CASE 1 – RESEARCH DESIGNS
The different kind of research rated from high to low for quality of evidence (cause -> effect)
❖ Systematic reviews & Meta analysis
A systematic review answers a specific research question by evaluating and summarizing all the
studies on the topic. A meta-analysis goes on to use statistical methods to combine the data from the
studies.
➔ PROS: comprehensive evaluation, reduction of (publication) bias, heterogeneity
➔ CONS: retrospective nature (limited to bias in original studies), potential
misinterpretation, inability to answer all questions, complex calculations (in meta-
analysis)
EXPERIMENTAL DESING
❖ Randomized controlled trials (RCT)
Randomized controlled trials (RCT) are prospective studies that measure the effectiveness of a new
intervention or treatment. Although no study is likely on its own to prove causality, randomization
reduces bias and provides a rigorous tool to examine cause-effect relationships between an
intervention and outcome. This is because the act of randomization balances participant
characteristics (both observed and unobserved) between the groups allowing attribution of any
differences in outcome to the study intervention. This is not possible with any other study design.
➔ PROS: Causality, minimization of cofounding variables, control group, blindings, minimize
selection bios (randomization), replicability.
➔ CONS: cost and time, high internal validity but lower external validity (= lower
generalizability), sometimes not practical (long term effects/extensive follow up), participant
dropout and compliance.
Cluster/group randomized trial: clusters (schools, hospitals, villages) are randomly allocated to
different interventions. All individuals in each cluster receive the same treatment. => public health
,NON EXPERIMENTAL DESIGNS: OBSERVATIONAL STUDIES – INDIVIDUAL LEVEL
❖ Cohort study
Cohort studies are a type of longitudinal study—an approach that follows research participants over
a period of time. In a cohort study, the population under investigation consists of individuals who are
at risk of developing a specific disease or health outcome. In a cohort study, the participants do not
have the outcome of interest to begin with. They are selected based on the exposure status of the
individual. They are then followed over time to evaluate for the occurrence of the outcome of
interest.
➔ PROS: establish temporal sequence, investigation of multiple outcomes, relative risk
estimation (RR), are well suited for rare exposures of those with long latency periods.
➔ CONS: require long-term follow-up, biased estimates due to loss to follow-up, confounding
impact, selection bias (=not representative for the general population)
PROSPECTIVE: Researchers collect information on exposure and potential confounding variables at
the start of the study and follow participants over time to measure outcomes.
RETROSPECTIVE: Exposure and outcome data are collected from existing records, such as medical
records, databases, or archival sources
❖ Case control
Used to examine factors associated with disease or outcomes. It involves comparing two groups:
cases, who have the outcome of interest, and controls, who do not. Researchers analyse the medical
and lifestyle histories of both groups to identify factors associated with the disease or outcome.
Case-control studies are often referred to as retrospective studies because they look back at
historical data to assess associations
➔ PROS: efficient for rare outcomes or diseases with long latency periods. Allows estimation of
odds ratio, conducted quickly and low cost. Data at individual level.
➔ CONS: Prone to bias (recall, selection and info), not suitable for very rare expositions. Cannot
use incidence to calculate RR! Cases and controls are selected, and have not occurred.
❖ Case series / Case reports
Collection and analysis of detailed information on individual cases of a particular disease, condition,
or medical phenomenon. A case report is a detailed report of the diagnosis, treatment, response to
treatment, and follow-up after treatment of an individual patient. A case series is group of case
reports involving patients who were given similar treatment.
➔ PROS: prepared for illustrating novel, unusual/rare, or atypical features identified in patients
in medical practice, and they potentially generate new research questions.
➔ CONS: not representative for broader population (selection bias), cannot establish causality
or determine prevalence of a disease in a population.
Cross-sectional studies: data is collected from a population at a single point in time (measured at the
same time). These studies are often used to estimate the prevalence of a condition, identify
associations between variables, or generate hypotheses for further research.
➔ PROS: easy, cheap and fast
➔ CONS: causality??
NON EXPERIMENTAL DESIGNS: OBSERVATIONAL STUDIES – POPULATIONAL LEVEL
, Ecological studies: examines relationships between exposures and health outcomes at the population
level. Typically correlation are used to identify potential associations. Often used for hypothesis
generation. Useful for public health, not individual level.
