ALERT! Warning: your browser isn't supported. Please install a modern one, like Firefox, Opera, Safari, Chrome or the latest Internet Explorer. Thank you!
Log inLog in to ACORN Wiki. | RegisterSign on to ACORN Wiki.

Measuring Change

All measurement of treatment outcomes presumes an amount of pre-post change, however pre and post may be defined. Clinical trials have the luxury of pre-determining a fixed dose and length of treatment and then simply measuring change from the beginning to the end of the trial.

Clinical trials, however, are plagued by the fact that not all participants complete the prescribed treatment. Simply obtaining a pre and post measure of change at predetermined points in time fails to measure outcomes for patients who did not complete the expected length of treatment. Of course, in the real world practice, this problem is much greater because the length of treatment varies greatly from one patient to another.

In order to avoid the limitations of pre and post measurement at predetermined points in time, most researchers advocate the use of repeated measures administrated throughout the treatment episode and the use of an intent-to-treat analysis when determining the outcome of care. An intent-to-treat analysis simply measures change from the first measurement to the last measurement in the treatment episode, regardless of when that occurred.

This method is most appropriate for effectiveness studies, which seek to evaluate treatment outcomes in real world settings. See EfficacyAndEffectiveness for further discussion of the distinction between efficacy and efficacy research.

The intent-to-treat method rewards clinicians who are expert at keeping patients engaged in treatment until optimal benefit is realized. This method also tends to reward clinicians who administer the outcome measure as frequent intervals, as these clinicians are best able to monitor patient improvement and identify patients at risk for premature termination and consequently poor outcome. Frequent measurement also assures that the pre-post change score reflects all of the change created during the treatment episode. See TherapistFeedback for further discussion of the benefits of clinicians receiving feedback from patients via the routine use of outcomes questionnaires.

The simplest method to estimate the magnitude of improvement is to calculate the pre-post raw score change. However, simple change scores also have several serious limitations, including artifacts to regression to the mean. See topics ChangeScore and RegressionToMean for further discussion.

The effect size statistic is a method for reporting change scores in a standardized manner that permits comparisons across different samples using different outcome measures.

A measure of change is not of much use unless there is some basis for comparison, such as a control group. However, in real world clinical practice, a control group is not possible (or ethical!). However, the use of case mix adjustment permits outcomes for any patient to be compared to the outcome to all other patients in the comparison sample after controlling for differences in intakes scores, diagnoses, etc.

Primary topics:

You may also view all topics in alphabetical order in WebTopicList.

-- JebBrown - 06 Jan 2007
This site is powered by FoswikiCopyright © by the contributing authors. All material on this collaboration platform is the property of the contributing authors.
Ideas, requests, problems regarding ACORN Wiki? Send feedback
Syndicate this site RSS