HaDPop Glossary Flashcards
Census
The simultaneous recording of demographic data by the government at a particular time pertaining to all persons who live in a particular territory
Crude birth rate
The number of live births per 1,000 population in a given year
General fertility rate
The number of live births per 1,000 women aged 15-44 year olds in a given year
Total fertility rate (TFR)
The average number of children that would be born to a hypothetical woman in her life
Crude death rate
The number of deaths per 1,000 in a given year
Age-specific death rate
The number of deaths per 1,000 in a specific age range in a given year
Standard mortality ratio
Compares numbers of observed and expected deaths is age-sex distributions in a population were identical
Incidence
The number of new cases of a disease in a given time period
The tap
Prevalence
The number of people affected by a disease - no time period; NOT A RATE (the bath tub)
Prevalence = incidence x length of disease
P = I x L
Incidence rate
(number of new cases) / (population x time)
Normal unit = people-years
Incidence rate ratio
(Exposed incidence rate) / (not exposed incidence rate)
Absolute risk
Absolute risk measures the size of a risk in a person or group of people (Eg absolute risk of an individual developing Alzheimer’s as they age)
Relative risk
Relative risk compares a risk in two different groups of people
Eg Relative risk of diabetes between obese and normal weighted people
Confounding factor
A factor which is associated to both disease of interest and the exposure of interest without being part of the causal pathway
Confidence interval
A range of values so defined that there is a specified probability that the true value of a parameter lies within it (Eg 95%)
Statistical significance
If the results of a test have statistical significance, it means that they are not likely to have occurred by chance alone. In such cases, we can be more confident that we are observing a ‘true’ result
Cohort study
This study follows a people over a period of time to see how their exposure affects their outcome,
Starts with OUTCOME FREE individuals
Survival bias
The logical error of only concentrating on those who survived a situation, and overlooking those who didn’t. Can result in false conclusions being drawn
Observational study
A study where the exposure cannot be varied. Instead people are just observed
Odds ratio
Odds ratio compare the odds of an outcome in an exposed group compared to an unexposed group
Case-control study
The study compares a group of people with a condition with a group who do not, the study then looks back over time to see how conditions/ exposures vary
Retrospective study
Relies on data previously collected (as medical records reports). Recall bias (where information was reported incorrectly) can make this type of study inaccurate
Recall bias
Where information was recorded and reported incorrectly. Particularly important when considering retrospective studies
Prospective study
Has a specific outcomes and recruits suitable participants. Observes exposures and outcomes in these people over time (months/years)
Nested case-control study
A case-control study nested inside a cohort study
Epidemiology
The study of health related states or events in a specific population and the application of this study to control the health problems
Experimental study
This is a study where the conditions are under direct control of the investigator. This usually involves giving a group of people a variable which does not normally arise. A common use of experimental studies is to see how a treatment affects people compared to a group of people not receiving treatment
Non-Randomised study
In this type of study, individuals are not randomly assigned to a particular intervention type. Allocation bias and confounding factors may cause distorted results
Randomised controlled trial
This is a study where individuals are randomly assigned an intervention type - this could be two treatments, treatment + no treatment or treatment + placebo. This study type is the best way to find whether the treatment is effective
Open-Label
This means both investigators and participants know who is getting which treatment - there is no blinding. This can cause bias as patients may change their behaviour or investigators may cause measurement bias. If clinicians are involved, they could introduce non-treatment effect as they change their treatment or attention to the patient
Blinding
This is when you do not tell someone whether a person is receiving a treatment or not, and in some cases you don’t even tell them the outcome of their treatment. This person being blinded could be the investigator or patient (single-blind) or it could be both of them (double-blind)
As-treated analysis
The analysis of a randomised controlled trial where only those who have completed the trial are included - drop-outs are excluded. Makes the treatment look better than it really is
Pharmaceutical companies often use this analysis when selling their products.
This analysis loses the effect of randomisation.
Intention-to-treat analysis
The preferred analysis of a randomised controlled trial as it analyses everybody who is assigned to a treatment group at the beginning of the study - if people drop out, their follow up results are counted as their start results (issued no change following treatment - this prevents making the treatment look more effective than it is).
This analysis gives a more realistic view of what the treatment is like in real-world situations - accounts for likelihood for people continuing the full treatment. Preserves randomisation
Clinical equipoise
The reasonable uncertainty or genuine ignorance as to which treatment/intervention is better
All studies should follow :
1) Beneficence - Do good
2) Non-maleficence - Non-harming
3) Patient has own choice + makes their own decisions (if suitable)
4) Justice - Be fair
