Final Exam Flashcards

1
Q

Efficacy

A

The extent to which a specific intervention, procedure, regimen, or service produces beneficial results under IDEAL CONDITIONS.

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

Effectiveness

A

A measure of the extent to which a specific intervention, procedure, regimen, or service, when deployed in the field in ROUTINE CIRCUMSTANCES, does what it is intended to do for a specified population.

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

Hierarchy of Evidence Quality

A

1) Systemic Reviews and Meta-Analysis
2) Clinical Trials (in humans)
3) Longitudinal Cohort Studies
4) Case-Control Studies
5) Descriptive and Cross-Sectional Studies
6) Case reports and Case series
7) Personal opinion, subjective impressions, anecdotal accounts

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

Impact Factor

A

Total number of citations to articles appearing in that journal/ Total number of articles published

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

Three types of papers published in journals

A

1) Research Reports
2) Reviews of literature to summarize knowledge in a particular area
3) Commentaries

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

Components of a Research Report

A

1) Title
2) Author
3) Date of submission and acceptance
4) Abstract or Summary
5) Introduction
6) Materials and Methods
7) Results
8) Discussion
9) Conclusion

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

Scales of Measurement (3)

A

1) Nominal
2) Ordinal
3) Continuous
- Interval
- Ratio

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

Measures of Central Tendency

A

1) Mode
2) Median
3) Mean

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

Measures of Variability

A

1) Range
2) Interquartile Range
3) Variance
4) Standard Deviation
5) Coefficient of Variation
6) Standard error of the Mean

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

Standard Deviation

A

Used to measure variability of Individual subjects around a sample mean

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

Standard Error

A

Used to assess how accurately a sample mean reflects a population mean

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

Central Limit Theorem

A

In random samples of N observations, sample means will be normally distributed.

significance: allows us to make inference about the population from which our samples are drawn.

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

Confidence interval for a mean

A

Lower confidence bound and an upper confidence bound with the population mean contained within this interval (1-alpha) percent of the time.

For normal and t distributions, the confidence interval is centered around the mean. Confidence intervals get larger as you decrease the amount of error.

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

Research Hypothesis

A

Prediction based on the theory being tested on preliminary observations; on concentrations or guesses. (ex: gender affects intelligence)

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

Null Hypothesis

A

A mathematical statement, usually in population parameters, of no difference. Can be directional or non-directional.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
16
Q

Alternative Hypothesis

A

An objective hypothetical statement of the research hypothesis similar to the null hypothesis. We PROVE the alternative hypothesis by showing that the null hypothesis is not true.

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

Type I Error

A

Probability associated with rejecting the null hypothesis when it is true. (saying there are effects when there are none) ex: saying males are smarter than females when they are not.

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

Type II Error

A

The probability associated with accepting the null hypothesis when it is actually false. (saying there are no effects when there actually are)`

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

Power

A

The probability of rejecting the null hypothesis when it is false. The ability of a statistical test to detect a specified difference if that difference exists. Directly proportional to the sample size.

1-beta. Beta is Type 2 Error aka Null hypothesis is false but you accept it.

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

Alpha (Type 1) and Beta (Type 2) are _______ proportional to each other and the sample size N

A

inversely

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
21
Q

Methods to increase power

A

1) Increase Type 1 Error you are willing to tolerate
2) Increase the sample size
3) Increase the deviation from the null hypothesis you are willing to tolerate (big differences are easier to prove than small differences)
4) Decrease variability
5) Use a directional alternative hypothesis if appropriate.
6) Use the most efficient (most powerful) statistical test.

