Exam Flashcards

Question

Probability sampling

Answer 1

Random and not based on choice of researchers - Simple random – Stratified – Systematic – Cluster

Answer 2

Researchers pick – Convenience – Snowball – Purposive

Answer 3

People volunteer to participate in the study – One sample of the target population is more likely to be included/excluded than others – Self-selection or study inclusion/exclusion  Participants in screening programs tend to be healthier than those who don’t volunteer or comply  Impact on disease specific and overall mortality

Answer 4

Example of self-selection whereby outcome, over time, directly affects the exposure More healthy people may be found in potentially hazardous environments

Answer 5

Systematic difference between those in the study / intervention / exposure and those not Overcome by; – Randomisation – Transparent selection process

Answer 6

Process of allocating participants to groups is compromised Overcome by; – Randomisation – Concealment of the randomisation process

Answer 7

Once allocated, occurs when any differences in outcomes may be attributed to the ‘intervention’ or exposure – Guards against ‘placebo’ effect – Provides ‘natural’ prognosis Overcome by; – Blinding

Answer 8

Participants perform/behave differently due to being involved in the study (they know they are being observed)

Answer 9

Responses (positive/negative) to the perceived intervention

Answer 10

Attrition rate, or drop-outs, within a study Important to identify reasons for withdrawals due to; – Missing data – Adverse events – Motivation – Other… Overcome by; – Intention-to-treat analysis

Answer 11

CONSORT encompasses various initiatives developed by the CONSORT Group to alleviate the problems arising from inadequate reporting of randomized controlled trials (complete and transparent reporting)

Answer 12

Also known as ascertainment bias -An investigator may distort or misclassify the outcome measured if participant group is known Overcome by; – Blinding

Answer 13

Errors in measuring outcomes that lead to misclassification  Non-differential vs differential misclassification

Answer 14

When measurement error and misclassification occurs equally in all groups being compared  Due to; – Random error – Instrument bias  Results in dilution of ‘true’ effect

Answer 15

Measurement error and misclassification occurs to a greater extent in one group over others  Due to systematic error  Examples include; – Recall bias – Response bias – Interviewer / observer bias

Answer 16

Differences in accuracy or recollections of events/exposures from participants  Bias is unintentional and often based on expectation – MMR vaccination and autism

Answer 17

Often occurs in patient self-reported data  Bias is intentional – Portraying oneself in good light – Lack of understanding

Answer 18

Recording of information in different ways between interviewers / observers  Corrected by; – Standardised questions – Inter-observer / inter-rater reliability

Answer 19

‘When the Okies left Oklahoma and moved to California, they raised the average intelligence level in both states.’ 1. Individuals who are misclassified (Okies) who are moving, are below average for the current context (Oklahoma) 2. Individuals who are misclassified (Okies) are above the average for the context that they enter (California)

Answer 20

Occurs when the effect of the study factor on the outcome is mixed in the data with the effect on another (third variable, or confounder) Overcome by stratification

Answer 21

Only studies which show a certain result are published

Answer 22

Detailed report by one or more health professionals on the profile of a single patient Case series is a report on a series of patients with an outcome of interest -Strengths: Hypothesis generating, quick, cheap -Limitations: Generalisability, bias

Answer 23

 Also known as correlational studies  Units of analysis are groups, rather than individuals  Compares disease frequencies between; – Different populations during the same period of time, or – Same population at different time periods Strengths: Fast, easy, cheap Limitations: Bias, association only, possible misclassification

Answer 24

Exposure and outcome determined simultaneously  Cross-sectional studies measure; – Prevalence of disease – Presence/absence of exposure  Disease and exposure can be assessed at the same point in time in a cross-sectional study Strengths: Estimate prevalence of outcome, identifies association Limitations: Can’t generate cause and effect, only offers a snap-shot

Answer 25

 Compares the occurrence of possible cause in ‘cases’ and ‘controls’  Data is collected at one point in time  Exposures are collected at a previous point in time  Case-control studies are retrospective as the investigator is looking backward from disease to possible cause Strengths: Good for rare outcomes & long diseases, quick, cheap Limitations: Bias

