Research Methods - Encountering evidence Flashcards
What is evidence-based medicine? (EBM)
• “the conscientious, explicit, and judicious use of current best evidence in making decisions about the care of individual patients” (David Sackett)
• Clinical decisions should be based on best evidence
• The problem determines the nature/source of evidence needed
• Best evidence is integrative
• Evidence should be identified, integrated,
appraised and applied
• Performance should be continually evaluated
Clinical decisions should be based on … …
Best evidence - the problem determines the nature/source of evidence needed
Best evidence is integrative
Evidence should be identified, integrated, appraised and applied
Performance should be continually evaluated
What is research ethical approval?
- Research ethics govern the standards of conduct for researchers
- Research ethics protect the dignity, rights and welfare of human and animal research participants
- Research ethics approval is provided by a committee (who review a research protocol, and to ensure appropriate ethical standards are being upheld)
Which studies require research ethical approvals?
Studies involving human/animal participants if - gathering novel data or information, creating knowledge that can be generalised beyond the participant sample or setting
Which studies require research ethical approvals?
Studies involving human/animal participants if:
- gathering novel data or information
- creating knowledge that can be generalised beyond the participant sample or setting
Which projects do not require ethical approvals?
Health improvement studies:
- service evaluations/service improvement/quality improvement (finds out something that service which can be/is used to improve that service)
- clinical audit (as per service evaluation, but evaluating service against a benchmark (e.g. national standard)
NB - NHS audits should be registered with the NHS trust in which they are being conducted
Which committee needs to be involved if the participants are NHS patients and/or staff / work conducted on NHS site/s?
NHS health research authority
Which committee needs to be involved if the participants are individuals (or animals/cells) recruited anywhere except the NHS site / no NHS site involved?
School research ethics officer (SREO) / BSMS RGEC (Research governance and ethics committee)
IRP project types that require ethics approval?
Epidemiological studies can be either …. Or ….
Descriptive: amount and distribution in populations (person, place and time)
Analytic: test hypothesis, identify and quantify risk or exposure
Descriptive study types include (4)…
Case reports, case series, ecological studies, cross sectional or prevalence survey
Observational study type examples (4)
Case control, cohort and ecological studies, cross sectional or prevalence surveys
Experimental study type examples:
Clinical trials, preventive trials, community intervention trials
Exposure is …
Risk factor for a disease/outcome e.g. obesity, socioeconomic status, gender
- can increase or decrease risk of disease/outcome
- those which decrease risk of disease/ (negative) outcome are protective
Outcome is …
Disease, condition or event of interest e.g. HIV/AIDS, depression, mortality relating to cardiovascular disease
… measures the frequency of “cases” of a disease (/condition/outcome) in a given population at a designated time (e.g. asthma) - the numerator
Prevalence measures the frequency of “cases” of a disease (/condition/outcome) in a given population at a designated time (e.g. asthma) - the numerator
Calculation of prevalence requires a suitable …
denominator (e.g. GP registered patients, schoolchildren) - the number of people who are ‘at risk’ of the disease
Prevalence
Number of people who have disease at given time/ number of people at risk of having disease at given time
Prevalence is expressed as either a …, a …. Or a … per unit of population
Prevalence is expressed as a percentage (e.g. 70%), a proportion of 1 (0.7 is equivalent), or a proportion per unit of population (700 of every 1000 people)
Why might we be interested in prevalence of asthma?
- Informing clinical diagnosis e.g. the more prevalent asthma, the more likely it is that this explains why the teenager is breathless…
- Understanding the disease/condition and its risk factors (exposures) and outcomes e.g. identifying risk factors for breathless teenagers, identifying changes over time/place
- Informing prevention and public health interventions e.g. educating teenagers/parents/public about asthma and its causes/consequences
- Informing service planning and commissioning e.g. ensuring enough people/resources available to treat number of breathless teenagers
Prevalence measures the frequency of … of a disease
“Cases” of a disease (/condition/outcome) in a given population at a designated time (E.g. asthma) - the numerator
Types of prevalence (3)
Point, period, Lifetime
Uses of different types of prevalence
Point and period - for better patient recall and tracking changes
Lifetime - short/fluctuating/episodic conditions
What is point prevalence?
The proportion of a population that has the characteristic at a specific point in time
What is period prevalence?
The proportion of a population that has the characteristic at any point during a given time period of interest
What is lifetime prevalence?
