CHY Flashcards
High yield
1
Q
95% confidence intervals
A
Range of values that is 95% likely to contain the true value
2
Q
Standard deviation
A
Value that shows how much variation there is from the mean
- 68% of values in dataset lies between -/+ 1 SD
- 95% between 2SD
- 99.7% between 3SD
3
Q
P value
A
Likelihood that the observed result is due to change
P > 0.05 is not statistically significant
4
Q
Odds ratio
A
Ratio of odds that something will happen, to the odds that it won’t happen.
Used in case control studies.
5
Q
Blinding
A
Patients/clinicians/researchers are prevented from knowing certain information that may lead to conscious or subconscious bias on their part
6
Q
Concealment of allocation
A
The procedure for protecting the randomisation process
Person randomising the patient does not know what the treatment allocation will be
Prevents selection bias affecting which patients are give which treatment (the bias the randomisation is designed to avoid)
7
Q
Intention to treat analysis
A
Analysis based on the initial treatment intended from allocation,
Intention to treat (ITT) analysis means all patients who were enrolled and randomly allocated to treatment are included in the analysis and are analysed in the groups to which they were randomized, not the treatment eventually administered.
I.e. “once randomized, always analyzed”
8
Q
Treatment fidelity
A
How accurately the intervention is reproduced from a manual, protocol or model
9
Q
Number needed to treat
NNtT
A
Number of patients that are needed to be treated with the experimental therapy to prevent one negative outcome
E.g. After myocardial infarction, 18 people need to take an angiotensin-converting enzyme (ACE) inhibitor to prevent one heart attack (NNT = 18)
It is defined as the inverse of the absolute risk reduction.
1/Absolute risk reduction
10
Q
Purpose of randomisation?
A
To try and ensure that any characteristics of the sample population that may affect the results (confounders) are distributed equally between the 2 study groups, and avoid selection bia
11
Q
Which type of studies are often affected by recall bias?
A
Case control
12
Q
What is recall bias?
A
The participant either cannot remember back to when they were exposed or their outcome changes their perception of the exposure
13
Q
Internal validity
A
Accuracy
How well the study was conducted.
Taking confounders into account and removing bias.
14
Q
External validity
A
Generalisability
How well it can be applied to different scenarios, patients, environments.
15
Q
List 4 methods that can be used to limit confounding variables.
(may be used in cohort studies!)
A
Restriction:
- Limit participants of study that have possible confounders
- Means less data available
- This method can be difficult with multiple confounders
Matching:
- Make comparison groups (with and without the confounder)
- Used for things like age and sex in case-control studies
Stratification:
- Analyse exposure with sub-groups of the confounder
- Adjust for confounding (if there are few variables)
- Recombine data
- Means sometimes these subgroups have very few participants in them
Multiple variable regression:
- Coefficients are established for the confounder groups
- Allows for better adjustment
16
Q
Case-control studies are prone to recall bias in general
A
Cohort studies are prone to selection bias in general
17
Q
Population attributable risk (PAR)
A
The portion of the incidence of a disease in the population (exposed and non-exposed) that is due to exposure
It is the incidence of a disease in the population that would be eliminated if the cause of exposure was eliminated
This can be expressed as a value or percentage
18
Q
Allocative efficiency
A
Investing in worthwhile interventions
19
Q
Technical efficiency
A
Investing in the interventions that make the best use of scarce resources
20
Q
SPIKES
A
Setting Perception Invitation Knowledge Empathy Summary
21
Q
Kubler-Ross model of grief
A
Denial Anger Bargaining Depression Acceptance
22
Q
Bowlby model of grief
A
Numbing
Yearning/searching
Disorganisation
Reorganisation
23
Q
Most common cancers in men
A
Lung Prostate Colorectal Stomach Liver Bladder Oesophagus Non-Hodgkin Kidney Leukaemia
24
Q
Most common cancers in women
A
Breast Lung Colorectal Cervical Thyroid Uterine Stomach Ovarian Liver Non-Hodgkin
25
Most common cancers in children
Leukemia
Brain
Lymphoma
Soft tissue sarcoma
26
Type I error
False positive
27
Type II error
False negative
28
Standardised mortality ratio (SMR)
Ratio between the observed number of deaths (O) in a study population to the number of expected deaths (E)
SMR = O/E
The rate of deaths in one population compared with national averages corrected for age, sex, social class, etc.?
29
Direct standardisation in the context of death rates?
Age is major conounding factor in death rates
Direct standardisation requires that we know the age-specific rates of mortality in all populations under study
- Weighted average if stratum-specific rates
30
Indirect standardisation in the context of death rates?
Age is major conounding factor in death rates
Indirect standarisation only requires that we know the total number of deaths and the age structure of the study population
- How many deaths would we expoect in age group?
- How many were there?
- Work outobserved/expected mortality rate
31
When is indirect standardisation preferable?
Small numbers in particular age groups
32
Likelihood ratio
The likelihood that a given test result would be expected in a patient with the target disorder compared to the likelihood that the same result would be expected in a patient without the target disorder
33
Length bias
Overestimation of survival duration among screening-detected cases by the relative excess of slowly progressing cases
34
Consequences of length bias?
Diseases with a longer sojourn time are 'easier to catch' in the screening net.
On average, individuals with disease detected through screening 'automatically; have a better prognosis than people who present with symptoms/signs.
If we simply compare individuals who choose to by screened with those who don't we will get a distorted picture.
35
Lead time bias
Overestimation of survival duration among screen-detected cases (relative to those detected by signs + symptoms) when survival is measured from diagnosis
36
Consequences of lead time bias?
Survival is inevitably longer following diagnosis through screening because of the 'extra' lead time
Because of this the appropriate measure of effectiveness is deaths prevented, not survival
37
Over-diagnosis bias?