Answer 26

 Involves follow-up of people with a common characteristic  The incidence of an outcome is compared between those exposed and those not exposed to a risk factor during the study time Strengths: Identifies natural history, temporal sequence Limitations: Loss to follow up, expensive, time consuming Retrospective cohort: Participants identified on the basis of previously recorded exposure

Answer 27

 An RCT is an experimental comparative study in which participants are allocated to treatment/intervention or control/placebo groups using a random mechanism  Participants have an equal chance of being allocated to an intervention or control group Strengths: Reduced risk of bias, cause and effect Limitations: Expensive, follow up duration, ethics

Answer 28

 Participants receive a series of treatments  Participants are then ‘crossed-over’ to receive the alternate treatment Strengths: Patients serve as own control, sample size Limitations: Feasibility, ethics, order of events?

Answer 29

Clusters rather than individuals are randomized – Geography – Communities – Social – Educational – Occupational

Answer 30

– A variety of separate requirements contributing to the cause  Obesity, insulin resistance, hypertension, low LDL cholesterol, high triglycerides

Answer 31

– When all components are part of the one sufficient cause that will lead to the effect  Metabolic syndrome

Answer 32

– When an outcome can’t develop in its absence  Environmental, biological, social determinants of health – e.g. breast cancer and BRCA gene

Answer 33

 Temporal relation (this is necessary)  Plausibility  Consistency  Strength  Does-response relationship  Reversibility  Study design  Judging the evidence

Answer 34

 The cause must precede the effect  Sometimes difficult to demonstrate with casecontrol and cross-sectional studies – Patients with stomach cancer have low levels of Vitamin C…

Answer 35

 Is the association consistent with other knowledge?  Issues to consider include; – Mechanism of action – Evidence from animal, biological studies

Answer 36

 Have similar results been shown in other studies?

Answer 37

 Strength of a relationship as measured by the relative risk (RR)  Rule of thumb: RR ≥ 2 considered to demonstrate a strong relationship

Answer 38

 A dose–response relationship occurs when changes in the level of a possible cause are associated with changes in the prevalence or incidence of the effect

Answer 39

 When the removal of a possible cause results in a reduced disease risk, there is a greater likelihood that the association is causal – Use of NSAIDs and gastrointestinal bleeding / ulcers – Challenge / re-challenge strategy

Answer 40

How strong is the study design

Answer 41

 Temporal relationship is essential, then… judge the evidence

Answer 42

 Surrogate endpoint, or marker – Measure of an effect of a treatment / exposure that may correlate with a real clinical endpoint – But the relationship / correlation is not certain  Why use surrogate endpoints? – Regulatory approval (phase III trials) – Reduction in duration and cost of studies

Answer 43

The main thing you are investigating

Answer 44

Side effects

Answer 45

Confounding factors

Answer 46

Numerical  Continuous (blood pressure)  Discrete (number of children) Categorical  Nominal (blood type)  Ordinal (measure of pain)

Answer 47

1. Scheduled standardized interview – Structure is uniform 2. Non-scheduled standardized interview – Phrasing may be altered 3. Non-standardized interview – Conversation

Answer 48

 The design process; 1. Introductory statement explaining its purpose 2. Demographic questions 3. Simple factual questions 4. Complex questions asking for opinions or attitudes 5. Closing question / statement

Answer 49

– How useful did you find meeting with the practice nurse before attending your appointment and did the online system speed up the process?

Answer 50

– We have recently introduced practice nurses into our clinics to ensure that patients who are discharged from the emergency room have continually care from the ED to home. What are your thoughts on this innovation?

Answer 51

 Have you ever had a chest x-ray? – Yes / No  What was the natural colour of your hair when you were 20 years old? – Red / Blond / Brown / Black  How many packs of cigarettes would you smoke in a week? – One / two / three / four / five +  Which of the following symptoms are you currently experiencing? – Headache / Dizziness / Cough / Temperature

Answer 52

refer to OneNote

Answer 53

refer to OneNote

Answer 54

refer to OneNote

Answer 55

refer to OneNote

Answer 56

In statistics, inter-rater reliability is the degree of agreement among independent observers who rate, code, or assess the same phenomenon – Reliability across multi investigators

