Baselines

Research

To demonstrate an effect of treatment (or lack thereof), a goal of treatment studies across disciplines, initial data must be obtained prior to implementation of the manipulated (independent) variable. These values collected before the implementation of a treatment (e.g., medication, a newly developed clinical tool, a specific treatment approach) are considered baseline data. Study designs and methodologies vary greatly based on the question being asked or the phenomena under [Page 263]investigation, but all treatment studies require some form of baseline data for comparison.

Treatment research can take on an AB design, where A represents the initial baseline data and B represents data taken during a treatment condition. In this case, a baseline condition exists for which data are collected in the absence of a treatment. A treatment condition is then employed, and data are again collected. Data are then analyzed to determine if there is a difference between the A and B conditions (i.e., if treatment made an impact). ABA designs can also be used, in which the final A represents a removal of the treatment condition in an effort to see if assessed values return to baseline after treatment is discontinued.

In studies in which the removal of the treatment condition is unethical or in which values are not expected to return to baseline condition (as is the case with effective speech and language therapy), multiple-baseline designs may be used. In this design, implementation of the treatment condition is staggered across behaviors, contexts, or participants. Staggering the treatment condition helps ensure that any change noted is attributable to the treatment and is not affected by additional factors.

In studies with large numbers of participants, the baseline data can be subject to statistical analysis, leading to an understanding of the population from which the study participants were taken. Baseline data can also be used to ensure that comparison groups are adequately representative of the population at large (i.e., if baseline values being tested are known for the population).

Clinical Practice

In clinical practice, there is an expectation that well-chosen therapy tasks, with appropriate mediation from a trained clinician, will result in progress toward the goals set at the implementation of therapy. The initial assessment provides an overall picture of a client’s strengths and weaknesses, enabling a clinician to set realistic and meaningful goals for each client. For each goal, baseline values are collected related to the client’s initial levels of performance on goal-related tasks (e.g., percentage correct on a naming task).

...