Overestimation of survival duration among screen-detected cases caused by inclusion of pseudodisease - subclinical disease that would not become overt before the patient dies of other causes.
Occurs when screen-detected cancers are either non-growing or so slow-growing that they never would cause medical problems
38
Statutes (laws) that oblige doctors to disclose information
Public Health Act 1984
Road Traffic Act 1988
Prevention of terrorism act 1989
39
3 different types of errors
Knowledge based - Such as forming wrong intentions or plans as a result of inadequate knowledge/experience
Rule based - Encounter relatively familiar problem but apply wrong rule, either misapplication of a good rule or application of a bad rule.
Skills based - Attention slips and memory lapses, involve the unintended deviation of actions from what may have been a good plan. We are all prone to these types of errors, mainly due to interruption and distractions.
40
Latent error
Develop over time until they combine with other factors or active failures to cause an adverse event
Long lived and often can be identified and removed before they cause an adverse event
41
What are violations?
Deliberate deviation from some regulated code of practice or procedure
They occur because people intentionally break the rules
42
List 4 types of violations
Routine - Regularly performed shortcuts due to system, process or task being poorly designed or actions. May become tacitly accepted practice over time
Reasoned - Occasional reasoned deviation from a protocol or procedure which we believe we have good reason for making (e.g. time constraints), may be in patient's best interests
Reckless - Deliberate deviations from a protocol/code of conduct and include acts where opportunity for harm is foreseeable and ignored, although harm may never be intended
Malicious - Deliberate deviations from a protocol/code of conduct, where the intention is to cause harm
43
What is a near miss?
A situation in which events arise during clinical care but fail to develop further
44
Swiss cheese model of accident causation
Although many layers of defence lie between hazards and accidents, there are flaws in each layer that, if aligned, can allow the accident to occur.
45
Main causes of individual vs system errors?
Individual error - Errors of individuals, blames individual for forgetfulness, inattention or moral weakness
System error - Conditions under which an individual works, tries to build defences to eliminate errors or mitigate their effect
46
List 3 antenatal screening tests that identify major abnormalities
Alpha fetoprotein - Raised in neural tube defects and some GI abnormalities
Downs test - Alpha fetoprotein and HCG
Ultrasound - Growth check, cardiac abnormalities, diaphragmatic hernia
47
Neonatal tests
Blood spot test:
- PKU
- Cystic fibrosis
- Sickle cell disease
- Congenital hypothyroidism
Physical examination
48
4 types of neglect
Physical neglect
Educational neglect
Emotional neglect
Medical neglect
49
Which act says 16 year old has full capacity?
The family law reform act 1969
50
What is Gillick competency?
Child (under 16) can consent to medical treatment if deemed competent by medical professional, without need for parental permission or knowledge
51
What are the Fraser guidelines?
Doctor can give contraceptive advice and treatment to a person under 16 if she is mature and intelligent, likely to continue to have sex, and if the treatment if in her best interests
52
Effects of targeting population for prevention?
Large potential benefit to community
Low potential benefit to individual
May be low perceived benefit to individual
53
Effects of targeting high risk groups for prevention?
Larger potential benefit to individual
Smaller effect on population rate of stroke
Many of the conditions you treat are asymptomatic
May of the treatments have side effects
54
What did the PROGRESS trial show?
Reducing blood pressure after stroke reduces risk of stroke recurrence
55
Occupational causes of asthma
Bakers
Welders
Paint sprayers
Laboratory workers
56
Occupational causes of COPD
```
Coal mining
Agriculture
Construction
Dock workers
Brick making
```
57
Simple coal worker's pneumoconiosis
After around 10 years coal mining, small nodules are present
Shouldn't cause major impairment in lung function
Some coal workers have symptoms of chronic bronchitis (cough)
58
Pneumoconiosis
Occupational restrictive lung disease caused by inhalation of dust (coal dust, silica, asbestos)
59
Silicosis
Occupational lung disease caused by inhalation of crystalline silica dust, and is marked by inflammation and scarring in the form of nodular lesions in the upper lobes of the lung.
It is a type of pneumoconiosis.
60
2 types of asbestos fibres
Serpentine - Curly, white asbestos (relatively harmless), cleared with mucociliary escalator
Amphiboles - Short, sharp, blue/brown asbestos (have malignant potential)
61
What are preference sensitive and probability sensitive decisions?
Preference sensitive - The person might feel strongly about the side effects of the treatment
Probability sensitive - Sensitive to changes in the chance of different outcomes
62
Clinical decision support systems (CDSS) are designed to aid clinician decision making
Computerised
Paper based
Reminder systems
Developed to aid with particular decisions
63
Examples of CDSS
Reminder systems - Screening, vaccination, testing, medication use
Decision systems (diagnosis and treatment) - Model individual patient data against
epidemiological data
Prescribing - Advice on drug and dosage, highlights potential drug interactions
Condition management - Assists monitoring patients
64
Equity in financing
Geographic allocation of funding by weighted capitation
Resourcing determined by population weighted by need
65
What is class equality/inequality in health care?
Evidence of social class equality in the use of primary care and social class inequality in the use of secondary care
66
What is efficiency?
Target resources to those activities that give the greatest health gain for the money spent as this will maximise population health gain
Informing these choices required estimation of value of what is given up when a patient is treated (opportunity cost) and the value of what is gained in terms of improvements in the health of patients
67
Fair-innings arguement (1997)
Older people have had a long life already, therefor fairer to divert resources to younger people
Elderly also have a disproportionate share of the available resources allocated to them
68
Contradictions for the Fair-innings argument?