Answer 57

– Reliability of a single investigator

Answer 58

A statistic which can inform researchers about the inter-rater agreement

Answer 59

For binary (dichotomous) outcomes  Absolute risk reduction/increase (ARR/I)  Relative Risk (RR)  Relative risk reduction/increase (RRR/I)  Number needed to treat/harm (NNT/H)

Answer 60

Is simply the difference in the proportion of subjects with the outcome of interest in each group. ARR = X (control) – Y (intervention) remember to put it in %

Answer 61

Is defined as the probability of an event in the active treatment group divided by the probability of an event in the control group. A relative risk of 1 is the null value or no difference RR = Y ÷ X or (intervention ÷ control)

Answer 62

Can be thought of as a standardised measure of the absolute risk reduction It can be expressed as the absolute risk reduction divided by the probability of an event in the control group RRR = 1 – RR or x-y / x * 100

Answer 63

Is the number of patients who would have to receive the treatment for 1 of them to benefit The number needed to treat is the reciprocal of the absolute risk reduction NNT = 1 ÷ ARR

Answer 64

An odds ratio is a statistic that quantifies the strength of the association between two events A*D / B*C (refer to OneNote)

Answer 65

 Most research uses analysis from a sample to estimate the value in the true population  A CI is the range of values within which we are confident that the true population value lies  Bounded by the lower confidence limit (LCL) and upper confidence limit (UCL) - If they cross 1 then there is no difference  We can also use CIs to compare two sample means and determine whether the difference between groups is statistically significant  Critical value here is ZERO  If the CI for the difference in means includes zero, the difference is not statistically significant  If the CI for the difference in means does not include zero, the difference is statistically significant

Answer 66

 Mean difference (MD) is the absolute difference between the means of two continuous variables  A mean difference of 0 means no effect

Answer 67

 P = Probability  A p-value indicates the probability that an observed test result is due to chance (i.e. not a true result)  Standard accepted cut-off point is p<0.05 – This means that there is a 5/100 (5%) chance that our result is due to chance – We are willing to accept that we will be wrong 5% of the time

Answer 68

 The effects of possible prognostic factors, which are relative to one another, can be interpreted using a Hazard Ratio (HR)  Hazard ratios are similar to a Relative Risk, with the difference being that the Hazard Ratio is derived from the time-to-event analysis  A Hazard Ratio of 1 (HR=1.0) suggests no benefit

Answer 69

Case is a source of infection for others * Failure to detect early and treat is detrimental Immunity * Prior exposure may confer immunity. * Vaccination is important measure * Heard immunity Urgency in response * Prompt response is important. Surveillance and preparedness is key Multiple prevention measures is critical * Prevent exposure and transmission * Treatment is a key prevention * Increase resilience of population

Answer 70

Direct transmission * Direct physical contact Indirect transmission * Vehicle-borne -Contaminated inanimate materials or objects (fomites). * Vector-borne * -Mechanical: No reproduction of agent (i.e. fly) in vector -Biological: Reproduction (i.e. Mosquitoes) in vector Airborne * Droplet * Microbial aerosols usually the respiratory tract. * Dust * The small particles from soil clothes, bedding or contaminated floors by wind or mechanical agitation.

Answer 71

refers to the constant presence of a disease or infectious agent in a population within a geographic area .

Answer 72

refers to a disease that occurs infrequently and irregularly.

Answer 73

refers to an aggregation of cases grouped in place and time that are suspected to be greater than the number expected

Answer 74

is a noticeable, often small but sudden, increase over the expected number of epidemiologically linked cases

Answer 75

refers to an increase, often sudden, in the number of cases above what is normally expected in that population in that area.

Answer 76

is an epidemic with a P ( p for passport) - A new pathogen that spreads from person to person across the globe".

Answer 77

-Reproduction number (R) is the average number of new infections caused by 1 infected individual -Basic reproduction number (R0) is the reproduction number (R) when the entire population is susceptible.

