Statistics Flashcards

(49 cards)

1
Q

Define selection bias.

A

A distortion in results due to non-random selection of participants.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
2
Q

True or false: Confirmation bias leads researchers to favor information that confirms their hypotheses.

A

TRUE

This bias can skew data interpretation and conclusions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
3
Q

Fill in the blank: Publication (or reporting) bias occurs when only _______ results are published.

A

Positive

Failure to publish results from valid studies, often as they showed a negative or uninteresting result.

Important in meta-analyses where studies showing negative results may be excluded.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
4
Q

What is observer bias in studies? Particularly in non-blinded trials.

Also called Expectation bias (Pygmalion effect)

A

Observers may subconsciously measure or report data in a way that favours the expected study outcome.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
5
Q

Observer bias is also known as the _______effect.

A

Pygmalion effect

Researcher favouring expected study outcome > than true performance

Pygmalion was a sculptor who fell in love with a statue he had carved.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
6
Q

Define attrition bias.

A

A bias resulting from participants dropping out of a study.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
7
Q

True or false: Recall bias affects participants’ ability to remember past events accurately.

A

TRUE

Common in retrospective studies where memory is relied upon.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
8
Q

Fill in the blank: Funding bias can arise when research is influenced by _______.

A

financial sponsors

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
9
Q

How does the Hawthorne effect (or social desirability bias) impact the behaviour of research participants?

A

It changes their behaviour due to the knowledge that it is being studied

E.g. Similar to when being assessed.

Like wearing a ‘hawthorn’, becoming centre of attention.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
10
Q

A 60-year-old woman participates in a new breast cancer screening programme that utilises an advanced imaging technique to detect tumours earlier than standard mammography.

As a result, she is diagnosed with breast cancer at an earlier stage compared to women using standard mammography.

However, statistical analysis shows that the overall mortality from breast cancer in the screened population remains unchanged compared to the unscreened population.

If a person says that this new test increases survivability and prognosis of breast cancer, what bias are they engaging in?

A

Lead-time bias

Where they let the ‘time’ (earlier detection) ‘lead’ in their thinking, to think this is a positive intervention, rather than the actual prognosis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
11
Q

Provide an example of Continuous Data.

A
  • Blood pressure
  • Temperature
  • Cholesterol levels
  • Age

Continuous data allows precise measurement and statistical analysis.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
12
Q

Give an example of Discrete Data.

A
  • Number of hospital visits
  • Number of medications taken
  • Number of lymph nodes involved

This data consists of whole numbers without fractions.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
13
Q

Which type of data is represented by whole numbers without fractions

e.g. The number of lymph nodes affected by cancer

A

Discrete Data

vs continuous (with fractions)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
14
Q

Data that represent categories without any intrinsic order or ranking

Examples include blood type (A, B, AB, O) and gender (male, female).

A

Nominal Data

By name

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
15
Q

Give an example of Nominal Data.

These are labels or names without order.

A
  • Blood type
  • Gender
  • Presence/absence of a symptom
How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Data with a meaningful order or ranking, but intervals between ranks are not necessarily equal/clarified

Examples include pain severity and cancer stages.

A

Ordinal Data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
17
Q

Provide an example of Ordinal Data

The categories have a clear order, but not necessarily numerical

A
  • Pain severity (mild, moderate, severe)
  • Cancer stages (Stage I, II, III, IV)
  • Patient satisfaction ratings (good-poor)

Vs nominal - only named types without an order, or categorical - named and by order

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
18
Q

Both nominal data and ordinal data are examples of ______________data.

A

Categorical Data.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
19
Q

A type of numerical data that can take any value within a range and have meaningful intervals

Examples include blood pressure and temperature.

A

Continuous Data

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
20
Q

A type of numerical data that take specific, separate values that has no fractions.

Examples include the number of hospital visits and medications taken.

A

Discrete Data

21
Q

Define Student’s T-test.

A

A statistical test used to compare the means of two groups.

22
Q

Indications to use the:

1) Unpaired student t-test
2) Paired student T-test

A

1) Unpaired (Independent) t-test: Used when comparing means from two independent groups, e.g. blood pressure in patients receiving drug A versus drug B.

2) Paired t-test: Used for related or matched participants, e.g. pre- and post-treatment measurements in the same patients.

23
Q

True or false: The Chi-square test assesses relationships between categorical variables.

A

TRUE

It evaluates how likely it is that an observed distribution is due to chance.

