The Arts in Psychotherapy 40 (2013) 224– 229

Contents lists available at SciVerse ScienceDirect

The Arts in Psychotherapy

Effects of music therapy on perception of stress, relaxation, mood, andside effects in patients on a solid organ transplant unit: A randomizedeffectiveness study

Ian Crawford, MT-BC, Tyler Hogan, MT-BC, Michael J. Silverman, PhD, MT-BC !

University of Minnesota, United States

a r t i c l e i n f o

Keywords:MoodMusic therapyRandomizedRelaxationSolid organ transplantStress

a b s t r a c t

Solid organ transplant donors and recipients typically present with a variety of potential stressors at bothpre- and posttransplant, which may lead to symptoms such as psychological distress and depression. Thepurpose of the study was to determine the effects of single-session music therapy on stress, relaxation,mood, and perception of side effects in hospitalized solid organ transplant donors and recipients. Partic-ipants (N = 38) were randomly assigned to experimental or wait-list control conditions in a pre–posttestsingle-session design. As sense of control has been associated with positive health outcomes in solid organtransplant patients, participants were offered the choice of receiving either patient-preferred live musicor a brief harmonica lesson followed by a blues performance session with guitar accompaniment. Thoughthere were no between-group pretest differences, there were significant between-group posttest differ-ences in relaxation, stress, and mood, with experimental participants having more favorable scores thancontrol participants. Thirty seven participants selected patient-preferred live music while one patientselected the brief harmonica lesson. From the results of this randomized controlled study, it seems thatmusic therapy can be an effective psychosocial intervention concerning relaxation, stress, and mood forsolid organ transplant patients. Additionally, most participants selected receptive music therapy in theform of patient-preferred live music rather active music therapy in the form of a brief harmonica lesson.Results are congruent with existing literature. Limitations of the study, suggestions for future research,and implications for clinical practice are provided.

© 2013 Elsevier Inc. All rights reserved.

Literature review

In 1990, medical specialists performed approximately 15,000organ transplants in the United States. By 2009, the number oftransplants had increased to more than 28,000. Such an increaseis notable and serves as a testament both to the abilities of modernscience as well as to the generosity of organ donors. However, thenumber of transplants performed still pales in comparison with thenumber of people on the waitlist for organ donations. In 2009, thewaitlist for organs was over 104,000 names long, with 4000 namesbeing added each month (National Kidney Foundation, 2011).

Organ donation websites reveal the ongoing need for donors, aswell as the joy and excitement experienced by transplant recip-ients. A new organ can lead to an improved and extended lifefor someone experiencing organ failure. However, the transplantexperience is not complete after surgery or hospital discharge.Numerous stressors can be present after transplantation, leadingtransplant patients to be at risk for maladaptive coping (Campbell

! Corresponding author. Tel.: +1 612 624 1091; fax: +1 612 626 2200.E-mail address: silvermj@umn.edu (M.J. Silverman).

& Etringer, 1999) and heightened anxiety and depression levels.Dew et al. (1996) found that one-third of the heart transplantrecipients in their study experienced high levels of psychologi-cal distress in the year following transplantation. Financial burdencan be a contributing factor that can lead to such distress. Med-ications required to support a heart transplant can cost up to$2000 per month (Dew et al., 1996). In addition to medications,patients might also pay for hospital visits and stays, medical main-tenance, and likely experience unemployment or may be unable towork.

Emotional and behavioral factors can also contribute to distressin transplant patients (Campbell & Etringer, 1999). After receivingan organ transplant, recipients typically undergo a lifestyle changein order to maintain the new organ. These often unexpected andundesirable changes may consist of frequent returns to the hospitaland a life-long requirement of immuno-suppressants. These medi-cations can have side effects including hypertension, osteoporosis,and diabetes (Campbell & Etringer, 1999). Moreover, patients wholack adequate coping resources are likely to experience depressionas a result: Campbell and Etringer (1999) reported that over 50% ofthe patients in their study were prescribed antidepressants at leastonce after transplantation.

0197-4556/$ – see front matter © 2013 Elsevier Inc. All rights reserved.http://dx.doi.org/10.1016/j.aip.2013.02.005

I. Crawford et al. / The Arts in Psychotherapy 40 (2013) 224– 229 225

Researchers have documented various psychosocial interven-tions that may alleviate symptoms experienced by transplantpatients. Mindfulness-based stress reduction (MBSR) has resultedin improved sleep and lowered depression and anxiety scores intransplant patients (Gross et al., 2004). Engle (2001) noted that cog-nitive behavioral interventions and emotional support groups havealso resulted in decreased anxiety among transplant patients.

