ORIGINAL RESEARCH

BRIEF MEDITATION TRAINING CAN IMPROVEPERCEIVED STRESS AND NEGATIVE MOOD

James 0. Lane, PhD;Jon li. Seskevich, R\, iss\. iiA; Car! F. Pieper,

Objectives • To test a brief, non-sectarian program of meditationtraining for effects on perceived stress and negative emotion, andto determiiie eflects of practice frequency and test the moderatingeftects of neuroticism (emotional lability) on treatment outcome.Design and Setting • The study used a single-group, open-label,pre-test post-test design conducted in the setting of a universitymedical center.

Participants • Healthy adults (N=2nO) interested in learningmeditation for stress-reduction were enrolled. One hundred thir-ty-three (76% females) completed at least 1 follow-up visit andwere included in data analyses.Intervention • Participants learned a simple mantra-based med-itation technique in 4, 1-hour small-group meetings, withinstructions to practice for 15-20 minutes twice daily.Instruction was based on a psychophysiological model of medi-tation practice and Its expected effects on stress.Outcome Measures • Baseline and monthly follow up measures

of Profile of Mood States; Perceived Stress Scale; State-TraitAnxiety Inventory (STAI); and Brief Symptom Inventory (BSI).Practice frequency was indexed by monthly retrospective rat-ings. Neuroticism was evaluated as a potential moderator oftreatment effects.

Results • All 4 outcome measures improved significantly afterinstruction, with reductions from baseline that ranged from 14%(STAI) to .36% (BSI). More frequent practice was associated withbetter outcome. Higher ba.seline neuroticism scores were associ-ated with greater improvement.Conclusions • Preliminary evidence suggests that even briefinstruction in a simple meditation technique can improve nega-tive mood and perceived stress in healthy adults, which couldyield long-term health benefits. Frequency of practice does affectoutcome. Those most likely to experience negative emotionsmay benefit the most from the intervention. (Altern Ther Health

James D. Lane, PhD, is a professor of medical psycholog)' inthe Department of Psychiatry at Duke I'niversity MedicalCenter in Durham, NC. Jon E. Seskevich, KN, BSN, BA, is anurse clinician in the Department of Advanced PracticeNursing at Duke University Hospital, Durham, and Carl F.Pieper, DrPH, is an assistant research professor in theBiostatistics and Bioinformatics Department at DukeUniversity Medical Center.

Meditation training has become a popular inter-vention for the prevention and treatment ofstress-related diseases and for the manage-ment of stress that often accompanies seriousmedical conditions. Published research stud-

ies describe the application of meditation training as primary oradjunctive treatment for high blood pressure and other coronarydisease risk factors,'' chronic pain.'" and cancer,'" as well as forstress management in high-stre.ss occupations.'"^'' The 2 mostcommonly studied meditation training programs according to ourreview of the research literature are Transcendental Meditation®(TM®; ht tp: / / tm.org) and the Mindfulness-Based StressReduction (MBSR) program of the Center for Mindflilness at the

University of Massachusetts Medical School. North Worcester,Mass (http://wwuMimassmed.edti/cfhi/srp/index.aspx). These 2proprietary programs have been distributed widely and are avail-able in many locations throughout the United States and othercountries: however, the TM and MBSR programs share severalcharacteristics that may limit their use in the general population.

Both programs are based on traditional practices of Asianreligions. The TM technique has its origins in the religious prac-tices of India and includes a formal initiation conducted inSanskrit. Initiates receive secret Sanskrit mantras, traditionallyused to invoke Hindu deities, to use as the focus of meditationpractice. MBSR is based on traditional Buddhist meditation prac-tices and incorporates concepts and material from Buddhist philo-sophical writings in tbe class presentations.'" Such religiousfoundations may make these programs unacceptable to individu-als who practice other religions or no religion and may preventtheir adoption by publicly funded programs and public schools. Inaddition, both TM and MBSR programs require considerablecommitments of time and money to complete training. The TMprogram requires 10 hours of lectures and instruction in 7 sepa-rate steps.'' The standard MBSR program requires about 30 hoursto complete, with 8 weekly classes and a full-day retreat. Theseprograms typically cost several hundred to several thousand dol-

38 ALTERNATIVE THERAPIES. lAN/EEB 2007, VOL. 13. NO. 1 IVaining for Stress and Negative Mood

lars for initial training. Clearly, the time requirements and expensewould be barriers for many individuals, and the high cost alonecould limit their use in many situations.