The proportion of a population who has ever had the characteristic at some point in their lifetime
Calculating prevalence:
Disease prevalence:
Prevalence of disease = (A+C)/ N
Prevalence of disease in exposed = A / (A+B)
Prevalence of disease in unexposed = C / (C+D)
Exposure prevalence:
Prevalence of exposure = (A+B)/ N
Prevalence of exposure in diseased = A / (A+C)
Prevalence of exposure in non-diseased = B/ (B+D)
Cross-sectional study design - a defined population is surveyed to measure … … …
Cross-sectional study design - a defined population is surveyed to measure variable/s of interest (e.g. disease/outcome (E.g. asthma), exposure (E.g. sedentary lifestyle,sex)
In cross-sectional study, the sample is selected using …
Inclusion and exclusion criteria (Determined by study aims/research questions/hypotheses) - could be general population or clinic-based
Prevalence is reported for the sample and/ or subgroups based on … status
Prevalence is reported for the sample and/ or subgroups based on exposure status
Is the cross-sectional study descriptive or analytic?
Can be analytic or descriptive
- Descriptive (about amount and distribution in population person, place and time)
- Analytic (test hypotheses, identify or quantify risk or protective factors)
Strengths and weaknesses of cross-sectional studies
+ Can compare prevalence in exposed and non-exposed to risk factors
Quick and inexpensive study type
Can be used to initially explore/inform a hypothesis, prior to another type of study
- Not suitable for rare diseases
- Cannot separate cause (exposure) and effect (outcome) as they are measured at the same time
- Cannot measure rate of new cases arising and any changes therein
Sources of cross-sectional study data?
Secondary - collected by someone else previously for a different purpose, then used by the researches in the specific study
Primary - collected by the researches directly from main sources for the specific purpose
Examples of secondary cross-sectional data
Mortality registers, hospital/medical records, census data, previous research studies
Examples of primary cross-sectional study data?
Survey data
Pros of secondary cross-sectional study data
Cheap, if anonymous, minimal to no ethical/governance approval needed
Negatives of secondary cross-sectional study data
Limited by what data already gathered, poor accuracy and missing data
Pros of primary cross-sectional study data
Control over how variable/s of interest measured
Negatives of primary cross-sectional study data
More expensive and time-consuming, difficult to achieve representative sample
Research methods =
Foundational skills for evidence-based medicine
Prevalence = number of people with a condition in a defined … at a given … in time e.g. number breathless teenagers now
Prevalence = number of people with a condition in a defined population at a given point in time e.g. number breathless teenagers now
Examples of prevalence: What are the types?
Current number of teenagers breathless now?
Current number of teenagers breathless July-august 2020?
Current number teenagers breathless ever in their lifetimes 2020?
Point – current number teenagers breathless now
Period- current number teenagers breathless July-August 2020
Lifetime- current number teenagers breathless ever in their lifetimes 2020
Define prevalence
Number of people with a condition in a defined population at a given point in time
How to calculate prevalence
Prevalence = number with disease at given time / number at risk of disease at given time
What is a key study design to establish prevalence?
Cross-sectional study
Advantages vs Disadvantages of a cross-sectional study
Advantages – cheap, quick
Disadvantages – some issues with usefulness for establishing prevalence of rare or time-limited diseases
What does incidence measure?
Measures the frequency of “new onset cases” of a disease (/condition/outcome) (E.g. Covid-19) in a given population over a designated period of time (E.g. 12 months)
what studies don’t need research ethics approval?
service evaluation
service improvement
quality improvement
audit
when do studies need research ethics approval?
if patients or staff are involved
point prevalence
The number of cases of a specific condition or disorder that can be found in a population at one given point in time.
period prevalence
A measure of how many individuals were affected by the disease during a specified time period.
lifetime prevalence
the proportion of people in the population who have had a disease/exposure at some point over their lifetime
what are point and period prevalence better for?
better patient recall
better at tracking changes
what is lifetime prevalence better for?
tracking short, fluctuating or episodic conditions
cross-sectional study
A study in which a representative cross section of the population is tested or surveyed at one specific time to measure variable of interest
Pros of Cross-Sectional Studies
quick
cheap
can compare prevalence in exposed and non-exposed
can initially form a hypothesis before another type of study
Cons of Cross-Sectional Studies
not good for rare exposures
can’t separate cause and effect
can’t measure incidence
descriptive study
a study designed to describe the amount and distribution of variables in populations
analytic study
an epidemiological study aimed at testing hypotheses, can identify and quantify risk
primary data sources
surveys, interviews, observation, community forums, focus groups
secondary data sources
mortality registers, medical records, census data, previous studies