Treating on the basis of need might mean older people don’t receive lower priority
Years of life saved shouldn’t matter, the quality of life is more important e.g QALYs
Fairness is not the only thing that matters, other things do too e.g equal treatment
69
Advantages of cohort studies
Good for rare risk factors
Can assess multiple risk factors at once
Yields true incidence rates and relative risk
Best for common diseases
70
Disadvantages of cohort studies
Expensive compared to case-control studies
Time consuming - Risk factor may take decades to cause disease (long latency period)
Not good for rare diseases
Confounding factors
Cannot determine odds ratio
71
Advantages of case-control studies
```
Good for rare diseases
Good for tracing source of an outbreak
Relatively inexpensive and quick compared to cohort studies
Yields odds ratios
Good for long latency periods
```
72
Disadvantages of case-control studies
```
Relies on recall or existing records
Inferior to cohort
Cannot determine relative risk directly
Not good for rare exposures
Selection of controls difficult
```
73
What should systematic reviews have in their inclusion criteria?
Papers that have not been published (publication bias)
Papers that are not in English
74
R0
Basic reproduction rate
The average number of individuals directly infected by an infectious case during the infectious period, in a totally susceptible population (number of secondary cases following introduction of infection)
75
What factors affect R0?
The rate of contacts in the host population
The probability of infection being transmitted during contact
The duration of infectiousness
76
Effective reproduction rate (R)
Estimates the average number of secondary cases per infectious case in a population made up of both susceptible and non-susceptible hosts
77
Equation for effective reproduction rate
R = R0x (x is the fraction of the host population which is susceptible e.g. half population is 0.5)
R>1 - number of cases increases
R<1 - Number of cases decreases, needs to be maintained for elimination
R=1 - Epidemic threshold
78
Herd immunity equation
H = (R0-1)/ R0
79
What are some of the free and private travel vaccines available?
Free - Diptheria, polio, tetanus, typhoid, hepatitis A, cholera
Private - Hepatitis B, japanese encephalitis, meningitis, rabies, TB, yellow fever
80
Define disability
Restriction or lack (resulting from an impairment) of ability to perform an activity in the manner or within the range considered normal for human being
81
Dual process model
Better model that says we cope w/ grief/bad things by carrying out loss orientated + restoration oriented behaviours
Loss oriented: grief work, breaking bonds, denial
Restoration orientated: doing new things new roles + relationships
These 2 processes occur simultaneously through everyday life experiences
82
What different forms can the stress response take?
1. Direct action
2. Seeking information
3. Doing nothing
4. Developing way of coping
83
Likelihood ratio
The likelihood that a given test result would be expected in a patient with the target disorder compared to the likelihood that that same result would be expected in a patient without the target disorder.
84
What is the likelihood ratio for positive test results?
LR+
Chance of testing positive if you have disease/ chance of testing positive if you don’t have disease
85
What is the LR for negative test results?
LR-
Chance of testing negative if you have disease/ chance of testing negative if you don’t have disease
86
What is the significance of LR+ and LR-?
The larger the LR+ the greater chance you have disease if your test is positive
The smaller the LR- the less chance you have disease if your test is negative
87
What is the pre-test probability?
Chances of disease before test = the disease prevalence in your population
88
Length bias
Screening is best at picking up long-lasting and slowly-progressive conditions rather than rapidly-progressing poor prognostic ones
Overestimation of survival duration among screening-detected cases by the relative excess of slowly progressing cases
Consequences:
Diseases with a longer sojourn time are 'easier to catch' in the screening net.
On average, individuals with disease detected through screening 'automatically; have a better prognosis than people who present with symptoms/signs.
If we simply compare individuals who choose to by screened with those who don't
we will get a distorted picture.
89
Lead time bias
Overestimation of survival duration among screen-detected cases (relative to those detected by signs and symptoms) when survival is measured from diagnosis.
Consequences:
Survival is inevitably longer following diagnosis through screening because of the 'extra' lead time
Because of this the appropriate measure of effectiveness is deaths prevented, not survival
90
Overdiagnosis bias?
Overestimation of survival duration among screen-detected cases caused by inclusion of pseudodisease - subclinical disease that would not become overt before the patient dies of other causes.
Occurs when screen-detected cancers are either non-growing or so slow-growing that they never would cause medical problems
91
How can waiting lists be measured?
```
Average or median waiting time
Average or median wait of those on list
Number on the list (size)
Proportion waiting more than x days
Proportion of those on list waiting more than x days
Time to clear list
```
92
Identify two potential disadvantages to individuals or to society that may result from the introduction of a screening programme
False positive results leading to anxiety and unnecessary interventions.
Opportunity cost of the programme for NHS (and individual).
Increased diagnostic activity.
93
How might measuring BMI in school children be used?
Overweight (or underweight) children are identified and parents take appropriate action.
Schools take action to tackle obesity
The importance of obesity is highlighted in the community
NHS and/or Local Authority have information that can be used to plan services to tackle obesity
Local obesity figures and performance can be monitored
94
What conditions are screened for in newborns?
Congenital hypothyroidism;
Phenylketonuria;
Medium Chain Acetyl Decarboxylase (MCAD);
Hearing problems;
Cystic fibrosis;
Sickle cell disease;
95
Give five examples of the problems faced by families affected by AIDS in low income countries.
```
Children as care givers
Orphans
Psychosocial distress
Poverty
Lack of schooling
Reduced access to health care services
Discrimination
Child labour and exploitation
Sexual exploitation
Street children
```
96
HIV/AIDS in Africa is changing the pattern of the population structure.
What is the effect of the changes on the age structure and why should it have this effect?
Reduction in the young and middle aged adults rather than children (2 marks)
Infects sexually active people with an approx. 10 year period to death so time to reproduce is not affected (2 marks)
97
Where in Africa does HIV/AIDS have most impact?
Sub-Saharan
98
Give two reasons why waiting times may benefit the NHS.