Answer 78

* An "epidemic curve" shows the frequency of new cases over time based on the date of onset of disease. * The shape of the curve in relation to the incubation period for a particular disease can give clues about the source. -refer to the OneNote for different curves

Answer 79

Point source (common source) * Single source of contamination (food borne outbreaks) * Many people get sick concurrently (1 incubation period) * Steep upslope and More gradual downslope * Exposure period can be predicted from curve

Answer 80

* Can affect many people, over a long period of time E.g. person to person spread * Prolonged exposure * Can quantify transmissibility of propagated outbreaks using R

Answer 81

1. Confirm existence of an outbreak 2. Verify the diagnosis 3. Construct a working case definition 4. Find cases systematically, line list 5. Perform descriptive epidemiology 6. Develop hypotheses 7. Evaluate & refine hypotheses, perform additional studies as necessary 8. Implement control and prevention measures 9. Communicate findings 10. Follow up and maintain surveillance -details in OneNote

Answer 82

A systematic review is a summary of the medical literature that uses explicit and reproducible methods to systematically search, critically appraise, and synthesize on a specific issue. It synthesizes the results of multiple primary studies related to each other by using strategies that reduce biases and random errors. Key factors to consider to determine how good it is – Inclusion criteria – Inadequate literature search – Publication bias – Inadequate assessment of study quality

Answer 83

Meta-analysis is a research process used to systematically synthesise or merge the findings of single, independent studies, using statistical methods to calculate an overall or 'absolute' effect.2 Meta-analysis does not simply pool data from smaller studies to achieve a larger sample size.

Answer 84

A systematic review attempts to gather all available empirical research by using clearly defined, systematic methods to obtain answers to a specific question. A meta-analysis is the statistical process of analyzing and combining results from several similar studies.2 Apr 2018

Answer 85

A literature review explores and evaluates the literature on a specific topic or question. It synthesises the contributions of the different authors, often to identify areas that need further exploration. Comparison to systematic review in OneNote

Answer 86

Good for detecting publication bias

Answer 87

refer to OneNote for picture

Answer 88

 A meta-analysis does not just simply add up numbers across studies!  It identifies results from individual studies and calculates a weighted average  Simply adding studies would break; – Power of randomisation – Reduce variance – Overestimate significance

Answer 89

used in systematic reviews to determine the cumulative findings in a multitude of studies, examples in OneNote dichotomous (1) continuous (0)

Answer 90

 Heterogeneity refers to the diversity that exists between studies in a review – clinical – methodological – statistical  Identifying heterogeneity – visual inspection of the forest plots – chi-squared (chi2) test (Q test) – I2 statistic to quantify heterogeneity

Answer 91

Sensitivity analyses examine how the results vary under different assumptions – e.g. re-analysing data using ‘low’, ‘medium’ and ‘high’ quality studies good quality trials vs poor quality trials

Answer 92

Subgroup analyses are meta-analysis on subgroups of the studies – e.g. sex, age, drug doses

Answer 93

– Usually performed when a patient who has a clinical problem & is more likely to have the disease

Answer 94

– Performed on healthy people, before they have symptoms

Answer 95

Pre-test probability + test info = Post-test probability  Pre-test probability – Clinical assessment – Personal experience – Published data (prevalence of a disease)

Answer 96

-Sensitivity and specificity always and only relate/refer to the test, they don't relate to the patient -How accurate is the test that we're using -Sensitivity measures true positive (if somebody has the disease how likely are they to test positive)

Answer 97

-Sensitivity and specificity always and only relate/refer to the test, they don't relate to the patient -How accurate is the test that we're using -Specificity measures true negative (if somebody doesn’t have the disease how likely are they to test negative)

Answer 98

These are kinda like sensitivity and specificity but they relate to the participant/patient -Positive predictive value (the chance that if somebody tests positive that they'll actually have the disease) Sensitivity and specificity relate to the test and the test won’t change. The problem with PPV and NPV is that demographics and disease prevalence does change based on setting which impacts the PPV and NPV

Answer 99

These are kinda like sensitivity and specificity but they relate to the participant/patient -Negative predictive value (the chance that if somebody tests negative that they'll actually not have the disease) Sensitivity and specificity relate to the test and the test won’t change. The problem with PPV and NPV is that demographics and disease prevalence does change based on setting which impacts the PPV and NPV

Answer 100

 Positive Likelihood ratio (LR+) probability of positive test result in a patient with the disease / probability of positive test result in a patient without the disease or sensitivity / (1 – specificity) - When the positive likelihood ratio is close to 10 it means that when somebody tests positive we can be pretty certain they have the disease

