epi test 2

Phase 1: Safety

Safety, dosage, side effects, 20 - 80 subjects 

Phase ll: Efficacy

Effectiveness, relative safety, 100 - 300 patients

Phase lll: comparison to current treatments 

Efficacy and safety, large scale, multi-center may lead to FDA approval

Phase IV: Long term and rare effects 

Post marketing surveillance; long term effects in large populations

Internal validity = study provides true results 

Study design, conduct, and analysis answer the research question truthfully 

External validity = study is generalizable 

• Findings applicable to other settings 

• Consistency of findings

VALIDITY: selection bias (participation) 

• Participants are likely to differ from nonparticipants 

• Representativeness of participants affects external validity (“generalizability”) not necessarily internal validity 

How can we prevent, detect, and treat disease? (experimental): interventions -

new screening techniques, medication, surgical interventions, diet and exercise programs, educational interventions, genetic screenings 

What causes disease? (observational):

risk factors - age, gender, bacteria or virus, exposure to toxins, diet, exercise/activity, genetics, stress, childhood trauma, body mass index and smoking/alcohol  

Advantages of randomization: 

• Randomization prevents potential biases from influencing assignment 

• Randomization increases compatibility of treatment groups for Known AND Unknown confounders 

• It's the only way to effectively control for unknown factors 

• Compatibility is NOT ensured, but any differences are due to chance

Practical issues affecting validity of experimental studies:

who to include?, assigning treatment, blinding, compliance 

Masking aka blinding: 

Single blind: participant doesn’t know treatment 

Double blind: neither participant nor investigator know treatment 

• Applies to data collection and analysis 

Validity: assigning "treatment" groups

• “Experimental” treatment

  • May have more than one tested at a time 

• Control 

  • Nothing 

  • Placebo 

  • Active alternative 

    • “Usual care”; variety of treatments

May have more than one type

unit of randomization- individual

intervention is allocated to individual participants

unit of randomization - community

intervention is allocated at an aggregate level (city, clinic, school)

preventive

intervention is given to a target population to prevent disease

therapeutic

intervention is given to people with a disease to cure or improve outcomes

Trial design

parallel: interventions are administered to each group concurrently (simultaneously)

crossover: groups receive treatments sequentially, one after the other

simple (part of trial design)

each group gets 1 treatment (active vs. placebo)

factorial

each group gets 2 or more treatments

FACTORIAL DESIGNS

• studying effects of two independent treatments (A & B)

• randomize into one of four intervention groups

• measure outcome of interest after interventions

stratified randomization

a method used in clinical trials and experiments to ensure that important baseline characteristics (known as prognostic factors or covariates) are evenly balanced between treatment and control groups

true or false: the distinguishing feature of an experimental study is that an intervention / exposure is assigned by the investigator

true

in a particular clinical trial, every participant receives every treatment in the study but in a different order. what type of trial design does this study utilize?

crossover

what type of bias can be caused by stringent exclusion criteria in randomized controlled trials

selection bias

true or false: the intent-to-treat approach to analysis of randomized controlled trials does not eliminate data from subjects who failed to comply with the intervention

true

if you strongly believe that one treatment will yield better results than another before conducting a trial, you lack

equipoise

Masking a.k.a. Blinding: SINGLE BLIND

participant doesn’t know treatment 

Double blind:

neither participant nor investigator know treatment 

what is screening?

A set of procedures for early detection and treatment of a disease affecting a population

(SCREENING) The application of a test or procedure to asymptomatic, apparently well individuals to separate those with a:

High probability of having or developing a given disease from those with a

low probability of having or developing the disease

characteristics of screening tests

Validity: the ability to correctly predict who has the disease and who does not 

Reliability: consistency 

test retest

• the consistency of a measurement from one time to another

inter-observer reliability

• the degree to which different raters/observers give consistent estimates of the same phenomenon

Reliability

Ability of a test to give consistent results when performed more than once on the same individual under the same conditions

Validity

the ability to correctly identify who has the disease and who does not

sensitivity:

the ability of a test to correctly identify those who have the disease

specificity:

the ability of a test to correctly identify those who DO NOT have the disease

as sensitivity increases …

the number of false positives likely rises too (specificity decreases) and vice versa

Lead time bias

screening finds the disease earlier, so survival appears to be longer, even though death occurs at the same time

Patient self selection (volunteer) bias:

individuals who participate in the screening programs may differ from those who do not on characteristics that may be related to survival