Act as rationing
| Makes best use of resources
99
Give four initiatives taken in the NHS to reduce waiting lists which might come into your discussion.
More doctors/healthcare staff
Day cases
Targets and penalties for missing them
Activity based remuneration for doctors
Activity based remuneration for hospitals
Star ratings
Other service providers (e.g. contracting with private providers; switch to primary care provision)
100
What was the outcome of the midwives' act (1902)?
Established normality in childbearing as the midwife's role - Refer to doctors as soon as abnormality occurs
This ensures equal access to midwives and doctors for childbearing women of all socioeconomic standing
101
Benefits of institutionalised childbirth?
Standardisation of care
Access to good facilities to support childbirth
Availability of populations of childbearing women and infants for the purposes of midwifery and obstetric training
Faster access to emergency care
Access of effective obstetric analgesia
102
Risks of institutionalised childbirth?
```
Medicalisation
Deposonalisation of birth
Lack of privacy
Inflexibility in labour and birth practices
Limitation of resources
```
103
What are the legal aspects of ART?
| What act is it laid out in?
Human fertilisation and embryology authority act 2008
- Welfare of child needs to be taken into account
- Need for ‘supportive parenting’ hence valuing role of all (hetero and homo) parents
- Sex selection/social selection of embryos is prohibited
104
Standardised mortality ratio?
Ratio between the observed (O) number of deaths in a study population to the number of expected (E) deaths
SMR = O/E
105
How can NHS reduce waiting times?
Manage demand - Ensuring each referral represents the most appropriate decision for the care of the individual patient
Manage the queue - Ensuring waiting lists are well managed and patients are called for treatment in appropriate order
Manage capacity - Providing efficient and effective services that meet the level of demand from appropriate referrals
Provide leadership - Ensuring that all parts of the local NHS work together to achieve waiting time improvements in the best interests of patients.
106
Possible criteria for priority on waiting lists?
```
Clinical urgency
Clinical severity
Potential health gain
Productivity and economic loss
Equity waiting e.g. poverty
Length of time waiting
```
107
What does the NICE guidelines state about acupuncture in lower back pain, osteoarthritis, and headaches?
Low back pain - Consider manual therapy, do not offer acupuncture
Osteoarthritis - Manipulation and stretching should be considered as an adjunct to
core treatments, do not offer acupuncture
Headache/migraine - Consider a course of up to 10 sessions of acupuncture over 5-8
weeks
108
According to the mental capacity act 2005, a person lacks capacity if they are unable to ...
Understand information that may be relevant to the decision, including the consequence
Retain information, even for a short time
Use or weigh information to make decisions
Communicate decision
109
Type 1 error (false positive)
The null hypothesis is REJECTED when it is true
i.e. Showing a difference between two groups when it doesn't exist, a false positive. This is determined against a preset significance level (termed alpha).
As the significance level is determined in advance the chance of making a type I error is not affected by sample size. It is however increased if the number of end-points are increased. For example if a study has 20 end-points it is likely one of these will be reached, just by chance.
110
Type 2 error (false negative)
The null hypothesis is ACCEPTED when it is false.
i.e. Failing to spot a difference when one really exists, a false negative.
The probability of making a type II error is termed beta. It is determined by both sample size and alpha
111
Risk
Probability of an event in a given time period.
112
Risk ratio
Ratio of the probability of an event (risk) between two groups.
1 indicates risk is the same.
Used in cohort studies.
Risk ratio = Risk exposed/ Risk unexposed
Risk ratio = Risk of CVD smoker/Risk of CVD non smoker
113
Population attributable risk?
The risk of disease will increase as the exposure prevalence or relative risk increases
114
Main risk factors associated with lung cancer?
```
Smoking (90%)
Radon
Asbestos
Environmental tobacco exposure
Genetics
Other lung diseases
Prior radiation in chest area
```
115
Different types of lung cancer?
Small cell (13%)
Non-small cell (87%)
- Adenocarcinoma (>40%),
- Squamous cell carcinoma (20%),
- Large cell carcinoma (2%)
Mesotheloma
116
Prevention paradox
A preventive measure that brings large benefits to the community offers little to each participating individual
117
How are confounding factors limited?
Restriction
- Limit the participants of your study who have possible confounders
- Means that you have less data and difficult with multiple confounders
Matching
- Create a comparison group that is matched on the possible confounder, make case and control group as similar as possible on the confounder and then ask about exposure status
- Used for strong confounders like age and sex
Stratification
- Analyse exposure:outcome association in different subgroups of the confounder, recombine data and use a weighted average of the strata
- Limitations : to take into account all confounders would require lots of strata and you may run out of data to fill all possible options in your strata
Multiple variable regression
- You can adjust for the effects of multiple confounders, try and produce a linear model between the outcome and the different exposures
- Allows for adjustment of estimates for confounding
118
Types of research ethics committees?
NHS Research ethics committees
Higher Education Institution (HEI) research ethics committees
Gene therapy advisory committee
Social care research ethics committee
Ministry of defence research ethics committee
119
What was the Calman-Hine report?
Examined cancer services in the UK, and proposed a restructuring of cancer services to achieve a more equitable level of access to high levels of expertise throughout the country.
120
What are the stages in the transtheoretical model?
Precontemplation
- No recognition of need/or interest in change
Contemplation
- Thinking about changing
Preparation
- Planning the change
Action
- Adopt new habits
Maintenance
- Active practice of new healthier behaviour
121
How can utility be assessed?
Quality adjusted life year (QALY)
Standard gamble - Assesses utility for a health state by asking how high a risk of death would accept to improve it.
Time-trade off - Utility assessed by asking how much time (life expectancy) would give up to improve it
122
What is restriction?
To manage confounding, limiting participant of study who have possible confounders.
Means you have less data
123
What is matching?