These limitations can be overcome. Meditation training pro-grams have been developed without specific religious associations.Furthermore, there is no empirical evidence that lengthy instruc-tion is necessary for effective meditation training, hideed, severalindividuals have suggested that non-sectarian forms of meditationcan be taught in 1 or a few brief meetings, or even can be learnedfrom a book.'""' These simpler programs have been shown to beeffective treatments for stress and stress-related disease.'"'"

The authors (JDI. and JES) developed a simple, brief, non-sec-tarian meditation training program for use in a clinical trial ofmeditation and relaxation as non-pharmacological treatments forhigh blood pressure. This development effort sought to create a"generic" alternative to the popular proprietary systems, one thathad the potential for time- and cost-effective instruction. Duringenrollment for the hypertension treatment study, a second sampleof 200 non-hypertensive adults was enrolled in an "open-label"study of meditation training for stress reduction. This study, calledthe "Calm Your Stress Study." was designed to provide preliminaryevidence ofthe effectiveness ofthe training program for the reduc-tion of subjective stress and negative emotion in a general popula-tion of self-referred adults who were interested in learningmeditation for stress management. The study had '.i objectives: toevaluate the potential changes in measures of perceived stress andnegative emotion after participants began to practice this genericmeditation technique: to determine whether the frequency of med-itation practice affected outcome during follow-up; and to testwhether an indi\iduars initial level of neuroticism could explainvariability in outcome following meditation training. The latter 2objectives were intended to open new lines of meditation research.Current recommendations for practice frequency and durationhave little or no evidence base without empirical data linking prac-tice to outcome. The current literature offers no explanation forwhy meditation training yields different outcomes among individ-uals, although the identification of characteristics related to treat-ment success could be used to target the individuals most likely tobenefit from instruction. The re.sults of this pre-experimentalstudy were intended to inform the design of additional controlledtrials if they were warranted by initial results.

METHODSStudy Design

The Calm Your Stress Study used a single-group, pretest-posttest design'' that included a pre-intervention baseline assess-ment and 3 follow-up assessments at monthly intervals followingthe start of meditation practice. Although this pre-experimentaldesign lacked controls for indirect effects of participation, theopen-label approach simulated the effects of training in naturalis-tic settings in a large and varied population. The planned assess-ments ofthe effect.s of practice frequency and tests of specifichypotheses related to the moderation of outcome effects by per-sonality and affect variables strengthened the overall design. To

control subject-experimenter artifacts and biases, instruction anddata collection were conducted by different individuals, in differ-ent locations, and at different times.

ParticipantsA sample of 200 adult men and women was recruited from

among employees, students, patients, and visitors of DukeUniversity and Duke University Medical Center through campusnewsletters, brochures, and handbills. Advertisements sought vol-unteers who were interested in learning a simpie meditation tech-nique for the purposes of stress reduction, lenient eligibility criteriawere used to encourage a heterogeneous sample. Participants wererequired to be at least 18 years of age and able to speak, understand,read, and write English, which was necessary for understandinginstruction and completing assessments. Potential volunteers whohad active psychiatric disease or used psychoactive medications,who were currently participating in other formal stress manage-ment programs, or who planned to move from the study area ortravel extensively during the course of participation were excludedfrom the study. Participants who completed the study received $40as compensation for the time required for study assessments.

MeasuresFour different instruments assessed subjective stress and neg-

ative mood before and after training. Although the domains ofthese instruments were expected to overlap, each provided a differ-ent perspective on the concept of subjective stress. When Instru-ments contained multiple scales, the overall summary score wasused for hypothesis testing, to control the type 1 error that couldarise from multiple independent tests.

General negative mood was assessed using the brief version ofthe Profile of Mood States (POMS; Multi-I lealth Systems Inc, NorthTonawanda, NY), which contains 30 adjectives that describe feelings(eg, friendly, tense, grouchy, angry). Participants rated each item ona 5-point scale to describe their moods during the past week, includ-ing the current day. Scores were calculated for 6 different moods oraflective states: Tension-Anxiety, Depression-Dejection, Anger-Hostility, Vigor-Activity, Fatigue-Inertia, and Confusion-Bewilderment. Total Mood Disturbance (TMD), the sum of thesescales (with Vigor-Activity subtracted), was used as the summaryscore for general negative mood. The POMS has good internal con-sistency and reliability and has been widely used in re.search.-*-'

The Perceived Stress Scale (PSS) was used lo provide a globalmeasure of perceived stress in daily life. The PSS is a 14-iteminstrument designed to measure the degree to which common sit-uations are appraised as stressful.^' The items ask about feelingsand thoughts during the past month and how often the respon-dent felt a certain way in a specific situation. Responses range from"never" to "very often" on a 5-point scale. The PSS has adequatereliability and validity."'

