Epidemiological terms Flashcards
(103 cards)
1
Q
What is the confidence interval?
A
*The 95% CI gives a range within which the true value lies and if we do the same experiment 100 times, the true value would like within that range 95% of the time.
Also an indication of precision (wide vs. narrow) and influenced by sample size.
Small sample size = wide CI
Large sample size = narrow CI
CI that crosses 1.0 indicates no difference between arms of the study e.g. exposed vs. non-exposed.
2
Q
What does the term statistical significance mean?
A
3
Q
What does a p-value mean?
A
“Probability value”. The likelihood that the result occurred due to chance rather than a true association.
E.g. p < 0.05 means that there is a less than 5% probability that the results is due to random chance rather than being a true association.
The larger the population, the more likely you are to find significant results of some kind, even if the result is clinically meaningless.
4
Q
What is a null hypothesis?
A
5
Q
What does power mean?
A
6
Q
What is a ‘Normal distribution’?
A
7
Q
What is a Poisson distribution?
A
8
Q
What is a t-test?
A
9
Q
When might you use a t-test?
A
10
Q
What is logistic regression?
A
11
Q
What is linear regression?
A
12
Q
What is Poisson regression?
A
13
Q
What is Cox proportional hazards regression?
A
14
Q
What is survival analysis?
A
15
Q
What are measures of central tendency?
A
16
Q
What are measures of variability?
A
17
Q
What is incidence?
A
18
Q
What is prevalence?
A
19
Q
What is risk?
A
20
Q
What is the Relative Risk and how is it calculated?
A
The relative risk (RR) is a ratio of probabilities of an event occurring in all exposed individuals versus the event occurring in all non-exposed individuals.
RR = (a/a + b) / (c/c + d)
If the disease condition (event) is rare, then the odds ratio and relative risk may be comparable, but the odds ratio will overestimate the risk if the disease is more common. In such cases, the odds ratio should be avoided, and the relative risk will be a more accurate estimation of risk.
I.e. for rare diseases use the OR and for more common diseases use the RR.
21
Q
What is the risk ratio and how does it differ from the rate ratio?
A
22
Q
What is the relative risk reduction?
A
23
Q
How is relative risk (RR) calculated?
A
24
Q
What is the absolute risk reduction?
A
25
What is the number needed to treat?
26
What is the number needed to harm?
27
What are odds?
28
What is an Odds Ratio?
An odds ratio (OR) is a measure of association between an exposure and an outcome.
The OR represents the **odds** that an **outcome** will occur given a particular exposure, compared to the odds of the outcome occurring in the absence of that exposure.
29
How is the OR calculated?
OR = (odds of event in exposed) / (odds of event in unexposed)
OR = (a/b) / (c/d)
= (ad)/(bc)
Odds ratios are most commonly used in **case-control studies**, however they can also be used in cross-sectional and cohort study designs as well (with some modifications and/or assumptions).
OR must be interpreted with statistical significant e.g. confidence interval.
30
What is the prevalence odds ratio?
31
What is the prevalence ratio?
32
What is the prevalence difference?
33
What is a hazard in the epidemiological context?
34
What is a Hazard Ratio?
A measure of how often (rate) a particular event (hazard) happens in one group compared to how often it happens in another group, over **time**.
Represents time survived to an event and may have nothing to do with death or prolonging life.
In cancer research, hazard ratios are often used in clinical trials to measure survival at any point in time in a group of patients who have been given a specific treatment compared to a control group given another treatment or a placebo.
**HR is not the same as RR.**
A HR of 1 means that there is no difference in survival between the two groups.
A HR of 2 = twice as many patients in the active group will have the event compared to the control in the next unit of time.
A HR of 0.5 = half as many patients in the active group are having the event compared to the control in the next unit of time
35
What is the Likelihood Ratio?
Used to assess the value of performing a **diagnostic test.**
Likelihood ratios compare the probability that someone with the disease has a particular test result as compared to someone without the disease.
It uses the sensitivity and specificity of the test to determine whether a test result usefully changes the probability that a condition (such as a disease state) exists.
Positive and negative LR.
LR+ = sensitivity / (1 - specificity)
LR- = (1 - sensitivity) / specificity
LR+ = 2 means a person with the disease is twice as likely to test positive.
LR- = 2 means a person without the disease is twice as likely to test negative.
Tests with a high LR+ and a low LR- have better discriminating ability
LRs are not influenced by disease prevalence. .
36
What is attributable risk?
37
What is the population attributable risk?
38
What is the population attributable fraction?
39
What is absolute risk reduction?
40
What is an attack rate?
41
What does the term sensitivity mean?
42
What does the term specificity mean?
43
What is the positive predictive value (PPV)?
PPV - The proportion of patients who have the disease who test positive.
44
What is the negative predictive value (NPV)?
NPV - The proportion of patients who do not have the disease who test negative.
45
What is standardisation?
46
What is direct standardisation?
47
What is indirect standardisation?
48
How would you calculate a standardised incidence or mortality rate (SIR/SMR)?
49
What is the mean difference?
50
What is the standardised mean difference?
51
What are measures of life expectancy?
52
What are measures of mortality?
53
What are ambulatory care sensitive conditions?
54
Explain chance in lay terms.
