Thought Field Therapy (TFT) as a Treatment for Anxiety Symptoms: A Randomized Controlled Trial

ORIGINAL RESEARCH

THOUGHT FIELD THERAPY (TFT) AS A TREATMENT FOR ANXIETY SYMPTOMS: A RANDOMIZED CONTROLLED TRIAL Audun Irgens, MD,1,# Toril Dammen, MD, PhD,2,3 Tor Erik Nysæter, MSc,4 and Asle Hoffart, PhD5,6

Objective: To investigate whether thought field therapy (TFT) has an impact on anxiety symptoms in patients with a variety of anxiety disorders. Design: Forty-five patients were randomized to either TFT (n ⫽ 23) or a waiting list (n ⫽ 22) condition. The wait-list group was reassessed and compared with the TFT group two and a half months after the initial evaluation. After the reassessment, the wait-list patients received treatment with TFT. All 45 patients were followed up one to two weeks after TFT treatment, as well as at three and 12 months after treatment. Subjects: Patients with an anxiety disorder, mostly outpatients. Intervention: TFT aims to influence the body’s bioenergy field by tapping on specific points along energy meridians, thereby relieving anxiety and other symptoms. Outcome Measures: Symptom Checklist 90-Revised, Hospital Anxiety and Depression Scale, the Sheehan Disability Scale.

INTRODUCTION Thought field therapy (TFT) is used as a therapeutic tool by some psychiatric and many complementary therapists in Scandinavia as a treatment to relieve anxiety and other symptoms. Roger Callahan, the creator of TFT, writes in his book Tapping the Healer Within (p. 4), that “A key to the treatment is influencing the body’s bioenergy field by tapping with your fingers on specific points of the body located along energy meridians.”1 This theory has not been empirically validated. Andrade and Feinstein have published a large, preliminary study from 11 centers in South America in which they compared TFT to cognitive therapy combined with medication.2 Although

1 DPS Aust-Agder, Sorlandet sykehus, Arendal, Norway 2 Institute of Basic Medical Sciences, Department of Behavioral Sciences in Medicine, Faculty of Medicine, University of Oslo, Oslo 3 Department of Psychiatry, Oslo University Hospital Ullevaal, Oslo, Norway 4 DPS Aust-Agder, Sorlandet sykehus, Arendal, Norway 5 Research Institute, Vikersund, Norway 6 Department of Psychology, University of Oslo The study has been supported financially by the Norwegian ExtraFoundation for Health and Rehabilitation through EXTRA Funds, and by the Josef and Haldis Andresens Legat. Registration number NCT00202709, http://Clinical.Trials.gov. # Corresponding Author. Address: DPS Aust-Agder, Sorlandet sykehus, Box 783 Stoa, 4809 Arendal, Norway e-mail: [email protected]

© 2012 Elsevier Inc. All rights reserved ISSN 1550-8307/$36.00

Results: Repeated-measures analysis of variance was used to compare the TFT and the wait-list group. The TFT group had a significantly better outcome on two measures of anxiety and one measure of function. Follow-up data for all patients taken together showed a significant decrease in all symptoms during the one to two weeks between the pretreatment and the post-treatment assessments. The significant improvement seen after treatment was maintained at the three- and 12month assessments. Conclusions: The results suggest that TFT may have an enduring anxiety-reducing effect. Registration number NCT00202709, http://Clinical.Trials. gov. Key words: Thought field therapy, TFT, psychotherapy, anxiety disorders (Explore 2012; 8:331-338. © 2012 Elsevier Inc. All rights reserved.)