Anxiety was measured with the State-Trait Anxiety Inventory(STAI: CPP, Inc, Mountain View, Calif), a widely used measurewith well-established validity and reliability,-* The state version(STAI-S) was used to assess changes in general anxiet)' over time.

Meditation Training Ibr Stress and Negative .Mood ALTERNATIVE THERAI'IES, )AN/FEB 2007, VOL. 13, NO.

This inslrument consists of 20 items containing words descriptiveof anxiety that are rated on a 4-point scale and asks participants torate their feelings at that moment.

The Brief Symptom Inventory (BSI; National ComputerSystems, Inc, Minneapolis, Minn),"' derived from the SCI.-9()-R,""assessed symptoms of psychological distress. The BSI is a r)3-itemscale of self-reported psychological symptoms experienced duringthe previous 7 days. Nine symptom dimensions are measuredusing items rated on a 5-point scale from "not at all'" to "extreme-ly." The General Severity Index (GSI) is a weighted score based onall items that combines information on the number of symptomsexperienced and their severity. The reliability and validity oftheBSI are well-documented.

Study participants completed retrospective assessments ofmeditation practice frequency at each follow-up visit. Frequency ofpractice was reported using a 7-point category scale that includedthe following choices: "twice a day every day"; "twice a day mostdays"; "at least once a day every day"; "once a day every day"; "oncea day most days"; "several times a week"; and "never."

Neuroticism was assessed using the Revised NEO PersonalityInventory (NEO PI-R; Psychological Assessment Resources, Inc,Lutz, I'la),- a personality inventory that operationalizes the 5-factormodel of normal personality. The NEO PI-R contains 240 descrip-tive statements that are completed by self-ratings on a 5-point scaleand typically takes 20 to 30 minutes to complete. Validity and relia-hilit\' have been established. The neuroticism (N) domain contrastsemotional stability with lability. Scores reflect the general tendencyto experience negative emotions such as "fear, sadness, embarrass-ment, anger, guilt, and disgust."* '"''''" The N domain includes sub-scores or facets for specifk emotions comprising anxiety, angryhostilit}', depression, self-consciousness, impulsiveness, and vulner-ability to stress, which were examined in exploratory analyses.

InterventionThe meditation technique was similar to techniques

described by Benson,'" Carrington,''' and Wilson," " Participantsselected for themselves a sound, word, or brief phrase ("mantra")to be used as the focus of meditation. They sat with eyes closed andbegan practice with a brief (less than 1 minute) period of relaxedabdominal respiration, then repeated their mantra silently andcontinuously for 15 to 20 minutes. Mantra repetition was notlinked to the breath. Participants were instructed to allow themantra to repeat at its own natural pace and simply to return tothe mantra repetition whenever they noticed that other thoughts,sensations, or feelings had come into consciousness and distractedattention from the repetition ofthe mantra. Each practice endedwith 1 to 2 minutes of quiet rest before other activities wereresumed. The emphasis ofthe instruction was effortless repetitionofthe mantra and recognition and control of intrusive thoughts.Participants were instructed that twice-daily practice for the 3months ofthe study would provide them with a full experience ofthe potential ofthe technique (cf Wilson""*).

Participants learned the meditation technique in the first of4, 1-hoiir meetings and began twice-daily practice the following

day. The remaining classes served to reinforce the instructionsand provide information and guidance to support the develop-ment of a regular practice regimen. Questions were answered,and common problems were presented and discussed.Meditation was presented as a cognitive exerci.se that enhancedthe experience ofthe quiet mind and taught recognition and con-trol of intrusive thoughts. Participants also were given a psy-chophysiological rationale for the stress-reducing effects ofmeditation. At the final meeting, the iEistructor addressed com-mon problems that arise during independent practice and dis-cussed the benefits and experiences that have been reported byothers who continued regular meditation practice.

After the final meeting, study participants continued to prac-tice on their own. They were told to contact instructors if they hadquestions or concerns about the technique or their practice. Nofurther formal contact was maintained during follow-up, however.

Procedures

Interested volunteers Lnade initial contact and completed pre-liminary screening by telephone, then scheduled an appointmentfor completion of informed consent procedures. Participants pro-vided detailed background and medical history information, andcompleted the NEO PI-R questionnaire.