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
22
Q

Dependent Variable

A

The variable we measure and compare

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
23
Q

Independent Variable

A

The variable we manipulate

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
24
Q

Descriptive Studies

A

Descriptive patterns of disease occurrence in relation to persons, place, and time. (Data provided essential to public health administrators and epidemiologists)

How well did you know this?
1
Not at all
2
3
4
5
Perfectly
25
Correlational Studies
Measures representing characteristics of entire populations are used to describe disease in relation to some factor of interest such as age, utilization of health services, consumption of food etc.
26
Statistics used in Correlational and Regression Studies
1) Pearson coefficient: r- strength of association 2) Coefficient of determination: R^2- strength of association 3) Regression Coefficient: B1: Change in the dependent variable for every one unit change in the independent variable 4) Intercept: B0: Value of the dependent variable when the independent variable is 0.
27
r
Person Coefficient Ranges from -1 to 1. Unitless. Measures the strength of a relationship.
28
R^2
Coefficient of Determination Measures the strength of a relationship by explaining the % variance of the dependent variable accounted for by the independent variable Values range from 0-1 or 0-100%.
29
B1
Regression Coefficient | Shows the change in the dependent variable for every one unit change in the independent variable.
30
B0
Intercept | Shows the value of the dependent variable when the independent variable is 0.
31
Strengths of correlational studies
1) Quick 2) Inexpensive 3) Usually using available data/information
32
Limitations of Correlational Studies
Inability to link disease and exposure in an individual, data is collapsed, not specific to a person.
33
Case-Control Studies
Observational analytic study. Subjects selected on the basis of whether they do or do not have a particular type of disease. (and then look back retrospectively at a risk factor and its effects) Developed in 20th century when there was a shift from acute disease to chronic diseases.
34
Case-Control Study Advantage
- Allows study of diseases with ling latency periods - Efficient in time and cost - Allows for adequate numbers of diseased and non-diseased individuals to be identified. - Evaluation of RARE diseases - Can evaluate a wide range of potential etiologic (cause) exposures, and the interrelationships amongst them.
35
Case-Control Study Disadvantages
- Disease and exposure have already happened when the patient enters the study - Susceptible to bias: a) slection bias b) recall bias (differential reporting)
36
Case-Control Study Uses
- Test a Specific Hypothesis | - Explore a range of exposures among affected or non-affected individuals.
37
Cohort Study Types
1) Prospective 2) Retrospective 3) Both Pro and Retro-spective
38
Measures of Associations
Summary statistic that describes the association between exposure and risk of developing disease. Relative Risk/Risk Ratio Odds Ratio
39
Relative Risk (RR)
Likelihood of developing the disease in the exposed group relative to those not exposed. Incidence in Exposed group: Ie Incidence in non-exposed group: Io
40
Prospective Cohort Studies Advantages
- Temporal sequence between exposure and disease established - Assessing the effects of RARE exposures - Adequate numbers of exposed and non exposed individuals identified - Allows examination of multiple effects of a single exposure - Minimize potential for selection bias
41
Prospective Cohort Study Disadvantages
- Follow up for many years - Time consuming - Expensive - Boas associated with loss to follow-up
42
Cohort Studies: Retrospective
- Exposure/Outcome Disease have already happened at the time of study - Quicker and cheaper - Efficient for diseases with a long latency period requiring many years to accrue sufficient endpoints - Same disadvantages as in case-control
43
Difference between case-control and Retrospective Cohort
Case-Control: What was the risk factor/cause? (Odds ratio) Retrospective Cohort: Already know the risk factor/cause, studies exposure to disease outcome (can use both OR and RR)
44
Dependent t-test (One-sample t-test)
Used to see if your sample is different from a specified population
45
Independent t-test (Two-sample t-test)
idk
46
degrees of freedom (df)
df=N-1 | df= (ncon-1)+(nexp-1)
47
t
t- (difference of the means)/(SE of the difference of the means)