While hospitalized organ transplant patients may receive excel-lent medical care, they do not often experience a high sense ofcontrol over their bodies or environments. However, in the lit-erature, researchers have repeatedly articulated the importanceof autonomy and sense of control for hospitalized patients. Ina review of literature, Thompson (1981) concluded that cogni-tive and behavioral control can lessen anticipatory anxiety andimprove post-event effects when facing aversive stimuli. Lachmanand Weaver (1998) found a sense of mastery or control was relatedto perceptions of better health, fewer depressive symptoms, andgreater life satisfaction. Among patients with cardiac disease, asense of personal control over life was found to be associated withfewer depressive symptoms (Penninx et al., 1998). Additionally,Bohachick, Taylor, Sereika, Reeder, and Anton (2002) reported thatpersonal control was related to positive psychological outcomesand found that patients with a higher sense of personal controlduring their hospital stays reported higher levels of optimism, sat-isfaction with life, and wellbeing at six-month follow-up.

Generalizing the concept of control and automony to musictherapy, Robb (2000, 2003a, 2003b) developed theroetical under-pinnings concerning the contextual support model of musictherapy as related to hospitalized pediatric cancer patients. WhileRobb primarily focused on hospitalized children, the applicationsmay be applicable to adult hospital patients. She noted the threebasic elements of the contextual support model were structure,autonomy support, and involvement. While structure and involve-ment are fundamental and their importance should not be negated,autonomy is the focus of the current study.

Patrick, Skinner, and Connell (1993) defined autonomy as theextent to which a person feels free to show behaviors of choice.In generalizing this theory to music therapy in a medical settingand incoporating Robb’s (2000, 2003a, 2003b) theories, a patientmight make a number of decisions during music therapy interven-tions, such as choice of song, tempo, genre, length of intervention,or even type of intervention. Thus, the patient can manipulate themusic therapy session, potentially providing a sense of control,mastery, and autonomy. Moreover, patients should first be pro-vided the choice of accepting or denying music therapy services, aluxury not often afforded to hospitalized patients by other health-care providers.

Although researchers have published numerous studies con-cerning the use of music and music therapy in medical facilities,to date, researchers have only published two studies measuringthe effects of music therapy with patients on a solid organ trans-plant unit. Using patient-preferred live music (PPLM), Madsonand Silverman (2010) found significant and positive differences inmeasures of anxiety, pain, relaxation, and nausea. However, exter-nal valdity was limited as the researchers did not utilize a controlgroup in the pre–posttest design of this exploratory study.

Ghetti (2011) examined the effects of Active Music Engagement(AME) and Emotional-Approach Coping (EAC) in music therapy onaffective states, pain, coping self-efficacy, satisfaction with hospi-talization, and willingness to ambulate in solid organ transplantpatients. The researcher utilized a three-group randomized con-trolled design. Kidney and liver transplant recipients (N = 29), whowere at least in the second day of postoperative hospitalization,were randomly assigned to a standard care group, AME group,or EAC and AME combined group. The AME group received anindividual, 30-min, patient-preferred music-making session with

the researcher, who encouraged the participants to sing or playalong with rhythm instruments as much as they felt comfort-able. Participants in the EAC and AME combined group receiveda similar intervention. Although these participants were encour-aged to engage in singing or instrument play during the preferredmusic-making session, the researcher prompted the participantsto choose from a list of musical selections that were thematicallylinked to the researcher’s perception of their emotional states. Theresearcher also discussed possible coping strategies in the session.The standard care group received no music therapy intervention.Ghetti found significant differences in positive affect between theAME and AEC and AME combined groups (p < .05), with AEC andAME combined group being more effective than AME or standardcare. She also found statistically significant differences in negativeaffect between groups (p < .05), with both AEC and AME combinedgroup and AME group reducing negative affect more than thestandard treatment group. AME participants reported lower lev-els of pain than AEC and AME combined participants or standardcare participants (p < .05). There was no significant between-groupdifference in coping self-efficacy, satisfaction with hospitalization,willingness to ambulate, or length of ambulation.