The outcome instruments (POMS, PSS, STAI. and BSI)were completed at a baseline visit before training, and partici-pants then were scheduled for meditation training. Participantswere taught in small groups. The 4,1-hour group meetings wereheld at lunchtime or after work on a Monday/Wednesday orTuesday/Thursday schedule over 2 weeks. When participantsmissed a meeting, a make-up session was scheduled, if possible.Participants who attended at least 3 ofthe 4 meetings were con-sidered to have completed training.

Follow-up visits were scheduled 4, 8, and 12 weeks after thedate ofthe first class. Participants returned to the laboratoryand completed the 4 outcome instruments. Although partici-pants were encouraged to call investigators with questions orconcerns about practice, such contacts were rare. Study person-nel did not contact participants except to schedule the follow-upvisits. Study participants were encouraged to complete the fol-low-up visits, even if they had stopped meditating. Repeated e-mail and telephone contacts were attempted as necessary tocomplete follow-up. Individuals who refused to reply to repeat-ed contacts or who asked not to be contacted again were consid-ered "dropouts" from the study. The study concluded with the12-week follow-up visit.

Data AnalysisStandard scoring procedures were used for ali self-report

instruments. Summary scores were used for hypothesis testing,when instruments contained sub-scales, tu reduce the number ofstatistical tests and control type 1 error; however, sub-scale scoreswere used for exploratory purposes. Hypothesis tests were con-ducted using procedures for mixed-models analysis of varianceor covariance to accommodate the repeated measurements over

40 ALTERNATIVE THERATIES, 2007, VOL. L5. NCI. I Medilalion Trainint; for Slress and Negative

time. Mixed models (Proc MIXED. SAS v 8; SAS Institute. Inc,Cary, NC) allowed the inclusion of all available data and did notdelete participants with incomplete data. Mixed models also per-mitted iiiclnsion of time-varying covariates and between-subjectvariables when appropriate. Given the single-group design,hypothesis tests focused on the main effects of time (baselineand follow-up) and interactions ofthe time factor with between-subjects factors. When main eftects and interactions were signifi-cant, contrasts of baseline and each ofthe '.i follow-up visits wereconducted lo locate changes temporally. Comparisons amongfoilovv-up time points also were conducted to explore the stabili-ty of changes over time. The criterion /'<.O5 was used to declarestatistical significance tor ali tests.

RESULTS

Participant Disposition and CharacteristicsTwo hundred people expressed interest in participating in

the study and scheduled appointments for orientatitin, informedconsent, and baseline data collection. The disposition of theseparticipants is shown in detail in Figure 1. Baseline data wereavailable for 192 participants because 4 participants failed tomeet study inclusion criteria, and 4 withdrew, requestingremoval of their study data. Thirty participants who completedbaseline never arranged to take the classes despite repeated con-tacts. Ofthe 144 participants who completed training, 133 com-pleted at least 1 follow-up visit. The average age (+ SD) of thissample was 38.0 ± 13.S years. The group was 76% female.Seventy-four percent of participants were white, 17% wereAfrican-American, and thf remainder were Asian, Hispanic, orNative American. Most (73%) were currently employed. 9% wereretired, and the remainder included unemployed individuals andstudents. Half of the group had 12 to 16 years of education, andthe other half had more than 16 years.

Effects of Meditation TrainingMeditation training led to highly significant reductions

Enr

ollm

ent

Inte

rven

tion

Foll

ow-u

pna

lysi

Screened for eligibility (N = 200)Excluded {ii = 8)

Failed screening (n = 4)

Driipped out. requestingremoval ot all data (n = 4)

Completed baseline and

scheduled for training (N = 192)Attended no meetings (n = 30)

Attended < 3 inct'tings (n = 18)Completed training (n - 144)

Completed training and

sciicduledforfolknv-upin - 144)No tblkm-up visits (n = 11)

1 follow-up visit (n = 12)2 follovv-up visits (n - 10)

3 follow-up visits (n = 111)

Data analyzed (N = 133)

FIGURE 1 Disposition of Study Participants

(/'<.OOO1) in all 4 measures of stress and negative emotion(Table 1). hnpro\'ements were noted at the first follow-up visit 4weeks after participants began meditation practice andremained stable until the fmal follow-up at 12 weeks. Theplanned contrasts of baseline vs each follow-up visit demon-strated significant reductions for all 4 outcome measures at all 3visits (all P<.0001). Post-hoc comparisons found no differencesamong the 3 follow-up visits for any measure, indicating stabili-ty over time. Post-treatment reductions were 30% of baselinelevels for POMS total mood disturbance, 23% for the PSS, 14%for STAI. and 36% for the GSI on the BSI.