Creating a comparison group that is matched on possible confounder.
Used for strong confounders like age and sex
124
What is stratification?
Analyse exposure:outcome association in different subgroups. e.g. CHD in pie eaters v non-pie eaters
125
5 ways to limit confounding variables?
```
Restriction
Matching
Stratification
Multiple variable regression
Standardisation
```
126
What is multiple variable regression?
Adjusting for the effects of multiple confounders, producing a linear model between outcome and different exposures
127
What is standardisation?
To limit confounding, usually for age.
Comparing 2 groups in age subgroups
128
What is standardised mortality ratio?
Comparing how many deaths you would expect to how many there actually were.
SMR > 100% = higher death than average.
This is an example of indirect standardisation
129
Equation for risk ratio
Risk in exposure / Risk in unexposed
RR = a/a+c divided by c/c+d
130
What is attributable risk (AR) aka risk difference?
Attributable risk (AR) or risk difference is the difference between the incidence rates in exposed and non-exposed groups.
In a cohort study, AR is calculated as the difference in cumulative incidences (risk difference) or incidence densities (rate difference).
This reflects the absolute risk of the exposure or the excess risk of the outcome (e.g. disease) in the exposed group compared with the non-exposed group.
AR is sometimes referred to as attributable risk in the exposed because it is used to quantify risk in the exposed group that is attributable to the exposure.
131
What is absolute risk reduction?
In general, absolute risk reduction is the difference between one treatment comparison group's event rate (EER) and another comparison group's event rate (CER).
The difference is usually calculated with respect to two treatments A and B, with A typically a drug and B a placebo.
ARR = c/(c+d) - a/(a+b)
132
Pre-test probability is equivalent to the prevalence of a condition
The proportion of people with the target disorder in the population at risk at a specific time (point prevalence) or time interval (period prevalence)
For example, the prevalence of rheumatoid arthritis in the UK is 1%
133
Which one of the following is the best definition of the p value?
A. The probability of obtaining a similar result, assuming that the null hypothesis is true
B. The probability that a replicating experiment would not yield the same conclusion
C. The probability of obtaining a result at least as extreme, assuming that the null hypothesis is true
D. The probability that the null hypothesis is true
E. The probability of obtaining a result at least as extreme, assuming that the null hypothesis is false
C. The probability of obtaining a result at least as extreme, assuming that the null hypothesis is true
The p value is the probability of obtaining a result by chance at least as extreme as the one that was actually observed, assuming that the null hypothesis is true. It is therefore equal to the chance of making a type I error (see below).
134
When does a type 2 error occur?
The null hypothesis is ACCEPTED when it is false.
A type II error occurs when the sample size in a study is too small to allow for a difference to be detected.
A type II error can be prevented by recruiting a larger sample size.
135
When does a type 1 error occur?
The null hypothesis is REJECTED when it is true
A type I error occurs when the investigators conclude that there is a difference in a study when in reality there is none. Investigators usually set an arbitrary value such as 0.01 or 0.05.
This means that the investigators accept that there is a 1% or 5% chance that any difference detected is due to chance.
136
Lead time bias
A lead-time bias is introduced in a study when a disease is detected an earlier stage and this gives the impression that the survival time from the detection point has increased. In reality, the increase in survival time is because the disease is detected earlier and the time of death remains unchanged.
137
In the analysis stage of an experiment how can confounding variables be controlled for?
In the analysis stage of an experiment, confounding can be controlled for by stratification.
NOT RANDOMISATION
138
What systems are in place in the NHS to try and prevent errors occurring?
National Patient Safety Agency (NPSA) - 2001. coordination of reporting and learning from mistakes that affect patient safety.
National Reporting And Learning System (NRLS) - 2004. National system for anonymous reporting of patient safety incidents, including near misses. All trusts now have a local system for reporting, linked to the national system. Also has a E- form for reporting incidents anonymously directly to the NPSA.
Medicines and Healthcare - Products Regulatory Agency (MHRA) - Ensures medicines, healthcare products and medical equipment meet appropriate standards of safety, quality, performance and effectiveness and that they are used safely. Monitoring of medicines and acting on safety concerns. Responsible for adverse incident reporting system for medical devices.
139
How to know if a hospital is safe?
Hospital mortality data
Data on other measures of safety:
- Reports of never events and serious incidents
- NHS safety thermometer
- Patient safety dashboards
Monitoring and inspections by regulators
- Care quality commission (CQC)
- NHS improvement
140
Effects of computer support on prescribing?
```
Reduced time to achieve therapeutic stabilisation
Reduced risk of toxic drug level
Reduced length of hospital stay
Increased size of initial dose
Increased serum drug concentration
No change in adverse effects of drug
```
141
Staph aureus infection
Transmission - Contaminated food by skin/nasal flora
Produces enterotoxins
Incubation - 2-4 hours
Symptoms - Rapid onset, projectile vomiting and diarrhoea
142
Salmonella infection
Transmission - Ingestion of contaminated food, faecal contaminations, person- person, infected animals
Can cause enteric fever or enterocolitis
Incubation period - is 12-72 hours
Symptoms - Vomiting, diarrhoea, fever, headache, chills
143
Cryptosporidium infection
Transmission - Animal-human, person-person, contaminated water or land, associated with foreign travel
Incubation - 2-5 days
Symptoms - Watery or mucoid diarrhoea, severe illness in immunocompromised
144
Escherichia coli infection
Transmission - Contaminated food, person-person
Incubation - 1-6 days
Symptoms - Haemorrhagic colitis, 5% get haemolytic uraemic syndrome
145
What is an outbreak?
Incident in which two or more people, thought to have a common exposure, experience a similar illness or proven infection
146
What might outliers in an outbreak represent?