Although previous results from Madson and Silverman (2010)and Ghetti (2011) have supported the use of music therapy for solidorgan tranplant patients, there is a need for additional researchspecific to automony and music therapy with this population.Therefore, the purpose of the current study was to determine ifmusic therapy interventions can reduce stress and perception ofside effects and increase relaxation and improve mood of hos-pitalized solid organ transplant donors and recipients during arandomized controlled effectiveness study. Specific research ques-tions were as follows:

1. Will experimental participants have higher self-report measuresof relaxation and mood and lower measures of stress and per-ception of side effects than control participants?

2. Will patients choose PPLM or a brief harmonica lesson followedby a blues jam as their intervention?

Method

Research participants

Participants (N = 38) were 19 female (six control) and 19 male(eight control) patients on the solid organ transplant unit of a mid-western teaching hospital. As both donors and recipients werehospitalized on the same unit and the researchers wanted to beas inclusive as possible in an attempt to provide music therapy toanyone who may be interested, donors were also eligible for studyinclusion. The researchers of the present study utilized the inclu-sive approach of Dew et al. (2007), who recommended that donorsreceive comparable care to organ recipients throughout the trans-plantation process. Some participants had recently undergone atransplant (10 control), while others were returning to the hospitaldue to infection or other complications (five control). Thus, in thispurposely inclusive design, the only inclusion criterion was to bean inpatient on the unit receiving an initial music therapy session.Descriptive statistics concerning participants’ ages and number ofdays in the hospital prior to researcher contact are depicted inTable 1.

Design

Participants were randomly assigned via a computer programto either the experimental group (n = 22) or wait-list control group(n = 16). In an attempt to provide music therapy to all interested

226 I. Crawford et al. / The Arts in Psychotherapy 40 (2013) 224– 229

Table 1Descriptive statistics of participants’ ages and number of days spent in hospital.

Experimental group(n = 22)

Control group(n = 16)

M SD M SD

Age 50.64 16.00 54.19 12.44Days in hospital 6.68 6.64 4.31 2.24

patients on the unit, the researchers used a single-session pre- andposttest, wait-list control design. Experimental participants com-pleted the pretest, received music therapy, and then completed theposttest. After control participants completed the pretest, the prin-cipal investigator (PI) or secondary investigator (SI) returned after20 to 30-min and control participants then completed the posttest.Due to the use of a wait-list control group design in single-sessiontherapy, control participants received music therapy immediatelyafter completion of the posttest. The university and hospital sharedan Institutional Review Board that approved the study before datacollection initiated. The researchers completed all necessary andrequired research training to conduct this study.

Procedure

Upon entering a patient’s room, the PI or SI introduced him-self and asked if the patient would like to receive music therapy. Ifthe patient responded in the affirmative, the PI or SI asked if thepatient would like to take part in a research study. Thus, in anattempt to provide music therapy to as many patients as possible,patients had the option to receive music therapy but not partici-pate in the research study during this exploratory investigation. Ifa patient chose to be a research participant, the PI or SI explainedand obtained informed consent and the participant completed thepretest. If the patient was in the experimental condition, the PIor SI initiated 20 to 30-min of music therapy and the participantcompleted the posttest after the treatment. Participants in the con-trol group completed the pretest and 20 to 30-min later completedthe posttest without receiving music therapy. After completing theposttest, all participants in the wait-list control group receivedmusic therapy. Additionally, patients could receive multiple dosesof music therapy. In the cases of multiple music therapy doses, thePI or SI only collected data during the patient’s first music therapydose.

Regardless of condition, participants could receive either PPLMor a brief harmonica lesson (see Appendix A) followed by a bluesperformance session with guitar accompaniment. If a partici-pant chose PPLM, the researchers inquired about the participant’smusical preferences and played songs suited to those prefer-ences (Walworth, 2003). If the harmonica lesson was chosen,the researchers provided a brief harmonica lesson followed by aperformance session with guitar accompaniment. The lesson andperformance session were based on 12-bar blues progressions. Par-ticipants were allowed to keep their new harmonicas (for purposesof infection control and later use) and also received a CD contain-ing four pre-recorded blues accompaniment tracks at no cost. ThisCD could be used as accompaniment in the hospital or at home

Table 2Posttest descriptive statistics.