their study was not peer-reviewed, it showed remarkable results in favor of TFT and indicated that TFT not only works faster and better than cognitive behavioral therapy (CBT) but also has longer-lasting effects. They hypothesize that the tapping in TFT sends afferent signals that interact with activation signals produced when the patient thinks about an emotional problem and reduces symptoms in this way.2 In an article from 2010, David Feinstein describes several empirical studies on possible mechanisms of change that may support this hypothesis.3 Several other empirical studies describe possible mechanisms by which TFT, or the closely related Emotional Freedom Techniques, may reduce anxiety and other symptoms. In many of these studies, the influence on the amygdala has been highlighted as a possible common denominator.4-6 Other studies show that these therapies may influence the stress hormone cortisol,7 or brain wave activity, measured by electroencephalogram.8,9 In its original form, TFT involves tapping on specific points on the face, hands, and body while the patient produces the thoughts, feelings, or memories that are the target for treatment (eg, anxiety, sadness, and restlessness).1 There is an additional procedure of circling the eyes, counting, and humming, while tapping on the hand and concentrating on the target thought. Patients are trained to do the tapping themselves. We have found seven studies and three doctoral dissertations in which patient samples have been treated with TFT.2,10-18 None of these are randomized, controlled studies with longterm follow-up, with the exception of Andrade and Feinstein’s

EXPLORE November/December 2012, Vol. 8, No. 6 331 http://dx.doi.org/10.1016/j.explore.2012.08.002

study, which the authors state was meant to be a pilot study. However, these studies clearly indicate that TFT may have effect on various conditions. Of special interest here is the study by Schoninger and Hartung on public-speaking anxiety.18 Fortyeight subjects were randomized to either one session of TFT or wait-list control. The patients treated with TFT did better than the control group, and the results were maintained at the fourmonth follow-up. In addition, a randomized controlled study on EFT showed efficacy in reducing specific phobias of small animals.19 Criticisms of TFT have focused on whether it can be regarded as a combination of systematic desensitization, distraction, and demand characteristics;20 and several other problems.15,21-23 It is likely that these criticisms are related to the lack of established theoretical mechanisms behind the effect of TFT, to the fact that there are few randomized controlled studies on TFT, and that almost all studies are performed by researchers who themselves are involved in applying TFT in their therapeutic work. Given its rather widespread use, there is a need for more controlled studies of TFT. A natural next step is to investigate TFT in an anxiety disorder sample and examine whether it leads to improvements beyond those attributed to the passage of time and whether the improvements are stable. The purpose of our study was to examine the possible effect of TFT as a treatment for patients with anxiety disorder. Patients were first randomized to either immediate treatment with TFT, or to a two and a half-month waiting period with subsequent TFT treatment. All patients were assessed immediately after treatment, and three and 12 months after the end of treatment. The study hypothesis was that two sessions of TFT would be more effective than a wait-list control in relieving anxiety symptoms, and that the improvements during TFT treatments would be maintained at three- and 12-month follow-up periods.

METHODS Participants The study took place at the psychiatric department at Sorlandet Hospital in Arendal, Norway, from May 2002 to June 2003. Fifty-three patients were consecutively referred for participation in the study by colleagues in psychiatric practice and general practice over a period of eight months. As there are no studies on the use of TFT for anxiety disorders in general, we could not do a power analysis. The choice of sample size was therefore a compromise between access to patients and time available for the study. Inclusion criteria were the presence of anxiety disorders and willingness to be randomized. The only exclusion criterion was ongoing psychosis. All 53 referred patients were included in the study. After a structured diagnostic interview they were randomized to either treatment with TFT or to a two and a half months wait-list condition. Patients were consecutively randomized based on a computer-generated list. The randomization procedure was not blinded. The whole procedure of inclusion, randomization, group-allocation and follow-up was performed by the principal investigator. Eight patients were excluded from the study. One patient was symptom-free at the beginning of the study; four patients demanded to change groups (TFT n ⫽ 1; wait-list n ⫽ 3); three