24
Q

Fill in the blank: The null hypothesis states there is ______ difference between groups.

A

no significant

25
What type of data is needed for a **Chi-square test**?
Categorical data, such as counts or frequencies.
26
True or false: A lower p-value indicates stronger evidence against the null hypothesis.
TRUE
27
What is the main assumption of the **Student's T-test**?
Data should be normally distributed in each group.
28
What is a **Type I error** in hypothesis testing?
Occurs when the **null hypothesis is incorrectly rejected when it is true** - e.g. there is actually NO difference. ## Footnote This leads to a false positive result, indicating a significant difference when none exists.
29
What is a **Type II error** in hypothesis testing?
Happens when the **null hypothesis is incorrectly accepted when it is false** - e.g. there is actually A difference. ## Footnote This results in a false negative, meaning the study fails to detect a difference that actually exists.
30
A 65-year-old woman is participating in a clinical trial evaluating a new antihypertensive medication. The initial study reports no significant difference in blood pressure reduction between patients taking the new medication and those taking a placebo. However, subsequent larger-scale studies reveal that the new medication significantly reduces blood pressure compared to placebo. What is the most likely explanation for the initial study's findings?
Type II error - the null hypothesis is accepted when it is false
31
Define **sensitivity** in medical testing.
The ability of a test to correctly identify individuals with a disease.
32
What does **specificity** measure?
It measures a test's ability to correctly identify individuals without a disease.
33
Fill in the blank: **Positive Predictive Value (PPV)** indicates the probability that a _______ test result is accurate.
positive
34
What is the formula for **Negative Predictive Value (NPV)**?
NPV = True Negatives / (True Negatives + False Negatives)
35
Define **Positive Predictive Value (PPV)**.
The probability that subjects with a positive screening test truly have the disease.
36
What does a high **NPV** indicate?
It indicates that a negative test result is likely accurate.
37
Fill in the blank: A test with 100% **sensitivity** has _______ false negatives.
zero
38
Fill in the blank: **Specificity** is calculated as True Negatives divided by _______.
the sum of True Negatives and False Positives
39
What does a low **PPV** suggest?
It suggests that many positive test results are false positives.
40
What is the focus of **Phase 1** in drug development?
* Small studies (e.g. 100) on healthy volunteers * Assess pharmacodynamics * Assess pharmacokinetics ## Footnote Phase 1 studies are crucial for understanding how a drug behaves in the body.
41
What is the primary purpose of **Phase 2** in drug development?
* Small studies (e.g. 100-300) on actual patients * Examine efficacy * Examine adverse effects ## Footnote Phase 2 trials help determine if the drug works as intended in patients.
42
What characterizes **Phase 3** in drug development?
* Larger studies (e.g. 500-5,000 patients) * Examine efficacy * Examine adverse effects * May compare drug with existing treatments * Studies of special groups (e.g. renal, elderly) ## Footnote Phase 3 trials are essential for confirming the drug's effectiveness and safety before approval.
43
What happens if a drug is shown to be safe and effective after **Phase 3**?
Drugs may be approved for marketing ## Footnote Approval allows the drug to be sold and prescribed to patients.
44
Relative risk is calculated by the forumula : _____________
New risk (%) / Risk in control (%)
45
A smoking cessation program lowers the risk of lung disease from 15% to 9%. What is the absolute risk reduction and relative risk reduction achieved by the program?
ARR: 15%-9% = 6% RRR: ARR/control risk = 6%/15% = 0.4 = 40% | RRR = 1=RR
46
A recent study shows that smokers have a relative risk of 2.0 (200%) for developing coronary artery disease compared to non-smokers. Based on this information, what does the relative risk of 2.0 indicate - in terms of the risk that a smoker have compared to non-smokers?
Risk of 200% = patients have a baseline risk * 200%= **2x more likely to have it** (Which is a 100% increase)
47
500 patients receive the new antiviral and 500 patients receive standard care. In the treatment group, 25 patients develop PHN, while in the standard care group, 100 patients develop PHN. What is the **relative risk of developing post-herpetic neuralgia** with the new antiviral medication compared to standard care?
0.25 (25/500 over 100/500) | equation: New rate/control rate
48
A relative risk (RR) of 300% means that the risk of VTE in women taking combined oral contraceptives is _______ times higher than in those not taking them.
3
49
This is asking the relative risk reduction. ##footnote Check this is correct by multiplying the risk percentage to the control number to get the new number.