Dependent measure Experimental group(n = 22)

Control group(n = 16)

M SE M SE

Relaxation 2.11 0.30 4.60 0.35Stress 2.68 0.46 4.25 0.54Mood 1.68 0.31 3.56 0.36Side effects 3.29 0.49 4.42 0.58

for further harmonica practice. The tracks on the CDs were ofvarying tempos and, in an attempt to heighten moods using theiso-principle, were placed in order from slow to fast. Thus, a partic-ipant would be able to choose the track that best represented his orher current mood and progress in either direction toward the trackthat best represented his or her desired mood.

Instrument

The researchers collected pre- and posttest data with four sepa-rate 10-point Likert-Type Scales, where “1” represented a desirablecondition (i.e., no stress, good mood) and “10” represented anundesirable condition (i.e., high stress, poor mood). This scoringsystem was brief, congruent with existing and commonly utilizedhospital scales, and was recommended by the head nurse of theunit. Using these Likert-Type Scales, the researchers asked partici-pants to report their current levels of relaxation, stress, mood, andperception of side effects. The researchers did not utilize an estab-lished psychometric instrument in order to remain congruent withexisting hospital scales and in the interest of time. The PI or SIthen recorded any additional comments the participants wishedto share.

Results

Independent samples t-tests were conducted to determine ifthere were between-group pretest differences. There were nostatistically significant between-group differences in any pretestmeasures (all p > .18). Correlational analyses revealed significantcorrelations between all pre- and posttest measures. To determineif there were significant between-group posttest differences, theresearchers utilized an ANCOVA with pretest scores as a covariate,posttest scores as the dependent measure, and group as the fixedfactor in SPSS Version 19.0. Results were statistically significantfor relaxation (p < .001), stress (p = .036), and mood (p < .001), withthe experimental group having more favorable posttest scores thanthe control group. There was no difference across the two thera-pists. Posttest descriptive statistics are depicted in Table 2. Resultsof statistical analysis are depicted in Table 3.

Concerning the second research question regarding PPLM versusthe harmonica lesson, only one participant chose the harmonicalesson. This participant was in the wait-list control condition.

Seven experimental participants made comments on theposttest. As these comments tended to be brief, a detailed thematicanalysis was not conducted. However, all participants’ comments

Table 3Results of statistical analyses.

Dependent measure Overall statistic Experimental versus control group

(df) F p, partial !2 Mean difference Adjusted p values 95% CI

Relaxation (1,35) 28.683 .001, .450 "2.484 .001 "3.425, "1.542Stress (1,35) 4.770 .036, .120 "1.565 .036 "3.021, "0.110Mood (1,35) 15.451 .001, .306 "1.876 .001 "2.844, "0.907Side effects (1,35) 2.320 .144, .060 "1.130 .144 "2.665, 0.406

I. Crawford et al. / The Arts in Psychotherapy 40 (2013) 224– 229 227

were positive and supported music therapy on the solid organtransplant unit.

Discussion

Organ transplant patients who experience lower levels of anx-iety during hospitalization tend to recover more successfully thanthose with higher levels of anxiety (Dew et al., 1996). Additionally,patients with a higher sense of personal control during hospi-talization experience higher levels of optimism and wellbeing atsix-month follow-up (Bohachick et al., 2002). These results empha-size the importance of providing psychosocial interventions duringhospitalization, rather than after discharge. In the current study onan inpatient solid organ transplant unit, music therapy was able toimmediately improve experimental participants’ conditions.

The researchers designed the present effectiveness study toinvestigate the effects of music therapy on perceptions of stress,relaxation, mood, and side effects in solid organ transplant patientsusing a randomized controlled design. With significant between-group differences in posttest measures of relaxation, stress, andmood, it seems that music therapy may be a beneficial psychoso-cial intervention for hospitalized solid organ transplant patients.These findings are congruent with those of previous researchers,who found that music therapy can improve levels of pain, anxiety,relaxation and nausea on the solid organ transplant unit (Ghetti,2011; Madson & Silverman, 2010). However, positive changes inthe experimental condition may also be a result of the automony inthe form of patient choice of active music therapy (harmonica les-son and performance session) or receptive music therapy (PPLM).Thus, future researchers might measure the effects of interventionchoice compared without intervention choice.