Outcome Measure

Profile of Mood States

Total Mood Disturbance

Perceived Stress Scale

State Trait Anxiety inventorvState Scale

Brief Symptom InventoryGeneral Severity' Index

TABLE 1 Changes in Perceived Stress and Negative Mood Following Meditation Training

Pre-training

Baseline

24.7±1.3

28.1±fl.(i

39.6±0,8

0.70±0.03

Post-training Follow-up

4 Weeks 8 Weeks 12 Weeks

i2.2±i.s* n,()±i.(^* n,fi±i,B* F

22.0±0.7* 22,0±0,7* 2I,0±0,7* F ^

34,4±0.9'' 34.3±0.9* 33.8±0.9* F.,

0.4{i±0.03* 0.47±0.a4* 0.42±0.04* F^

*For each outcome measure, mean values differ significaiiily from the baseline mean value bul are imt diiferenl from each other (.P<.05).

Time Effect

..= 41.48,/'<.OOO1n.i

.,)= 54,77. /*<.OOO!

^.^= 1797 P<.0001

Meditation Training for Stress and Negalive Mood ALTERNATIVF TilFRAI'IFS, 2007, VOL. 13, NO, I 41

Effects of Practice on OutcomeThe relationship between practice frequency and outcome

was tested using retrospective self-reports ofthe frequency ofmeditation practice collected at follow-up visits. Inspection ofthe 364 completed reports revealed a bimodal distribution ofresponses on the 7-point categorical scale, with an extreme lowpoint at category 4 ("once a day, every day." 1.4% of responses).Given the shape ofthe distribution, the 7 categories weredichotomized into bigb (once a day or more frequently) and low(less than once a day) practice frequency categories. I'hi.s divi-sion had face validity, representing low practice frequency as lessthan daily practice. In addition, this division yielded similar pro-portions for the 2 categories, 48% high vs 52% low.

Because practice frequency could vary over time during fol-low-up, practice category was included as a time-varying covari-ate in a re-analysis of treatment outcome data. Outcome valuesduring follow-up were converted to change scores by subtractingbaseline values, a necessary step because practice frequency datawere not available at baseline. The hypothesized eflects of prac-tice frequency on outcome change were evaluated by tests ofthemain effect of practice category and the interaction of practicecategory with visit, which would indicate a changing influence ofpractice frequency during follow-up.

None of the practice category by visit interaction effects wassignificant (all P>.3). but the main eftect of practice frequencywas significant for POMS total mood disturbance, PSS, and STAI(Table 2). Practice frequency did not affect BSI CiSI change scoressignificantly (/'>.25). More frequent practice was consistentlyassociated with greater reductions in scores for perceived stressand negative emotion and an overall better outcome.Improvements in these 3 nieasure.s were more than 50% greaterwhen a participant practiced at least once a day. compared to lessfrequent practice.

Neuroticism as a Moderator of Meditation EifectsTo test the moderating effects of initial level of neuroticism

on responses to training, N scores from the NEO Pi-R w ere

TABLE 2 Average f lianges in Outcome Measures l-roin Baseline to [•ollow-up tor llie Dichotomous Categories of Reported Dailv Meditation Practice

Outcome Measure

Protile ot Mood States

Total Mood Disturbance

Perceived Stress Scale

State Trait Anxiet}'Inventory Scale

Brief Symptom InveTitor)'General Se\'eritv index

Self-reported Frequency ofMeditation Practice

One or more Less thantimes daily once daily

-15.1 ±1.73 -10,4 ±1.7

-7.7 ± 0.8 -4.9 ± 0.7

-7.0±l.l -4.ti±l.l

-0,27 ±0,04 -0,23 ±0.03

DifferencePvalue

<.O2

<.00I

<,05

>.25

included as a ti.xed covariate in re-analyses of outcome data.Moderation was assessed by the test ofthe interaction ofN andvisit, which revealed whether the trait affected the changes instress or mood score over time.

NEO PI-R N scores were distributed normally in the studysampk'. The mean score (±SD) was 89.2 ± 23.8, and the medianwas 87, with an interquartile range of 33. This mean corre-sponds roughly to the 70th percentile on adult norms for menand women," N scores did not differ significantly between thesexes by /-test (/'>.3O), The N by visit interactions were signifi-cant for all 4 outcome measures (all /'<.OOO1). These interac-tions were graphed using parameters of the regression modelsto calculate estimates over time for N scores at the 9()th. SOth,and 10th percentiles ofthe sample distribution (Figure 2), Thegraphs indicate that higher N scores were associated with higherbaseline scores for each ofthe mood and stress measures andwith greater reductions in negative mood and stress after train-ing in all 4 measures.