PREVALENCE – INCIDENCE – RR – OR
Prevalence: the number of existing cases of a disease at a given time period (regardless of when they
first developed it).
- Point prevalence: The proportion of individuals in a population who have a particular disease
or condition at a specific point in time.
- Lifetime Prevalence: The proportion of individuals in a population who have ever been
diagnosed with a particular disease or condition at any point in their lifetime.
Incidence: the number of new cases of a characteristic that develop in a population in a specified
time period
- Cumulative incidence: the proportion of individuals who develop a new case of a disease or
condition within a specified time period among those who are initially disease-free.
RELATIVE RISK (RISK RATIO; RR): risk of an event in one group (e.g., exposed group) versus the risk of
the event in the other group (e.g., nonexposed group)
-> RR is calculated using incidence rates
-> Used in cohort studies, including prospective and retrospective
ODDS RATIO (OR): ratio of odds of an event in one group (exposure among cases) versus the odds of
the event in the other group (exposure among controls).
-> OR is calculated using prevalence data,
-> Used in case-control studies and situations where the outcome is rare.
These both show the strength/association between exposure and outcome. In general, RR is
preferred when the outcome is rare, as it provides a more accurate estimate of risk than OR in this
scenario. OR is a good approximation of RR when the outcome is rare, but as the prevalence of the
outcome increases, OR may overestimate RR.
Interpretation:
0 < RR < 1 → Protective factor = indicates a decreased risk among exposed individuals
RR = 1 → No effect = no association between exposure and outcome
RR > 1 → Risk factor = increased risk among exposed individuals
--------------------------------------------------------------------------------------------------------------------------------------
The Case: study design for proving whole grain rich diet is preventive for diabetes
RTC = difficult with timing of study, how long follow up, how control group? Cohort is also difficult for
that long, follow up is almost impossible. Retrospective is difficult with recall bias. Cross-over design
for mechanism, but cannot look at incidence level. Best possibility: Case control study with the data
of prospective cohort from consumption of whole grain bread and probability of type2 diabetes
The different kind of research rated from high to low for quality of evidence (cause -> effect)
❖ Systematic reviews & Meta analysis
A systematic review answers a specific research question by evaluating and summarizing all the
studies on the topic. A meta-analysis goes on to use statistical methods to combine the data from the
studies.
➔ PROS: comprehensive evaluation, reduction of (publication) bias, heterogeneity
➔ CONS: retrospective nature (limited to bias in original studies), potential
misinterpretation, inability to answer all questions, complex calculations (in meta-
analysis)
EXPERIMENTAL DESING
❖ Randomized controlled trials (RCT)
Randomized controlled trials (RCT) are prospective studies that measure the effectiveness of a new
intervention or treatment. Although no study is likely on its own to prove causality, randomization
reduces bias and provides a rigorous tool to examine cause-effect relationships between an
intervention and outcome. This is because the act of randomization balances participant
characteristics (both observed and unobserved) between the groups allowing attribution of any
differences in outcome to the study intervention. This is not possible with any other study design.
➔ PROS: Causality, minimization of cofounding variables, control group, blindings, minimize
selection bios (randomization), replicability.
➔ CONS: cost and time, high internal validity but lower external validity (= lower
generalizability), sometimes not practical (long term effects/extensive follow up), participant
dropout and compliance.
Cluster/group randomized trial: clusters (schools, hospitals, villages) are randomly allocated to
different interventions. All individuals in each cluster receive the same treatment. => public health
,NON EXPERIMENTAL DESIGNS: OBSERVATIONAL STUDIES – INDIVIDUAL LEVEL
❖ Cohort study
Cohort studies are a type of longitudinal study—an approach that follows research participants over
a period of time. In a cohort study, the population under investigation consists of individuals who are
at risk of developing a specific disease or health outcome. In a cohort study, the participants do not
have the outcome of interest to begin with. They are selected based on the exposure status of the
individual. They are then followed over time to evaluate for the occurrence of the outcome of
interest.
➔ PROS: establish temporal sequence, investigation of multiple outcomes, relative risk
estimation (RR), are well suited for rare exposures of those with long latency periods.