There is currently no research involving the use of harmonicalessons as a music therapy intervention in a medical setting. Theresearchers in the current study offered participants the option ofreceiving either PPLM or a harmonica lesson followed by a bluesjam session. Only one participant chose the harmonica option, andthe lone harmonica participant was a member of the wait-list con-trol group. Thus, the researchers did not collect data regardingthe effects of harmonica lessons on the dependent measures.Results concerning the PPLM preference are congruent with relatedresearch indicating that medical patients prefer receptive formsof music therapy such as PPLM (O’Callaghan & Colegrove, 1998).Moreover, previous researchers have found that patients who havenever received music therapy were more likely to agree to partic-ipate in a music therapy session if the therapist did not requirethe patient to actively make music through singing or instrumentplay (Burns, Sledge, Fuller, Daggy, & Monahan, 2006). However, itmay be that after multiple doses of music therapy, patients aremore amenable to other types of interventions (i.e., songwriting,lyric analysis, therapuetic music videos, harmonica lessons). Thiscertainly represents an area for future inquiry.

It should be noted, however, that the PI observed positivechanges in both affect and verbalizations of the lone participantwho opted for a harmonica lesson during and after the musictherapy session. This participant smiled often and made manycomments about how much she enjoyed the session. She alsonoted the session was “relaxing.” The participant seemed surprisedand excited that she could learn to play the harmonica in such ashort amount of time. Additionally, a number of nurses enteredthe patient’s room during the session to compliment the patientand express awe at how talented she was. While these observa-tions were not measured or recorded objectively, the researchersfeel that the experience offered the participant a sense of con-trol, mastery, and success, and that further systematic investigationmay reveal additional positive results and potential mechanismsof change. While future research is certainly warranted, initial

anecdotal evidence seems to positively support the feasibility ofthe harmonica lesson. However, it may be that personality charac-teristics or mood affects participant choice of active music therapyinterventions or receptive music therapy interventions. Futureresearch is warranted to best determine appropriate interventionsbased on situtional factors, personality, and mood. Additionally,although data were not collected on this aspect, the researchershoped that the harmonica participants would continue to play theirnew instrument using the provided CD as a way to facilitate cop-ing and alter mood. Finally, the third author has had non-researchrelated clinical success using harmonica lessons with forensic psy-chiatric clients. It seems that the harmonica may be an accessible,popular, expressive, engaging, and inexpensive treatment for avariety of clinical populations. Future research on music therapyharmonica-based applications is warranted.

There are a plethora of potential reasons for the lack of partici-pants who chose the harmonica option in this study. Some patientsdeclined music therapy altogether, citing reasons such as sleepi-ness, too much pain, diarrhea, headaches, or simply not beingin “the right mood.” As playing the harmonica requires physicalactivity, including arm and hand movement and breath support, apatient who is already tired, medicated, or in pain, may not havethe desire to exert any additional energy. Another possible reasonfor declining the harmonica lesson is that patients might not wishto draw any additional attention to themselves. Playing a musi-cal instrument in a public place (such as a medical hospital) mightpotentially draw attention from other patients, hospital staff, andvisitors. Someone experiencing a low sense of control, in addition toother discomforts of hospitalization, may wish to avoid this type ofexcess recognition. Another possibility is that patients may havefelt musicially unskilled and, despite encouragement from theirmusic therapist, declined in order to avoid potential failure.

In addition to quantitative data, the researchers also collectedparticipant comments at the conclusion of each experimentalmusic therapy session. Due to the small sample size, generaliza-tions regarding these comments are premature and inappropriatebut comments were positive and tended to support music therapy.Perhaps future researchers could utilize a mixed-method design toobtain a more holistic and triangulated sense of how music therapymight effect patients on a solid organ transplant unit. Qualitativedata concerning participants’ music therapy experiences could beused to determine and empirically study potential mechanisms ofchange.

Implications for clinical practice consist of including as muchautomony to medical music therapy sessions as possible. In the cur-rent study, the researchers gave participants a choice concerningactive or receptive music therapy and the choice itself – regard-less of specific decision – may have been a key contributor toimproving dependent measures for experimental conditions. Formedical music therapists using active music engagement, harmon-icas might be an attractive option as they are small, inexpensive,and, given proper teaching techniques, relatively easy to play.

