Robinson, Rosen, & Hogg, 1976). The VAS is a 10-centimeter line marked from 0, meaning no 60 JOURNAL OF HOLISTIC NURSING / March 2001 Downloaded from jhn.sagepub.com at PENNSYLVANIA STATE UNIV on March 4, 2016 pain, to 100, meaning pain as bad as can be. The respondent was asked to mark or identify the position between the two extremes that best described the degree of pain intensity being experienced at that moment. The test-retest reliability of the VAS as used with this sample was r = .621. Anxiety was measured using the State-Trait Anxiety Inventory (Spielberger, 1983). Developed in 1966, this inventory has been used more extensively in psychological research than any other anxiety measure (Buros, 1978). Trait anxiety is a measure of one’s general perceptions of stressful situations, whereas state anxiety is a measure of one’s anxiety at that moment in time (Spielberger, 1983). The test-retest reliability coefficients for each inventory ranged from .27 to .86, and the measures of internal consistency, such as the alpha coefficient, demonstrated a median coefficient of .93 for the samples of working adults. Construct validity has been determined for both instruments, as have high correlations (r = .90 to r = .94) between these instruments and other tools of anxiety measurement (Spielberger, 1983). The test-retest reliability coefficient for the state anxiety instrument in this sample was r = .647, and it was r = .886 for the trait anxiety instrument. For the participants in this study, internal consistency was demonstrated by the alpha coefficient that ranged from .82 to .93 on the state anxiety instrument and .90 to .95 for the trait anxiety instrument. Each instrument contains 20 questions that are answered using a 4-point Likert scale. On the state inventory, the responses range from not at all to very much so. On the trait inventory, the responses range from almost never to almost always. The least anxiety possible on each scale is a score of 20, whereas the greatest anxiety on each scale is a score of 80. A demographic questionnaire developed by the research team asked the participants to identify their age, gender, type of employment, hobbies, history with lower back pain, any pain medication(s) they were taking, and five inconveniences they were experiencing due to their lower back pain. An exit questionnaire, also developed by the research team, asked participants to indicate medication(s) taken during the study, how the treatments had affected the major inconveniences of lower back pain they had identified at the beginning of the study, and if they would recommend shiatsu treatments to friends or relatives with a similar complaint. The final component of the exit questionnaire was an open-ended question that invited participants Brady et al. / SHIATSU AND LOWER BACK PAIN 61 Downloaded from jhn.sagepub.com at PENNSYLVANIA STATE UNIV on March 4, 2016 to write anything they wished about shiatsu, the therapists, or the research study. Procedures This study was approved by the university’s institutional review board for human subjects. Volunteer participants were recruited for a convenience sample using printed advertising as well as word of mouth. Potential participants contacted The Shiatsu Clinic & School and were informed of the requirements for participation in the study. In addition, participants were told of the usual pricing structure for shiatsu treatments, the reduced price for those who volunteered for the study, and the refund schedule should a participant not complete all four treatments. Prior to their first treatment, participants were asked to read and sign an informed consent form and were given a copy of the signed form to keep. Participants were randomly assigned to a therapist. Random assignment was performed by the participant rolling one of a pair of dice; even numbers were assigned to the male therapist, odd numbers to the female therapist. In addition, whenever the therapists had a client who could qualify for the study, they informed the client and invited him or her to join the study. These subjects were not randomly assigned because they were already at the clinic with one or the other of the therapists. Both therapists are nationally certified in therapeutic massage and bodywork and licensed massage therapists in the state of Iowa. These therapists are both on the faculty of The Shiatsu Clinic & School. Each subject received massage from the same therapist and in the same private room throughout the study. Prior to each of four shiatsu treatments, each subject completed Spielberger’s (1983) State-Trait Anxiety Inventory and indicated on the VAS (Huskisson, 1974) the degree of lower back pain being experienced at that time. A full-body shiatsu treatment taking 50 to 60 minutes followed. Immediately after the treatment, the subject again completed the State-Trait Anxiety Inventory and described the level of lower back pain using the VAS. Two days following the completion of the treatment, the subject was contacted by telephone and asked once again to describe, using the VAS, the level of lower back pain being experienced at that time. This pattern continued throughout four shiatsu treatments, all scheduled within an 8-week period of time for each subject. 62 JOURNAL OF HOLISTIC NURSING / March 2001 Downloaded from jhn.sagepub.com at