```
Baseline level of illness
Outbreak source
A case exposed earlier than the others
An unrelated case
A case exposed later than the others
A case with a longer incubation period
```
147
How can analytical epidemiological studies be useful to identify probably food source of outbreak?
Compare food history of ill and well persons
Point source outbreak - Cohort study
Common source of outbreak - Case-control study
148
Main dietary associated with cancer
Oesophageal - Alcohol, obesity
Stomach - Possible salted preserved foods
Pancreas - Overweight, obesity
Hepatic - Aflatoxin contamination
Colorectal - Preserved and red meat, alcohol, body fat
Breast - Alcohol, overweight
Urologic - high calcium
149
What are the different types of results?
Therapy - Look at relative risk reduction, absolute risk reduction, odds ratio, number needed to treat, confidence intervals
Diagnosis - Look at sensitivity, specificity, positive predictive value, negative predictive value, likelihood ratios
Prognosis - Look at how likely the outcomes are over time and how precise the prognostic estimates are (relative risk or odds ratios)
Harm/aetiology - Look at relative risk, odds ratio, number needed to harm
150
The National Service Framework on Children and Young People has made suggestions for the improvement of children’s health including accidents.
Give three key messages that the NSF suggests to improve child health.
Tackle health inequalities including poverty
Improving access to services/information
Focus on early intervention/Sure Start
Promote physical mental and emotional wellbeing
Promote healthy lifestyles
Setting standards
151
Compare the importance of childhood accidents with other causes of child ill health and death.
Accidents are a major cause of death in children
They cause twice as many deaths in children as cancer and leukaemia
Cause 6 times as many childhood deaths as murder (homicide)
The incident rate of childhood accidents has not fallen in recent decades as quickly as, for example, diphtheria and pneumonia
152
What are the arguments for and against age-based rationing being applied to micro-level resource allocation decisions?
For - Age should be relevant because older people are less likely to respond to treatment and have a poorer prognosis in general due to increased complication risk
Against - Age alone is not a good predictor of prognosis/complications hence need case-by-case decisions, decisions based on age may be hidden form of discrimination
153
Which act protects discrimination based on age, race, sex, gender, disability, religion etc.?
Equality Act 2010
154
What do clinical and economic evaluations investigate?
Clinical - efficacy and effectiveness
Economic - efficiency
155
Advantages of systematic reviews over other research designs for answering clinical questions.
A systematic review is a rigorous summary of all the research evidence that relates to a specific question.(1 mark)
By bringing together all the relevant evidence disadvantages of single studies can be guarded against.(1 mark)
156
Identify two potential weaknesses of the search strategy used by the authors of the review (for systematic review).
Explain why these features might bias the findings of the review
* They have only looked at publications in English
* They have only searched one database – e.g. Medline
* Relevant publications might be published in non English language journals.
* Not all relevant articles may be indexed in Medline.
* A thorough search will reduce the risk that studies that might influence the results of the review will be missed
* Trial reports that are translated for publication in English language journals tend to have bigger effect sizes (i.e. so English-language only searches tend to overstate the effect of an intervention).
157
What are the different consumer protection agencies?
- Care quality commission (CQC)
- National Institute for Health and Clinical Excellence (NICE)
- DoH and NHS England: oversight of cost and quality control
- GMC: regulates medical schools and practitioners
- Royal colleges: examine and certify practitioners
158
National service framework?
Set national standards and define service models for a service or care group
Put in place programs to support implementation
Establish performance measures against which progress within agreed timescales would be measured
159
How can consumer protection be improved?
Appraisal by peers
Revalidation by the GMC
Medical audit as a compulsory part of routine practice and annual job planning
GP and consultant contracts - Increasing transparency in comparative performance
in relation to activity, costs, and patient reported outcomes
Transparency and accountability
160
General fertility rate
General fertility rate = number live births per 1000 women between ages 15-44
161
Positive predictive value (FP)
Probability that those with positive result truly have the disease.
It is a measure of false positives.
Depends on population.
162
Negative predictive value (FN)
Probability that those with a negative result truly are disease free.
It is a measure of false negatives.
Depends on population.
163
Specificity (TN)
The probability of a negative, if the patient really is not ill
True negatives (TN)
TN / ( TN + FP)
true negatives / all truly disease-free
164
Sensitivity (TP)
The probability of a positive, if the patient really is ill.
True positives (TP)
TP / (TP + FN)
true positives / all truly disease
165
Summary of sensitivity, specificity, PPV, NPV.
```
Sensitivity = TP
Specificity = TN
PPV = FP
NPV = FN
```
166
Relative risk
One populations risk compared to another population.
The risk of developing disease for one population with an exposure compared with a population without the exposure?
E.g. smokers are 20 times more likely to contract lung cancer than non-smokers
E.g. women are 100 times more likely to get breast cancer than men.
167
Absolute risk
One population's risk in its own right.
AKA lifetime prevalence in that population.
E.g. women have a 11% risk of getting breast cancer.
168
Principles of ethical research
Do good (or at least do no harm).
- Equipose
- Stop trial early if necessary
- Individual vs population
Informed consent.
Protect anonymity and confidentiality.
Avoid deceptive practices.
Right to withdraw.
169
What does high incidence to prevalence ratio mean?
High incidence to prevalence ratio means high death or cure rate
170
Cancer
```
Risk factors:
Smoking
Obesity/diet
Alcohol
Lower socioeconomic class
```
Specific risk factors:
Previous radiotherapy
Occupational
171
Coronary artery disease
Highest cause of death in developed countries.