Answer 101

 Negative Likelihood ratio (LR-) probability of negative test result in a patient with the disease / probability of negative test result in a patient without the disease or (1 – sensitivity) / specificity - When the negative likelihood ratio is close to 0 it means that when somebody tests negative we can be pretty certain they don't have the disease

Answer 102

examples in OneNote Using multiple tests (when they test positive for them all they build of likelihood of having the disease)

Answer 103

Primary prevention aims to prevent disease / injury before it occurs – Vaccination against infectious disease – Education and awareness of healthy and safe habits – Legislation

Answer 104

Secondary prevention aims to reduce the impact of disease / injury that has occurred – Screening to detect disease at an early stage – Interventions to prevent further disease / injury

Answer 105

Tertiary prevention aims to alleviate the impact of ongoing illness / injury – Rehabilitation  Mental  Physical  Social

Answer 106

 Target a select group – usually vulnerable  Intravenous drug users – Needle-exchange program – Vaccination (against hepatitis B)  Screening pregnant women aged over 40 years – MSAFP – Ultrasound – Amniocentesis  Limitation: common disease / widespread cause

Answer 107

 Common disease / widespread cause  Mass / population strategy  Legislated use of seatbelts – Targeting only ‘high’ risk e.g. males 18-25 years would not work

Answer 108

going in for 1 test but decide to do 4 because we can – ‘Case finding’ – e.g. PSA test at ‘general health check-up’

Answer 109

– Screening those in a specific criteria – e.g. mammography in women 50-69

Answer 110

– Screening across an ‘entire population’ – e.g. Neonatal screening (i.e Guthrie test)

Answer 111

 Potential benefits – Early detection of the disease – Early treatment of the disease (↑ treatment options) – Psychological well being

Answer 112

 Potential limitations – Over-diagnosis (over-treatment) – Interval disease (cancers – too late???) – False negative/positive (implications) – Side effects (screening and treatment)

Answer 113

 Lead time  Length time  Volunteer  Over-diagnosis

Answer 114

 It does not take into account the natural history of the disease  The period between screens when disease is detected by screening and when it would have become symptomatic and been diagnosed in the usual way

Answer 115

 Occurs due to heterogeneity in the disease (i.e. fast and slow growing tumours)  Screening tests more likely to find slow growing disease, hence better apparent prognosis  Over-representation of slowly progressing disease among cases detected by screening

Answer 116

Identification of slow growing cancers that may never have become apparent (‘false positive’)

Answer 117

Can also be thought of as ‘false negatives’  Interval disease occur between screening as they are found in the time interval between screens  How do you overcome interval disease??? – Increase screening frequency – But, do you increase chances of over-diagnosis?

Answer 118

A number of measures can be used to quantify how ‘effective’ a screening program is; – RR, RRR – Gain in life expectancy – Cost per case detected – Cost per life saved – Gain in quality adjusted life years (QALYs) – Number needed to screen (NNS) a good screening test is... -Safe -Simple -Reliable -Accurate/valid

Answer 119

1. The condition should be an important health problem 2. There should be a recognisable latent or early symptomatic stage 3. The natural history of the condition, including development from latent to declared disease, should be adequately understood 4. There should be an accepted treatment for patients with recognised disease 5. There should be a suitable test or examination that has a high level of accuracy 6. The test should be acceptable to the population 7. There should be an agreed policy on whom to treat as patients 8. Facilities for diagnosis and treatment should be available 9. The cost of screening (including diagnosis and treatment of patients diagnosed) should be economically balanced in relation to possible expenditure on medical care as a whole, and 10. Screening should be a continuing process and not a ‘once and for all’ project.

Answer 120

Does the intervention ‘work’ under ideal, ‘laboratory’ conditions?

Answer 121

It's public health impact  If we administer the intervention in ‘real life’ situations, Is it effective? Barriers include...  Taste  Mode of delivery  Cost  Accessibility  Perception

Answer 122

If the intervention is effective, what is the cost to benefit ratio? Costs include: money, discomfort, pain, disability, quality of life and social implications

Answer 123

Peer-reviewed journal Implementation Science is an open access peer-reviewed academic journal in healthcare that was established in 2006