Massage therapy improves mood and behavior of students with attention-deficit/hyperactivity disorder

Adolescence, Winter, 2003 by Sonya Khilnani, Tiffany Field, Maria Hernandez-Reif, Saul Schanberg

Mood state. Information about mood state was collected through the developmentally appropriate use of pictorial self-reports. A faces scale was used to provide an estimate of the participants' experiences. This pictorial measure was a modification of the self-report method designed to assess pediatric patients' anxiety and pain during invasive medical procedures, such as bone marrow aspirations (LeBaron & Zeltzer, 1984). LeBaron and Zeltzer used faces showing increasing degrees of distress and found a strong correlation (r = .63, p < .001) between patient ratings and observer ratings of anxiety. In the present study, drawings of four faces ranging from sad (scored as one) to happy (scored as four) were presented to the participants before and after the first and last assessment sessions (for both massage and control groups). The participants were then asked which face best described the way they felt at that moment. This scale's utility in previous massage studies supported its use as a self-report measure in this investigation.

A modification of the Children's Pain/Fear Thermometer Rating Scale (a drawing of a thermometer with a vertical scale) was used as a second measure of mood state. Participants were asked to select the point on the scale, ranging from zero (i.e., not feeling good at all) to ten (i.e., best I have ever felt), that best described the way they felt. The original scale was used in a study of pain in pediatric cancer patients undergoing bone marrow aspirations (Jay, Ozolins, Elliot, & Caldwell, 1983). In that study, pain ratings were significantly correlated (r = .67, p < .02) with observed behavior in youths aged 8 years and older. In the current study, this scale was administered before and after the first and last assessment sessions (for both massage and control groups).

Longer-Term Measure

Teachers (who were blind to group assignment) were asked to complete the Conners Teacher Rating Scale, which provided an index of changes in classroom behavior across the course of the study. This measure was used at the first session and four weeks later for both groups.

Classroom behavior. The Conners Teacher Rating Scale (Conners, 1969) is one of the most widely used behavior rating scales for assessing externalizing symptoms in children. The following six factors from this 39-item scale were examined: hyperactivity, conduct problems, emotional-indulgent, anxious-passive, asocial, and daydream/attention problems. Test-retest reliability coefficients over a one-month interval ranged from .70 to .90 across factors (Conners, 1973). Inter-teacher agreement of .92 was reported for the entire scale (Trites, Blouin, & Laprade, 1982).

RESULTS

A t test was performed on age, and chi-square tests were performed on gender and ethnicity. No significant differences were noted between the massage group and wait-list control group on these background variables.

Repeated-measures analyses of variance (ANOVAs) were conducted with the massage therapy and wait-list control groups as the grouping variable and session (pre, post) and day (first, last) as the repeated measures. Significant interaction effects were followed by alpha-corrected post hoc comparison t tests.