidence is the number of new cases of disease in a defined population within a period of time.
What is prevelence?
Prevelence is the number of individuals known to have a disease in a given population at a specified time period.
What is sensitivity?
This is a measure of how good a test is at diagnosing a disease in those who have it.
What is specificity?
Specificity is a measure of how good a test is at detecting no disease in those who do not have it.
What is a type I (alpha) error?
This is a false positive, where the study concludes that something is so, when it actually isn't. This can be controlled by significance. This is the incorrect rejection of a null hypothesis.
What is a type II (beta) error?
This is where the study concludes with a false negative, that is something is so but the study says it is not. This can be controlled through sample size and power of the study. This is the failure to reject a false null hypothesis.
What are the three models of stress?
1) Stimulus model
2) Response model
3) Transactional model
What is the stimulus model of stress?
This is where an external stimulus creates stress and the response does not matter.
What is the response model of stress?
Stress is as a result of response to a stimulus and is from within the organism itself.
What is the transactional model of stress?
This is the prevelent model and is a combination of the stimulus and response models.
This is a an interaction between the stimulus and response.
It lead to the study of coping mechanisms.
What are the three stages of response to stress and who developed the model?
1) Alarm reaction
2) Resistance - Tries to adapt and cope.
3) Exhaution - Stress exceeds coping mechanisms.
How are associations caused by 'chance' dealt with in studies?
Dealt with by statistics, defining a probability level which we find acceptable.
This is the Alpha level of probability.
The p-value deals with which erronerous association in a study?
What are the two types of bias and what are the differences between them?
• Selection - Referral bias, loss to follow-up bias, publication bias.
• Information bias - Recall bias
Which factors can be accounted for during the analysi stage of a study?
Bias cannot be dealt with.
What are residual confounding errors?
These are when all known confounders have been accounted for but there is some unknown confounding factor which is skewing the results.
How can a study deal with both confounders and unknown confounders?
What is standardisation?
This is a set of techniques used to remove as many confounding factors when comparing two populations.
What is direct standardisation?
This is where the rates are applied to the study population to generate expected number.
What is indirect standardisation?
This is where rates are applied to the reference population.
What is sensitivity?
The probability that a test on a patient with the condition will give a positive result.