```
Non-modifiable risk factors:
Male
Asian ethnicity
Age
Family history
```
Modifiable risk factors:
High blood pressure
Cholesterol
Diabetic control
172
Lung cancer risk factors
SMOKING (RR 3, 90% of lung cancers)
Occupational (Asbestos, Dust diseases, Chromium, Arsenic, Radiation)
Age
Male > female (confounder = smoking)
173
Tuberculosis
1/3 of the world’s population
20% have HIV
Urban areas
At-risk countries (south Asia, sub-Saharan Africa)
174
Ethical principles
Autonomy: Patient’s right to self-rule of their body
Non-maleficence: Not doing harm
Beneficence: Doing good
Justice: Fair and equal access to healthcare
175
Ethical theories
Deontology – people should adhere to their duties and obligations.
Consistent and does not vary.
Follows rules rigidly.
Utilitarianism: Greatest good for the greatest number.
Consequentialism: Consequences of actions are the determinant of the validity of an action.
Intentionalist: Intentions of actions are determinant of their validity.
176
Duty of care
A moral or legal obligation to ensure the safety or well-being of others.
Maintain and improve safety and quality of care.
Raise and act on concerns about patient safety.
Use resources efficiently, to benefit the patient and the public.
177
Negligence
A breach of the legal duty of care which results in damage.
Often passive, can be omission.
Failure to exercise the degree of care that a careful or prudent practitioner would have exercised under like circumstances.
Criminal negligence: Non-consensual treatment/Not acting in best interests
178
Errors
Medical error: a mistake that could have been prevented
Can involve medicines, surgery, diagnosis, treatment or lab reports.
Human factors.
Near miss: an event that might have resulted in a hazard, but it did not because of timely intervention
179
Capacity
Factors meaning a patient has capacity:
- Understand the information relevant to the decision
- Retain that information
- Use or weigh up that information and come to a conclusion
- Communicate that decision
May have capacity for some decisions but not others.
Adults have assumed capacity.
180
Gillick competence
“A parent yields to the child’s right to make his own decisions when he reaches a sufficient understanding and intelligence to be capable of making up his own mind on the matter requiring decision”
Typically assessed in over 13s
Often can accept not deny treatment
181
Fraser guidelines
For seeking contraception though <16.
The girl understands the advice
Cannot be persuaded to tell parents.
Likely to continue having sex anyway.
If contraception not received, health will suffer.
Best interests to give contraceptive advice.
Consultation should be kept confidential regardless.
Unless <13, refer to child protection.
182
Current screening programmes
> Antenatal and newborn, routine, all newborns
> Cervical cancer: 25-65, F
> Breast cancer: 50-70, F
> Bowel cancer: FOB at 60-74, endoscopy at 55
> Abdominal aortic aneurysm: 65, M
183
The Black Report
Reasons for inequalities in health.
Statistical artefact - Not genuine.
Natural or social selection - Those who are less well become poorer.
Materialist explanation - Exploitation of the lower socioeconomic classes.
Cultural/behavioural explanations - Eg. drinking.
184
Prevention
Primary prevention:
- Before the disease has occurred
- E.g. Warnings on cigarette packets, vaccines, legislation, lifestyle campaigns
Secondary prevention:
- Reduce the impact of the disease
- E.g. Bike helmets, early detection and diagnosis though screening
Tertiary prevention:
- Reduce the long term effects of the disease
- E.g. Support groups, disease management
185
Technical efficiency
The relationship between resources and health outcome.
Maximal possible improvement in outcome from a limited resource.
Lowest dose for greatest effect.
186
Allocative efficiency
The efficiency in which healthcare resources are used to produce outcomes AS WELL AS the distribution among the community.
Maximal good for most people.
187
Define guidelines
Guidelines: statements that include recommendations to optimise patient care that are informed on a systematic review of the evidence alongside expert opinion
188
Define protocols
Protocol: a set of rules to be followed by healthcare providers for a specific situatio
189
Types of bias
Selection bias: those included in a trial do not reflect a typical population, e.g. excluding those aged over 65 years from a study into ischaemic heart disease
Information bias: wrongly assigning trial participants incorrect diagnoses/ exposures
Reporting bias: cases with disease remember exposures better than those who are disease free
Interviewer bias: an interviewer may ask patients with disease more questions about exposures than those who are disease free
Loss to follow-up: if a significant proportion of trial participants drop out as time progresses, then the final data cannot be reliably interpreted
Lead time bias: earlier diagnosis of a condition results in an apparently longer survival time (see Fig. 8.1)
Publication bias: trials with positive results are much more likely to be published
190
Power
Probability of rejecting the null hypothesis when it is false
191
Hazard ratio
Similar to relative risk but used when risk is not constant
192
Cohort studies
Cohort studies are usually but not exclusively prospective, the opposite is true for case-control studies. The following notes relate cohort to case-control studies:
- Outcome is measured after exposure
- Yields true incidence rates and relative risks
- May uncover unanticipated associations with outcome
- Best for common outcomes
- Expensive
- Requires large numbers
- Takes a long time to complete
- Prone to attrition bias (compensate by using person-time methods)
- Prone to the bias of change in methods over time
- Related methods are risk (prospective), relative risk meta-analysis, risk difference meta-analysis and proportions
193
Case control studies
Case-Control studies are usually but not exclusively retrospective, the opposite is true for cohort studies. The following notes relate case-control to cohort studies:
- Outcome is measured before exposure
- Controls are selected on the basis of not having the outcome
- Good for rare outcomes
- Relatively inexpensive
- Smaller numbers required
- Quicker to complete
- Prone to selection bias
- Prone to recall/retrospective bias
- Related methods are risk (retrospective), chi-square 2 by 2 test, Fisher's exact test, exact confidence interval for odds ratio, odds ratio meta-analysis and conditional logistic regression.
194
Bias
Bias is a systematic error that leads to an incorrect estimate of effect or association. Many factors can bias the results of a study such that they cancel out, reduce or amplify a real effect you are trying to describe.
Epidemiology categorises types of bias, examples are:
- Selection bias - e.g. study of car ownership in central London is not representative of the UK.