Limitations of the study include the single-session researchdesign. Perhaps increasing the music therapy dose would improveresults. However, single-session treatment is common in medicalmusic therapy sessions (Boyle, 2008; Miller, 2008), so one mightcontest that this study accurately represented contemporary clini-cal practice. Regardless, future research should include a follow-upmeasure to determine if treatment gains were maintained. How-ever, due to the use of a wait-list control group, follow-up measuresusing the true experimental design were not feasible in the caseof the current study. Another limitation is the use of Likert-Type scales to measure treatment effects. Though the researchersutilized Likert-Type scales to emphasize treatment over data col-lection within the time constraints of a single-session and remainconsistent with existing nursing assessments, Likert-Type scales

228 I. Crawford et al. / The Arts in Psychotherapy 40 (2013) 224– 229

limit external validity. Another limitation is that the researchersdid not collect data on the number of patients who declined orrefused treatment. The researchers estimate that a third of potentialparticipants declined treatment and study participation. Thoughthis approximated percentage is similar to Ghetti (2011), thesedata are becoming increasingly important to determine if inter-ventions are feasible. Finally, participants’ responses may be biasedto the researchers’ dual roles as clinicians and researchers. Futureresearchers should consider utilizing research assistants to providetreatment or measure treatment effects in an attempt to reduce oreliminate response biases.

Suggestions for future systematic inquiry include the useof established psychometric instruments instead of Likert-TypeScales, increased doses of music therapy, recording the numberof patients who decline music therapy and research participation,and determining if patients continued to play their harmonicasand found this to be an effective coping skill. Future researchersshould utilize larger sample sizes and consider multi-site studiesto increase external validity. Researchers might also examine medi-cation usage – it may be that with increased doses of music therapy,less medication is utilized to manage side effects. As decreasedmedication usage is typically a central goal for medical facilities,music therapy to promote coping, manage negative symptoms, andreduce medication consumption might be an avenue for generatingadditional medical positions.

This study was designed in order to offer various choices withina music therapy session to all patients on the solid organ trans-plant unit. Moreover, the use of a wait-list control condition, even inthe single-session design, enabled all participants to receive musictherapy treatment. This inclusive effectiveness design was ethi-cally acceptable to the researchers and the unit staff supported thisdecision. Denying music therapy services to patients was not anacceptable option for the researchers. Additionally, hospital staffsupported the researchers’ presence on the unit. This support mayhave been a result of the inclusive wait-list design allowing allpatients access to music therapy treatment and utilizing a briefinstrument to measure potential changes.

The purpose of this study was to determine the effects ofmusic therapy interventions on stress levels, relaxation, mood, andperception of side effects in hospitalized solid organ transplantpatients utilizing a randomized effectiveness design. There weresignificant between-group differences in posttest relaxation,stress, and mood scores. From the results of this study and relatedmedical music therapy research, it seems that PPLM can be aneffective psychosocial intervention for solid organ transplantpatients. In today’s evidence-based era, it is becoming increasinglyimportant to provide quantitative data supporting psychosocialtreatments. Thus, future systematic inquiry is warranted in anattempt to provide music therapy services to greater numbers ofmedical patients.

Appendix A. Harmonica lesson protocol

Materials:Two 10-hole diatonic harmonicas in the key of C major, 1 guitar

in standard tuning, 1 CD containing pre-recorded blues accompa-niment tracks in the key of G.Procedure:1. Holding the harmonica

• Form a C-shape with your left hand, with your fingernailsfacing the ceiling and your thumbnail facing the floor. Withthe holes of the harmonica facing you, hold the harmonicabetween the index finger and thumb. All four fingers shouldbe held together and parallel forming a flat surface fromleft to right (fingers should not be covering the back of the

harmonica). The numbers 1–10 should be visible on the topof the harmonica.

• Cup your right hand as if you are holding a small amountof water in the middle of your palm. The fingers should betogether. The right hand is then cupped over the back of theharmonica, held vertically, so that the fingers of your righthand rest on top of the fingers of your left hand.

• Wet the lips lightly and pucker over holes, so that sliding backand forth across the harmonica is easy.

2. Playing the harmonica• There are two sets of reeds inside the harmonica. In order to

make them vibrate you have two options: (1) blow (exhale)or (2) draw (inhale). Blowing on a C harmonica producesa C major chord. Drawing produces a G dominant 7 chord.Demonstrate this and have the student mimic your demon-stration. Have the student take a minute to play around,experiment, and get a feel for the instrument, while blowingand drawing across all 10 holes.

• Vibrato – There are several ways to perform vibrato on theharmonica. The simplest is hand vibrato. While either blowingor drawing air, open and close your right hand rapidly. To dothis, move your fingers while keeping the base of your palmanchored against your left hand. This should produce a slightvibrato.