DISCUSSIONMeditation practices such as TM and MBSR have been

employed for stress reduction in a variety of settings, althonghthe length and cost of training and the association of these pro-grams w'nh Eastern religions may limit their broad acceptance.The results of this open-label study demonstrate that even brieftraining in a simple non-sectarian meditation practice can beassociated with improvements in subjective stress and negativeemotions in a general sample of adults interested in learningmeditation as a stress-reduction technique. The group demon-strated bigbly significant reductions in all 4 outcome measures ofperceived stress and negative emotion. For 3 ofthe 4 outcomemeasures, reductions of 20% to 40% from baseline scores wereobserved. These beneficial changes were apparent at the initialfollow-up 1 month after training began and were maintaineduntil the end ofthe study at 3 months.

'Fhe observed effects cover a broad range of negative moodsand perceptions. TMD from the POMS is a summary score com-prised by anger, confusion, depression, fatigue, and tension, withvigor scores subtracted. Post-hoc examination of these separatescales revealed significant improvements in all 6, althoughchanges were greater for the 4 negative mood scales than the 2scales that assess the physical symptoms of fatigue and vigor(data not reported). PSS scores include both the number of com-mon stressors experienced in everyday life and the negativeimpact that these stressors have. Items on the STAI assess thelevel of anxiety at the moment of administration. The GSI oftheBSI combines 9 separate symptom dimensions: somatization,obsessive-compulsive, interpersonal sensitivity, depression, an.xi-ety, hostility, phobic anxiety, paranoid ideation, and psychoti-cism. Post-hoc exploration of these dimensions found that ai!were significantly reduced after treatment, witb all but one (som-atization) dropping by about 40% frotii baseline levels. Thus,improvements following meditation training appear to be wide-spread in the domain of stress and negative emotion.

42 ALTERNATIVE THERAPIES, )AN/EEB 2007, VOL. 13, NO. I Meditation Training for Stress and Negative Mood

High - H - Med -^

NEO PI-R Neuroticism

Low

Baseline 1 mo 2 mos 3 mosFollow-up

Black Iriangles-90th percentile: gray t)(>xes=:>()tli ptTceiilile;white trianglcs=10th pertentile of sample distribution nf N scores,

FIGURE 2 Moderation of Treatment Outcome by Baseline Levelof \ E 0 PI-R Neuroticism

The importance of regular meditation practice has beenasserted despite the lack of empirical evidence. Our results con-firm that more frequent meditation practice is associated withbetter outcome. Differences between "at least once a day" prac-

tice and "less than once a day" practice were statistically signifi-cant in 3 of 4 outcome measures (POMS, PSS, and STAI), with asimilar trend in the fourth (BSI). The beneficial effects of morefrequent practice persisted through the 3 months of follow-up.Similar practice effects have been reported in several recentstudies of MBSR for stress reduction,'" ' "' although the majorityof meditation studies have neglected to test the effects of prac-tice frequency. The covariations of practice frequency and treat-uient outcoiue reported here do not establish a "dose response"effect for practice or determine a minimal level of practice thatprovides useful results because individnal levels of practice wereself-selected: however, these results are consistent with a causaleffect of practice frequency on outcome and should stimulatefurther research on the dose-response question. Certaitily, weneed to understand how practice frequency affects outcome ifwe want to develop effective and efficient training programs forthe general public. Traditional guidelines for the frequency andduration of meditation practice will be strenjjthened by empiri-cal evidence to support them.

Finally, this study demonstrated that individual character-istics tnight affect the outcome of a meditation intervention. Inthese data, participants who scored higher on the NliO Pl-Rdomain of neuroticism demonstrated greater improvements fol-lowing meditation training. Graphs ofthe interactions suggestthat the higher-N participants reported higher initial levels ofnegative emotion and perceived stress before training and sub-sequently converged toward the lower-N participants. Theseresults complement those of a recent study that found that ini-tial trait anxiety .scores were correlated with increased autonom-ic relaxation indexed by heart rate variability when participantslearned Zen meditation,'" It may not seem surprising that par-ticipants who start with higher stress scores show greaterimprovement: however, it is reassuring thai these individualsare not resistant to the potential benefits of meditation practice,especially when they appear to have the most to gain. Futureinvestigations may clarifS' the characteristics of responsive indi-viduals so that treatments can be targeted more effectively.