48
Independent t-test assumptions
1) Continuos measurement scales 2) Samples are drawn from populations with normal distributions 3) Samples are drawn from populations with equal variances 4) Samples are independent Remember: People want to be measured on the same scale, normal, equal, and independent.
49
Reasons for failure to demonstrate significance
1) it's the truth- they're not significant | 2) sample size is too small
50
ANOVA
Analysis of Variance- Divides variance of the entire experiment into two or more components: - Variance due to treatment effects - Residual, Unexplained, or Error Variance
51
Anova Table F
= MStreatment/MSerror
52
ANOVA Assumptions
1) Interval/ratio (continuos) measurement scales 2) Populations follow a normal distribution 3) Populations have equal variance 4) Independent groups
53
Pairwise Comparisons
Planned vs. Unplanned Planned: theoretical basis for comparison PRIOR to study Unplanned: Effects suggested by the data
54
Single Variable Tests
Z test: - we know the population and the standard deviation - estimate the mean from a sample t-test: -estimate both SD and mean
55
Two Variable Statistical Tests
Y dependent/Xdependent Nominal/Nominal: Chi-Square Nominal/Continuous: Logistic Regression Continuous/Nominal: t-test or ANOVA Continuous/Continuous Correlation and Regression
56
Two -Variable Tests
- Both variable Continuous - Independent nominal/ordinal, Dependent Continuous - Independent nominal/ordinal, Dependent nominal/ordinal - Independent continuous and Dependent nominal/ordinal
57
Both Continuous
- Correlation (r) strength of linear relationship | - Regression (linear regression): Usually assigns the dependent variable to the variable that would be estimated.
58
One Nominal One Continuous Dependent Continuous Nominal Variable has TWO levels Tests the null hypothesis that there's no difference between the groups versus the alternative that the two groups differ
- t-test | - anova
59
Requirements for One Nominal/One Contnuous
-2 groups are each normally distributed -groups have roughly same SD (if above not met, non-parametric tests can be used) -groups are independent of each other.
60
One Nominal One Continuous Nominal Variable has TWO OR MORE levels Tests the null hypothesis versus the alternative that AT LEAST ONE of the groups differs from some or all of the others.
-ANOVA further tests have to be done to see which groups are different.
61
Both Dependent and Independent variables are nominal or ordinal
chi-square
62
Dependent is nominal/ordinal and Independent is Continuous
Logistic Regression
63
Parametric
Normal Distribution - Linear Regression & Correlation - Two sample t--test (Independent t-test) - ANOVA - One sample t-test (Paired t-test) (dependen t-test) - Repeated measures of ANOVA
64
Non-Parametric
No Normal Distribution - Chi-Square Test - Wilcoxon Rank Sum Test - Kruskal-Wallis Test - Wilcoxon Sign Rank Test - McNemar Test - Friedman Test
65
Wilcoxon Rank Sums Test
Uses Ranks: Is the average rank for each group the same?
66
Wilcoxon assumptions
- Independent Groups | - Dependent Variable ordinal (or stronger)
67
Kruskal-Wallis Test
- Extension of Wilcoxon Rank Sum Test - At two groups it is like the t-test and ANOVA - can handle more than two groups - some assumptions
68
Dependent Group Tests
One Sample or Paired Data: - Wilcoxon Sign Rank Test:Ordinal Data - McNemar Test: Nominal Data More than one Group: -Friedman Test
69
Sign-Rank Test
``` Repeated measures (dependent observations)-ordinal data Ex: Evaluation of 2 facult by dental students ```
70
Assumptions of Sign-Rank Test
``` Ordinal measures (within and between) Dependency between measurements ```
71
McNemar Test
Ex: Two anesthetics compared McNemar Assumptions: - Two observations are related (not independent) -Categories are mutually exclusive -Expected frequencies under the null hypothesis should be 5 or more ((0.5)*(A+B))
72
Prevalence
Number of people in a population who have a disease at a GIVEN POINT IN TIME. Prevalence measure the frequency of all current diseases (old and new)
73
Incidence
A measure f the number of lesions/PERIOD OF TIME | Limited in that they only measure the number of new or initial lesions per unit of time
74
Caries Prevalence: DMFT/DMFS
``` Prevalence of caries in an individual D: Decayed M: Missing F: Filled T/S: Teeth/Surfaces ```
75
A surface with both caries and a restoration is scored
D
76
Maximum value for DMFS
128
77
Molars and Premolars have
5 Surfaces
78
Anterior teeth have
4 Surfaces