• Articulation – Demonstrate the following basic articulations:# Draw notes: say the word “ta” while breathing inward for an

accented note. Use the word “ha” while breathing inward foran unaccented note. Be sure the student does not vocalizethese words while playing.

# Blow notes: say the word “too” while breathing outward foran accented note. Say the word “hoo” while breathing out-ward for an unaccented note. Again, be sure that the studentdoes not vocalize these words while playing.

• Tone quality – Tongue and lip shape play an important rolein tone quality on the harmonica. There are clean tones, dirtytones, fluttery tones, and many more.# Saying “wah-wah” (without vocalizing) into the harmonica

produces a speech-like sound.# Flutter-tongue produces an interesting oscillating effect.

Blow air across the roof of your mouth and tongue in orderto get the tongue to vibrate. It is the same principle of rollingR’s in Spanish.

# The shape of your tongue can also alter the tone quality.Voice, but do not vocalize, “eeh-oh” or “eeh-ah” while blow-ing or drawing through the harmonica.

3. Ready to jam!• Explain enthusiastically that there are no wrong notes on

the harmonica, and that you’re going to prove it rightnow.

• Ask the student to begin by drawing on holes 1 and 2 simulta-neously, and then to blow air through the same holes. Create avamp with the following rhythm on holes 1 and 2: two eighthnotes on draw followed by two eighth notes on blow. Use a“ta” articulation.

• Explain that you are going to take turns playing solos whilethe other person plays the vamp. Note: this is a great time toexperiment with different articulations, vibrato, and tongueshapes!

4. The Blues• Explain that you are going to play through a 12-bar blues

chord progression one time to give the student a feel for theform. Invite him or her to play along. Note: When playing theblues, the accompanist should play a perfect fifth above thekey of the harmonica. For instance, if the student is playinga harmonica in the key of C, the accompanist should play ablues progression in the key of G.

I. Crawford et al. / The Arts in Psychotherapy 40 (2013) 224– 229 229

• Teach the following rhythm with swung eighth notes, beingsure to count the rests clearly:

• Ask the student to play this rhythm four times on draw, thentwice on blow, then twice more on draw. It does not matterwhich holes are played. This will help the student follow thefirst eight bars of the blues progression. Explain that anythingcan be played during the last four bars (9–12). Cue the stu-dent to return to the original rhythm when the progressionrepeats.

• If the previous rhythm is played successfully, you may moveone step further and ask the student to play a few extra notesin each measure after playing the initial eighth-note motive.

5. Well done!• Close the session with a song in a non-blues style. This intro-

duces the student to non-blues harmonica playing. Because aC major harmonica can be played in the key of C major or Aminor, choose a song in one of these keys, or transpose a songinto one of these keys.

• Give the student a copy of a CD containing pre-recordedblues accompaniment tracks. The tracks can represent varyingstyles and tempos and be placed in order from slow/solemn toupbeat/energetic. The student may keep the harmonica freeof charge and use the CD while in the hospital or at home.The CD demonstrates the iso-principle, where the student canplay along with the track that best represents his or her cur-rent mood and continue from there in either direction towardthe track that represents his or her desired mood.

Other tips:• Use cold air to produce a loud/aggressive sound. Use warm air

to produce a quiet/gentle sound.• To produce clear single notes, tighten the corners of your mouth

and pull them inward.• If hand vibrato is too difficult: use your tongue to make ‘ah’ and

‘eeh’ sounds. Alternate rapidly to produce vibrato. If this is diffi-cult, tap slow quarter notes for the student and have him or herchange between ‘ah’ and ‘eeh’ once every beat. Next, have themchange twice per beat, then four times per beat, and finally eighttimes per beat.

• If the student is following along easily and wants to be chal-lenged, you can teach more rhythms. If the student reads music,write out a list of five or six rhythms, referring to the list as a‘grab–bag of rhythms’ to be drawn upon during solos.Song suggestions for non-blues harmonica soloing:

• Heart of Gold by Neil Young• Piano Man by Billy Joel• Knockin’ on Heaven’s Door by Bob Dylan• Love Me Do by The Beatles• Try For the Sun by Donovan• You Are My Sunshine

References

Bohachick, P., Taylor, M. V., Sereika, S., Reeder, S., & Anton, B. B. (2002). Social support,personal control, and psychosocial recovery following heart transplantation.Clinical Nursing Research, 11, 34–51.