These results are preliminary, given the open-label natureofthe study and the lack of experimental controls for possibleindirect effects associated with participation in the training pro-gram. But the results do suggest that simple, brief, non-sectari-an programs of meditation training tnay offer benefits to adultsin the general population who are interested in stress rednction.Such programs may have advantages over more established pro-grams in many situations, especially where the length and costof training or ties to Eastern religions are barriers to participa-tion. Simpler programs could be made available at lower cost,which could be a further advantage in publicly funded settingssuch as schools or community programs. Although the estab-lished programs have offered the opportunity for many individ-uals to learn meditation and receive the benefits of regularpractice, there may be significant value in the exploration of lessintensive, non-sectarian meditation training programs that canbring these results to a larger population.

Meditation Training lor Stress and Negative Mood ALTEIINATIVH THhRAPILS. )AN/FEB 2007. VOL 13, NO. I 43

AcknowletlgmentsThe auilii>rs ivisii Io aikiiiiw!nlne Tara P, (.'laylon and Mtlivsa Iliirtilc tiir their assislamt- in ilu'

coliettioii and iiiaiia^cnifni (il study daia. Tins siudy "'as Mipjiorlcd m parl bj a grant truni the

National Heart, l.ung. and Blood Inslirutc thai was im:iriii*d lo the lirsl aulhor.

REFERENCKS1. Sihiifider Rll, Staggers F, Alexander ( X , cl al. A randomistil lonirolled trial ul

slress reductiiin tor hyperleiisioii in older Alrii-aii Americans. llypiTli'imoii.1993:26(5):820-827.

2. Zaniarra )W, Si'hiieider Rl I, Besse^hiiii I, Uobinsoii DK, Saltrni) IW. llM'tiilnt.ss of lhetranscendental meditation program in the treatment nt patients with coro]iar\ arterydisease./im/Giri/iW. 1996:77(10):8ti7-870.

3. Casiilki-Hichmoiid A, Schneider KH. Alexander CN. et al. T.Hects nl stress rednttion oncarotid atherosilerosis in liy)iertensive Alricaii Americans. Stwki'. 20tXI;:il(!S):,'iBR-,'J7.i.

4. Kahat-Zinn J. An iintpatii'nt program in behavioral iiiedkint' tbr chroiiir pain patientsba^ed un the practice ol tnindfulness meditation: theoretical considerations and prelimi-nary ri-sults. Cai Hmpl'iy,hmln: l!IS2;4[l):33-47.

5. Kabat-Zinn j , l.ipvvorth I,, Bnrney K. ITif clinical use uf tnin dful ness medilation tiir lhe.self-rcgulati<in 1)1 chronic paiii./BWwi',VIiv/. lH8,^):8(2):lti:i-im

6. Kaplan KH,(iiil(iinb('rgDI.,t;alvin-NadeauM. The impatt lit a meditatiiin-liased stressreduction program on tibromyalgia. Gen Iliup Pm'hi'itry. 19HJ{;iri(r)):2H4-28;i.

7. Speca M, Carlson LE, Cioodey E, Anigeii M. A randomized, wait-list TOntriilled clinicaltrial: the effed ot a niiiidfulne.ss n]editati(in-hascd stress ri'duttion program DII nitiodand sytnploms ot stress in cancer otitpatletits, l'w'li(iso,ii Mill. 20On:62(5):613-622.

8. Carbon LE, I'rsuliak Z, Goodey E, ,\ngen M, Speca M. The etlkts iit a miridtiilness nied-iladnn-baseil stress reduction program on mood and symptoms of stre.ss in cancer oiii-patienls.ti-inunth follow-up. Swppir/<.Virft'o"((r.2IXIl,9(2):112-123.

9. Carlson LE, Speca M, Patel KD, Coodey R. .Mindtlilness-based stress reduction in rela-tion to quality of life, moDd. symptoms of stress, and iminune parameters in breast andprostate cancer outpatients. ftviAu.w/H Mci/, 2OO. :Uri(4):.'i71-81,

10. Shapiro Si., Boot/in RR, Hjjueredo A], bjptu AM. Schwartz CE. The etliiaryiif iiiindhil-nes.s-ba.sed stres.s reductiiin in the treatment of .sleep disturbance in women ivith breastcancer: an exploratory study./fti'(7wsi))nfe\.2OUJi;.'i4(l):8.')-Hl,

11. Carlson LK. Speca M, Patel K\\ (uwidey T:. Mindfulness-based stress reduction in rela-tion to quality oflife, mood, symptoms of stress and levels of tortisoi, dehy-druepianilro.sterone stiltatf (1)II[;AS) and melatonin in hreasl and prostate caneeroutpatients. ftv</i»"('Kn«'iii/(NrfH<'/i)i5'.2t)(M:2!)(41:448-474.

