Lecture 36 - Critical Appraisal of Studies that Measure Patients' Experience

what is the JAMA user’s guide to using articles that measure patient reported outcomes

1. is the measurement of health-related quality of life important

2. is the study valid

3. what are the results? - how can we interpret the scores

4. how can I apply the results to patient care

what are qualities of a well done study design assessing health status

clearly defined research question relating to health status

appropriate instrument for measuring health status

psychometric properties

definition and validation of the minimal important difference

sample size

use of allocation concealment and blinding

how should a trial assessing health status define their research question

pre-specify that change in health status is a primary (/co-primary) outcome

state conceptual and measurement model

define what is being measured and why, which domains are being assessed

hypothesize magnitude and direction of the treatment and induced change

what is face validity

the extent to which an instrument appears to be measuring what it intends to measure

what is content validity

the extent to which all relevant concepts or dimensions of health status important to patients are comprehensively sampled by the items and domains in the instrument

what is construct validity

the extent to which the responses to items correlate, as logic would dictate, with values obtained for the same patient using other commonly used clinical measures (i.e. does the instrument relate to other measures the way it should?)

what is reliability

the extent to which the instrument yields the same (or highly correlated) scores on repeated applications to the same stable patients

what is responsiveness

the ability of the instrument to detect differences in scores over time in individuals who have changed with respect to the concept being measured

what is precision

the extent to which an instrument is free of random error

what is reproducibility

for instruments scored by clinicians, intra-rater and inter-rate reliability should be high

what is internal consistency

different items within a single domain are homogenous or cohesive in their responses, suggesting measurement of aspects that of a single attribute or construct

what is interpretability

the extent to which easily understood meaning can be assigned to an instrument scores