- Observation bias (recall and information) - e.g. on questioning, healthy people are more likely to under report their alcohol intake than people with a disease.
- Observation bias (interviewer) - e.g. different interviewer styles might provoke different responses to the same question.
- Observation bias (misclassification) - tends to dilute an effect
- Losses to follow up - e.g. ill people may not feel able to continue with a study whereas health people tend to complete it.
Some strategies to combat bias:
- Multiple control groups
- Standardised observations (e.g. single, double, or triple blinding)
- Corroboration of multiple information sources
- Intention to treat analysis: done to avoid the effects of crossover and dropout, which may break the random assignment to the treatment groups in a study.
195
Confounding
In epidemiology a confounder is:
- Not part of the real association between exposure and disease
- Predicts disease
- Unequally distributed between exposure groups
A researcher can only control a study or analysis for confounders that are:
- Known
- Measurable
Strategies to reduce confounding:
- Randomization (aim is random distribution of confounders between study groups)
- Restriction (restrict entry to study of individuals with confounding factors - risks bias in itself)
- Matching (of individuals or groups, aim for equal distribution of confounders)
- Stratification (confounders are distributed evenly within each stratum)
- Adjustment (usually distorted by choice of standard)
multivariate analysis (only works if you can identify and measure the confounders)
196
Calculate odds of being a case among the exposed
a - exposed case
b - exposed control
c - unexposed case
d - unexposed control
Calculate odds of being a case among the unexposed.
Calculate the odds ratio of exposed vs non-exposed cases.
Odds of being case among all exposed = a/b
Odds of being case among all unexposed = c/d
Odds ratio: (a/b)/(c/d) = ad/cb (mathematical simplification)
197
What would an odds ratio (OR) of 0.6, 1.0 and 3.2 indicate?
OR of 0.6 indicates that exposure is negatively associated with disease (is a protective factor).
OR of 1.0 means that exposure is not related to disease.
OR of 3.2 would indicate that exposure is positively associated with disease.
198
13. When is it NOT appropriate to use the odds ratio to approximate the relative risk?
a. When the controls’ exposure history is representative of all subject without disease
b. When the case selection is representative of all subjects with disease (e.g. unbiased incident cases)
c. When the study disease occurs frequently
d. When the study outcome is rare
c. When the study disease occurs frequently
199
Risk difference
Synonyms: absolute risk reduction, attributable risk
Risk difference: the difference in the probability of an event occurring over a given time between two groups
200
What is number needed to?
A communication tool for the absolute difference in risk observed in different situations.
201
NNt calculation
NNt = 1 / |risk difference|
risk difference = ARR etc so 1/ARR is also correct
202
What is the inverse care law?
Inverse care law: the availability of good medical care tends to vary inversely with the need of the population served
203
CAT RCT: How precise are results?
Confidence intervals
P-values
204
CAT Cohort: Was the cohort recruited in appropriate way?
Selection bias
Is sample representative of general population?
205
CAT Cohort: Was exposure accurately measured to minimise bias?
Were subjects classified into exposure groups using same/appropriate procedure?
Do measurements reflect what you want them to?
- Inappropriate measurement could cause type 1 or 2 statistical errors
206
CAT Cohort: Was outcome accurately measured to minimise bias?
Reliable system established to ensure detection of all cases? (were measurement methods similar in different groups)
Blinding of subjects/observers
- Was outcome established blind to exposure
- Assess confounding variables and other differences in cohorts
207
CAT Cohort: Have authors identified important confounding factors?
Have they taken confounders in account in design or analysis?
Degree to which confounding variable is present.
- Define confounder: distortion of association between outcome and exposure by a third factor which has an association with both exposure and outcome.
- Confounder can lead to biased association between exposure and outcome.
208
CAT Cohort: Was follow up of subjects long enough/complete?
Not long enough for exposures affect on outcome to manifest and be detected.
Completeness important because those lost may have different outcomes than those who continue the study (intention to treat analysis)
- Incomplete follow-up allows for selection bias.
- May reduce power of the study
- A bias towards the null AND outcome exposure associations may have been underestimated.
209
CAT Cohort: Was the cohort recruited in appropriate way?
Importance of cohort being representative of population.
210
CAT Cohort: What are the results of the study?
Are they statistically significant.
211
CAT Cohort: Application?
Description of study cohort to allow comparison with local population.
- Assess if study cohort similar to local population
- Any treatment differences etc?
212
CAT Cohort: Appropriate study?
RCT are gold standard for assessing cause and effect.
However it is not ethical to allocate patients to certain exposure that may be harmful.
Not practical - patients may choose to change behaviour and this cannot be controlled.
213
CAT Case-control: Were the cases recruited in acceptable way?
Selection bias - may compromise validity
Are cases defined precisely?
- Representative of defined population?
- Established reliable system of selecting cases?
- Is time frame of study relevant to the disease/exposure?
- Sufficient number of participants (power)?
- Anything special about cases?
214
CAT Case-control: Was the controls selected in an appropriate way?
- Controls representative of defined population? - geographically/temporally
- Anything special about controls?
- Are they matched, population based or randomly selected?
- Sufficient number?
215
CAT Case-control: Was exposure accurately measured to minimise bias?
Measurement, recall or classification bias?
- Exposure clearly defined and accurately measured?
- Subjective/objective measurements?
- Do measures truly reflect this (validated)?
- Were methods similar in cases and controls?
- Did the study incorporate blinding where feasible?
216
Benefits of systematic reviews
Include all evidence to answer question.
Increase total sample size which improves power which improves certainty and precision.
217
What should be included in the search strategy of systematic reviews?
- Hand-searching
- Conference proceedings
- Contacting authors
- Papers in non-english
- Including unpublished studies
