Medical Tests and Medical Studies

Medical Studies

1. Magnitude of Effect
• medical studies often aim to describe the relationship between a treatment (exposure) and outcome
• the magnitude of a treatment effect can be measured in both absolute and relative terms


1. Absolute Measures of Effect
1. Absolute Risk Reduction
• difference between the incidence of a disease/event in the control group and the incidence of the same outcome in the treated group


2. Number Needed to Treat (NNT)
• how many patients must be treated to see the effect of the treatment
• reciprocal of the absolute risk reduction
• the ideal NNT = 1
• provides valuable insight into the clinical relevance of the treatment effect


2. Relative Measures of Effect
1. Relative Risk
• calculates the ratio of incidence proportions between 2 different groups (treated + control)
• relative risk reduction can make treatments seem more effective than they are


2. Odds Ratio (OR)
• represents the probability of an outcome
• odds that an outcome will occur given exposure to a variable of interest, compared to the odds of the outcome occurring in absence of the exposure
• most commonly used in case-control studies
• an OR < 1 means that the risk of the outcome is lower in exposed individuals than in unexposed individuals; an OR > 1 means that the odds are higher in exposed individuals


2. Evidence-Based Medicine
• integration of the best research evidence with clinical expertise and patient values
• different types of medical studies provide different levels of research evidence



3. Types Of Medical Studies
1. Systematic Reviews
• summary of the medical literature that uses explicit and reproducible methods to systematically search, critically appraise, and synthesize information on a specific issue
• synthesize the results of multiple primary studies related to each other by using strategies that reduce biases and random errors
• since no study, regardless of its type, should be interpreted in isolation, a systematic review is generally considered the best form of evidence
• may or may not include a statistical synthesis called meta-analysis, depending on whether the studies are similar enough so that combining their results is meaningful
• Cochrane reviews are the most well-known type of systematic reviews


2. Randomized Controlled Trials (RCT)
• provide the highest level of evidence supporting causality
• primary advantage of RCTs is that confounding variables are distributed equally between groups, as long as the number of study subjects is sufficiently large
• single-blind study is when the patients do not know which treatment they received, and is intended to mitigate the placebo effect
• double-blind study is when neither the treating physician nor the patient knows which treatment they received, and is intended to mitigate investigator bias
• in practice, blinding in surgery is rarely practical


1. Intention to Treat Principle (ITT)
• outcome comparisons are based on the original randomization, not on the treatment the patient actually received
• patients may crossover if they are not able to complete the intervention
• if the ITT principle is not followed, then the benefits of randomization may be lost because an equal number of confounders across comparison groups cannot be guaranteed


3. Meta-Analyses
• considered a subset of systematic reviews
• many medical studies are underpowered to answer a particular research question and results of multiple small studies may be conflicting
• meta-analysis is a statistical technique that pools similar studies in an attempt to increase the statistical power of the analysis
• several small studies are combined to make one larger study – a study of studies
• it is important that the pooled studies be homogenous and evaluate similar end points and patient populations
• a failure to identify all the existing studies can lead to erroneous conclusions
• many relevant studies may not have been published if they did not reach the necessary statistical threshold
• now the most frequently cited form of medical research


4. Cohort Studies and Case-Control Studies
• longitudinal, observational studies without randomization, making them vulnerable to unpredictable confounding factors
• evaluate associations between diseases and exposures
• very common in medicine because many questions cannot be ethically answered by RCTs
• provide level 2 or 3 evidence


1. Cohort Studies
• participants are followed over time to observe the incident rate of disease or outcome in question
• may be prospective or retrospective in design
• no intervention or treatment is administered as part of the study
• no control group
• Nurses’ Health Study, Framingham Heart Study (1948) are ongoing examples


1. Advantages
• can assess causality
• can examine multiple outcomes for a given causality
• can calculate relative risks or rates of disease in exposed and unexposed individuals over time


2. Disadvantages
• selection bias
• inefficient for rare outcomes, or outcomes that occur a long time after group assignment
• long duration of follow up is required, which is time consuming and expensive
• patients can be lost to follow up


2. Case-Control Studies
• retrospective studies that start with the outcome of interest and work backward to the exposure
• patients with a disease are identified and compared with controls for exposure to a risk factor
• controls are from the same source population but without the outcome
• one famous example linked smoking to lung cancer in the 1950s


1. Advantages
• useful for examining rare outcomes or outcomes with a long latency
• relatively quick and inexpensive
• can use existing records


2. Disadvantages
• control group selection may be difficult
• can be difficult to validate information retrospectively
• susceptible to bias
• relative risk or incidence of disease cannot be calculated
• odds ratio must be used to provide an estimate of relative risk


5. Cross-Sectional Study
• prevalence studies or censuses
• involve data collected at a single point in time
• cannot assess cause and effect (no temporality)
• mostly used for hypothesis generation instead of hypothesis testing
• one example of this type of study is to determine whether surgical patients are receiving adequate VTE prophylaxis


6. Case Reports and Case Series
• a case report highlights an unusual or unexpected event
• a case series demonstrates that the unusual event can occur more than once
• most useful for hypothesis generation
• an example is several reports of port site recurrence in the early days of laparoscopic colon cancer resections