➔ CONS: require long-term follow-up, biased estimates due to loss to follow-up, confounding
impact, selection bias (=not representative for the general population)
PROSPECTIVE: Researchers collect information on exposure and potential confounding variables at
the start of the study and follow participants over time to measure outcomes.
RETROSPECTIVE: Exposure and outcome data are collected from existing records, such as medical
records, databases, or archival sources
❖ Case control
Used to examine factors associated with disease or outcomes. It involves comparing two groups:
cases, who have the outcome of interest, and controls, who do not. Researchers analyse the medical
and lifestyle histories of both groups to identify factors associated with the disease or outcome.
Case-control studies are often referred to as retrospective studies because they look back at
historical data to assess associations
➔ PROS: efficient for rare outcomes or diseases with long latency periods. Allows estimation of
odds ratio, conducted quickly and low cost. Data at individual level.
➔ CONS: Prone to bias (recall, selection and info), not suitable for very rare expositions. Cannot
use incidence to calculate RR! Cases and controls are selected, and have not occurred.
❖ Case series / Case reports
Collection and analysis of detailed information on individual cases of a particular disease, condition,
or medical phenomenon. A case report is a detailed report of the diagnosis, treatment, response to
treatment, and follow-up after treatment of an individual patient. A case series is group of case
reports involving patients who were given similar treatment.
➔ PROS: prepared for illustrating novel, unusual/rare, or atypical features identified in patients
in medical practice, and they potentially generate new research questions.
➔ CONS: not representative for broader population (selection bias), cannot establish causality
or determine prevalence of a disease in a population.
Cross-sectional studies: data is collected from a population at a single point in time (measured at the
same time). These studies are often used to estimate the prevalence of a condition, identify
associations between variables, or generate hypotheses for further research.
➔ PROS: easy, cheap and fast
➔ CONS: causality??
NON EXPERIMENTAL DESIGNS: OBSERVATIONAL STUDIES – POPULATIONAL LEVEL
, Ecological studies: examines relationships between exposures and health outcomes at the population
level. Typically correlation are used to identify potential associations. Often used for hypothesis
generation. Useful for public health, not individual level.
PREVALENCE – INCIDENCE – RR – OR
Prevalence: the number of existing cases of a disease at a given time period (regardless of when they
first developed it).
- Point prevalence: The proportion of individuals in a population who have a particular disease
or condition at a specific point in time.
- Lifetime Prevalence: The proportion of individuals in a population who have ever been
diagnosed with a particular disease or condition at any point in their lifetime.
Incidence: the number of new cases of a characteristic that develop in a population in a specified
time period
- Cumulative incidence: the proportion of individuals who develop a new case of a disease or
condition within a specified time period among those who are initially disease-free.
RELATIVE RISK (RISK RATIO; RR): risk of an event in one group (e.g., exposed group) versus the risk of
the event in the other group (e.g., nonexposed group)
-> RR is calculated using incidence rates
-> Used in cohort studies, including prospective and retrospective
ODDS RATIO (OR): ratio of odds of an event in one group (exposure among cases) versus the odds of
the event in the other group (exposure among controls).
-> OR is calculated using prevalence data,
-> Used in case-control studies and situations where the outcome is rare.
These both show the strength/association between exposure and outcome. In general, RR is
preferred when the outcome is rare, as it provides a more accurate estimate of risk than OR in this
scenario. OR is a good approximation of RR when the outcome is rare, but as the prevalence of the
outcome increases, OR may overestimate RR.
Interpretation:
0 < RR < 1 → Protective factor = indicates a decreased risk among exposed individuals
RR = 1 → No effect = no association between exposure and outcome
RR > 1 → Risk factor = increased risk among exposed individuals
--------------------------------------------------------------------------------------------------------------------------------------
The Case: study design for proving whole grain rich diet is preventive for diabetes
RTC = difficult with timing of study, how long follow up, how control group? Cohort is also difficult for
that long, follow up is almost impossible. Retrospective is difficult with recall bias. Cross-over design
for mechanism, but cannot look at incidence level. Best possibility: Case control study with the data
of prospective cohort from consumption of whole grain bread and probability of type2 diabetes