55
What are measures of chance?
Confidence interval and p-value.
56
Explain bias in lay terms.
57
What are the main types of bias?
Selection
Recall
Misclassification
Expectancy
Observation
. . . others
58
What is selection bias?
59
What is recall bias?
60
What is misclassification bias?
61
What is expectancy bias?
62
What is observation bias?
63
What is survivorship bias?
64
What is confounding? Explain in lay terms.
65
How can confounding be reduced?
66
What can be done in research studies to understand sources of and reduce impacts of chance, bias and confounding?
67
What is effect modification?
68
How do you control for effect modification?
69
What is causality or causation?
70
What is a causal relationship?
71
What are the Bradford-Hill criteria?
**DC ASPECTS**
**D** - Dose-response (biological gradient)
**C** - Consistency - reproducibility; consistent results in different studies, geographical locations, populations, time periods
**A** - Analogy - relationship synonymous with other similar relationships
**S** - Specificity - single cause for single effect
**P** - Plausibility - biological rationale
**E** - Experiment - experimental studies reproduce results
**C** - Coherence - coherence between epi and lab findings
**T** - **Temporality (required) - cause precedes outcome.**
**S** - Strength - magnitude of effect measure
72
What is another way to consider causality other than Bradford-Hill?
* **Direct evidence **– strength, size of effect not attributable to plausible confounding, temporal/spatial evidence, dose-response, reversibility
* **Mechanistic** – plausible mechanism of action (biological plausibility)
* **Parallel evidence** – coherence, replicability, similarity
73
What is the number needed to treat?
The Number Needed to Treat (NNT) is the number of patients you need to treat to prevent one additional bad outcome (death, stroke, etc.).
To calculate the NNT, you need to know the Absolute Risk Reduction (ARR); the NNT is the inverse of the ARR:
**NNT = 1/ARR**
Where **ARR = CER (Control Event Rate) – EER (Experimental Event Rate).**
NNTs are always rounded up to the nearest whole number.
74
What is intention to treat analysis?
Intention-to-treat analysis is a **method for analyzing results in a prospective randomized study** where **all participants **who are randomized are included in the statistical analysis and **analyzed according to the group they were originally assigned, regardless of what treatment (if any) they received.**
75
What is per protocol analysis?
Per-protocol analysis **only analyses data from participants who follow the protocol,** excluding their data after they become nonadherent
76
What is a DALY?
**Disability adjusted life year.**
DALYs are a measure of **disease burden.** One DALY representes the loss of the equivalent of one year of full health through **death or disability.**
**DALY **= Years of Life Lost (**YLL**) + Years Lived with Disability (**YLD**)
DALYs can be used to measure **trends and severity** of illness. They can be measured at the **national level.**
77
What is YLL and how is the YLL calculated?
YLL = Number of deaths X standard life expectancy at age of death.
E.g. If a 60 year old woman dies from stroke and life expectancy is 90, there were 30 years lost to life.
Multiply by the number of deaths in that age group. E.g. 1000 X 30 = 30,000
If people die younger, the DALY is greater.
78
What is YLD and how is YLD calculated?
Depends on the frequency of ill-health caused by disease or injury and the level of severity of disability experience.
**YLD = I x DW x L**
I = incidence cases
DW = disability weight (0 = perfect health, 1 = dead)
L = average duration of disability
Highly prevalent conditions do not necessarily results in high levels of YLD if the disability weight is low. Long-term causes which are, on average, successfully treated or managed have relatively low disability weights.
DWs are decided by experts in disabilities and experts in the computation of DALYs.
79
What is a QALY?
80
What is a HALY?
Health-adjusted Life Year. DALY and QALY are types of HALY.
81
What is HALE?
82
Describe the system for the surveillance of cancers and cancer-related issue in Tasmania / Australia.
83
Define disease surveillance.
84
What data systems are commonly used in the surveillance of chronic diseases and their risk factors?
85
What is the difference between screening and case-finding?
86
Define the terms efficiency, efficacy, and effectiveness.
87
Define the terms cost-effectiveness, cost-benefit analysis and cost-utility analysis.
88
What is type 1 error?
A **false positive**.
**Shows a relationship** when one **does not truly exis**t. Look at **sample size**, multiple comparisons threat.
89
What is type II error?
A **false negative**.
**No relationship** detected when one does truly exist. Look at **study power**.
90
What is multiple comparisons threat?
When a study compares many different exposures/interventions, the chances of false positives increases (i.e. type 1 error). Also called alpha-inflation. Can be corrected e.g. Bonferroni correction.
Affects **internal validity**
91
What is multiple treatments threat?
When a study compares many treatments it becomes difficult to know how well each treatment works individually; it may be that they are only effective in combination.
Affects **external validity.**
92
What is study power?
The probability that the study will detect a predetermined difference in measurement between the two groups, if it truly exists, given a pre-set value of α and a sample size, N.
93
What is reliability?
94
What is internal validity?
95
What is external validity?
96
What is efficacy?
97
What is effectiveness?
98
What is efficiency?
99
What is cost-effectiveness analysis?
100
What is cost-utility analysis?
101
What is cost-benefit analysis?
102
What is cost minimisation analysis?
103
What is the incremental cost-effectiveness ratio?