12. Btddoi" AE. Mur])hy SO, Does iiiirnifulness de< rease sirens and foster empatliy amotignursinastiiiienls?/,VH)-i/.V/w.2(H)4;4:i(7)::!{ir>-:!t2.

Koth B, Hobbiiis I). .Miiidfuliiess-hased slress reduclioii and bealth-related quality oflife- tlndings from a bilingual inner-city patient population, I'Wfkawm Med. 2004:

kixsenzweig S, Reibel DK, (ireeson IM, BrainardGC, Jloiat M. Mindfulness-hased stressreduction lowers psycliological distress in tiiediijil sindenls. Tci,il> laini Mvil.

Shapiro SI., Schwartz GE, Bonner G. Etiects otniindfulness-based stress reduction onmeilical andprenicdiial students../ftVwi',Wrt/. t9H8:21(fi):,''.a]-.''i9!l.Kabat-Zitni I. WhernvT You Gu. Thin liiu .Inv Minilpilww MnliliiUnn in Evirvday Liji:New York: Hyperion Books: Iy94.Thr TrauMtndi'Tital Mt'ditatiiin I'ro^rani. Vailable at: http:/ / tni .org/. .\ici-ssi'dNoveniher29,200aUensoii II. Tlii' KeUixiiliim KnpDtisi. Ne«\ork; Morrow: t97:"i.Carrington P. Tlw Hxak nf McdiliUioir The Omph-h' b'linle W Mmlfni Mnliliihim. Bostiiii,Mass: Element Books; 1998.Wilsiin P. The Culm Tixhiuijtw: Medili,l!iiii Wilhuul Mi,^ie o' \iy\tiii\m. Neu "l'lirk: Barnes& Noble Books: 1999.( anijibell DT, Stanley JC. lixpcrinieiiUi} ami Qiiiiii-l'.xperiim'nliil /")(.\H,'/M fi'r Ren'itrdi.Boston, Mass: I loiighton Mitflin; 19lj:i.Mi'Nair DM, Uirr M, Droppleman I.E. EdiTS Manual for the Protile ot Mond .States.San t^iego, (.a Iif: F.cjucational and Industrial Testing Servite; 1992.Coiien ,S, Kamarck T, ,Mermelstein R. A global measure ot perceivcil stress. / lleitlth Sm

Spielberger CH. State-Trait Anxiety Inventory (Fiirni V). Mi-ido I'arW, C alii: MindGaideiL,ln(.:lf)H:i.Derogatis LK, Spenier fM. The Hthj SyiiipUm liivinhny (HSl)- \,l,uiiinlr,ilhi,, ami

I'roi nlure\ M(,,imil-i. Baltimore, Md: Clinical Psychometric Research; 19K2.Detogatis LK, ,SCL-9(I-H: Ail'tiinislnilim. Sdviiig. and I'rorrdiire Mimunl il. Towsoii, Md:Clinical Psyihometnts Kosearih: 198.1Derogatis LR. Melisaratiis N. The Briet Sym|)ioiii Inventory. ;in ititrod\ictory rejiorl.

Costa IT, jr, McCrae RR. Revised NEO Persiwality Inmilm tNEO Pi- li) and NliO Fiir-l-tiilor Imvniiny (i\'l:(>-i'iT) Pnije\sii>nal Miinmil. Odessa, Fla: Psycholo^cal AssessmentRe.sources, Inc; 1992.

Gross Ctt, Kreit7er Ml, Russas V, Treesak C, Erazier PA, Hertz MI. .Mindluhifss medita-tion to reduce symptoms after organ traiiS]ilant: a pdot study. .Adv XUiul lUidy Mrd.2004:20(2):20-29.MurataT, Takahashi T, Hamada T. et al. Individual trait anxiety levels characterizing theprojitrties of ?t'iimt'dilaiioii.,\'ruriyny(Ari/Mri/iii^. 2004:50(2}: 189-194.

5th Annual Conference

Integrative Medicine forHealthcare Organizations

This one-of-a-kind CAM event is designedparticularly for:

lcal and nursing directors;»es and clinical practitioners

onfirmed Faculty: yne Jonas, MD,lueli Institute for

•; Jennifer James, FhD,cultural anthropologist and author; and more

American HospitalAssociation

rma

althforum.comlick on Complementary and

Aitemative Medicine

707-644-1181sananth@healthforum.com

derwrjter

SAMUELl INSTITUTE