Boyle, M. E. (2008). Capturing unique and important data within the brief medical musictherapy service model. Paper presented at the 34th Conference of the CanadianAssociation for Music Therapy, Quebec City, Canada.

Burns, D. S., Sledge, R. B., Fuller, L. A., Daggy, L. K., & Monahan, P. O. (2006). Cancerpatients’ interest and preferences for music therapy. Journal of Music Therapy,42, 185–199.

Campbell, B., & Etringer, G. (1999). Posttransplant quality of life issues:Depression-related noncompliance in cardiac transplant patients. Transplanta-tion Proceedings, 31(Suppl. 4A), 59S–60S.

Dew, M. A., Jacobs, C. L., Jowsey, S. G., Hanto, R., Miller, C., & Delmonico, F. L. (2007).Guidelines for the psychosocial evaluation of living unrelated kidney donors inthe United States. American Journal of Transplantation, 7, 1047–1054.

Dew, M. A., Roth, L. H., Schulberg, H. C., Simmons, R. G., Kormos, R. L., & Trzepacz,P. T. (1996). Prevalence and predictors of depression and anxiety-related disor-ders during the year after heart transplantation. General Hospital Psychiatry, 18,48S–61S.

Engle, D. (2001). Psychosocial aspects of the organ transplant experience: What hasbeen established and what we need for the future. Journal of Clinical Psychology,57, 521–549.

Ghetti, C. M. (2011). Active music engagement with emotional-approach coping toimprove well-being in liver and kidney transplant recipients. Journal of MusicTherapy, 48, 463–485.

Gross, C. R., Kreitzer, M. J., Russas, V., Treesak, C., Frazier, P. A., & Hertz, M. I. (2004).Mindfulness meditation to reduce symptoms after organ transplant: A pilotstudy. Advances in Mind-Body Medicine, 20, 20–29.

Lachman, M. E., & Weaver, S. L. (1998). The sense of control as a moderator ofsocial class differences in health and well-being. Journal of Personality and SocialPsychology, 74, 763–773.

Madson, A. T., & Silverman, M. J. (2010). The effect of music therapy on relaxation,anxiety, pain perception, and nausea in adult solid organ transplant patients.Journal of Music Therapy, 47, 220–232.

Miller, K. (2008). Treatment planning and documentation in single-session and high-turnover situations. Paper presented at the 12th World Congress of MusicTherapy, Buenos Aires, Argentina.

National Kidney Foundation. (2011). Retrieved from: http://www.kidney.org/news/tgames2010/PressRoom/PressKit/FactSheets/FactSheet1.cfm

O’Callaghan, C., & Colegrove, V. (1998). Effect of the music therapy introductionwhen engaging hospitalized cancer patients. Music Therapy Perspectives, 16, 67–74.

Patrick, B. C., Skinner, E. A., & Connell, J. P. (1993). What motivates children’sbehavior and emotion? The joint effects of perceived control and autonomyin the academic domain. Journal of Personality and Social Psychology, 65, 781–791.

Penninx, B. W. J. H., Guralnik, J. M., Ferrucci, L., Simonsick, E. M., Deeg, D. J.H., & Wallace, R. B. (1998). Depressive symptoms and physical decline incommunity-dwelling older persons. Journal of the American Medical Association,279, 1720–1726. http://dx.doi.org/10.1001/jama.279.21.1720

Robb, S. L. (2000). Designing music therapy interventions on the behavior of hospi-talized children in isolation: Developing a contextual support model in musictherapy. Journal of Music Therapy, 37, 118–146.

Robb, S. L. (2003a). Coping and chronic illness: Music therapy for children and ado-lescents with cancer. In S. L. Robb (Ed.), Music therapy in pediatric healthcare:Research and evidence-based practice (pp. 101–136). Silver Spring, MD: AmericanMusic Therapy Association.

Robb, S. L. (2003b). Designing music therapy interventions for hospitalized chil-dren and adolescents using a contextual support model of music therapy. MusicTherapy Perspectives, 21, 27–40.

Thompson, S. C. (1981). Will it hurt if I can control it? A complex answer to a simplequestion. Psychological Bulletin, 90, 89–101.

Walworth, D. (2003). The effect of preferred music genre selection versus preferredsong selection on experimentally induced anxiety levels. Journal of Music Ther-apy, 40, 2–14.