332 EXPLORE November/December 2012, Vol. 8, No. 6

patients dropped out before treatment was initiated (TFT n ⫽ 1; wait-list n ⫽ 2). One patient in the immediate treatment group did not return after the first TFT treatment. Another patient in this group did not complete rating scales at the three- and 12month follow-up. One patient in the group receiving TFT after the waiting period did not provide data at the 12 month followup. We used last observation carried forward for the missing data for these three patients (Figure 1). A final sample of 45 patients was retained for assessment and treatment (TFT n ⫽ 23, wait-list n ⫽ 22); thirty-three women and 12 men. Mean age at study start was 37 years (range ⫽ 19-60 years). Mean duration of symptoms was 18 years (range ⫽ 2-45; Table 1). Mean number of psychiatric diagnoses were 4.18 (SD ⫽ 1.72) per participant, and the mean number of anxiety disorders were 2.02 (SD ⫽ 0.81). All anxiety disorders were represented in the patient sample. Most common were agoraphobia with or without panic disorder (n ⫽ 33), social phobia (n ⫽ 29), posttraumatic stress disorder (PTSD; n ⫽ 13), and generalized anxiety disorder (n ⫽ 10). Twenty-seven patients also exhibited some type of ongoing affective disorder; the most frequent was major depressive episode (n ⫽ 20). Measures Self-Assessment Questionnaires. Symptom Checklist 90-Revised (SCL-90-R)24 consists of 90 items, with a five-point scale for each item (0-4). The SCL-90 is divided into nine subscales. Only the global indicator of severity of the psychiatric illness (GSI) and the Positive Symptom Total (PST) were reported because the anxiety-related subscale showed the same tendency as the GSI (change on the GSI vs the anxiety subscale r ⫽ .70, P ⬍ .01). The GSI has been shown to be a reliable and valid measure of general symptomatic distress and is widely used for clinical and scientific purposes.25,26 The GSI was used as the primary outcome variable. The Hospital Anxiety and Depression Scale (HAD)27 measures the severity of symptoms of anxiety and depression. It consists of 14 items, seven for anxiety and seven for depression, with a four-point scale on each item (0-3). The HAD has shown to be sensitive to changes in response to treatment of anxiety disorders.28 The Sheehan Disability Scale (SDS) measures level of impairment on a scale with 11 points (0-10) on three items: work, social life and leisure activities, and family and domestic work.29 A Norwegian version of the SCL-90R has been validated by Lian et al.30 We have not found that the Norwegian versions of the HAD or the SDS have been validated. Nonetheless, the Norwegian version of the HAD is often used in clinical practice, and the SDS is a visual analogue scale with a small amount of text. We originally planned to include patients with social phobia, agoraphobia, and/or (PTSD); therefore symptom scales for these disorders also were used in the study. During the recruitment period, we found that our patients had a variety of anxiety disorders and decided to change the inclusion criteria to include all anxiety disorders, regarding PTSD as an anxiety disorder. We have therefore omitted the specific symptom scales. We used 11 questions on quality of life, originally used in another study. These are not validated, and were therefore omit-

TFT as a Treatment for Anxiety

Enrollment

Assessed for eligibility (n=53)

Excluded (n= 5) ♦ Not meeting inclusion criteria (n=1) ♦ Other reasons (n=4)

Randomized (n= 48)

Allocation Allocated to TFT (n=24) ♦ Received TFT (n= 23 ) ♦

Allocated to wait list (n=24) ♦ Received wait-list (n= 22)

Did not receive TFT (n=1) Did not come to treatment (n=1)

♦

Did not receive wait-list (n=2) Did not come to treatment (n=2)

Follow-Up Lost to follow-up (n=2). Rejected contact after first treatment (n=1) Rejected contact after first follow-up (n=1)

Lost to follow-up (n=1) Died before last follow-up (n=1)

Analysis Analysed (n= 23) ♦ Excluded from analysis (n=1) Did not come to treatment (n=1)

Analysed (n=22) ♦ Excluded from analysis (n=2) Did not come to treatment (n=2)

Figure 1. Flow diagram.

ted from this article. The omitted scales show the same tendency as the six scales we have reported.

the items must be scored as completely fulfilled to give a diagnosis, except for antisocial personality disorder.

Interviews. Diagnoses were based on the DSM-IV.31 The MiniInternational Neuropsychiatric Interview (M.I.N.I.) version 5.0.032 was used to assess axis I disorders. It is a semi-structured interview that covers mood disorders, anxiety disorders except specific phobias, substance-related disorders, eating disorders, schizophrenia, and other psychotic disorders. In screening for personality disorders we used the Iowa Personality Disorder Screen (IOWA),33 which consists of 11 items. A cut-off of two positive answers on items 1-6 was used in this study. This has been found to yield optimal screening properties, with a sensitivity of 96% and a specificity of 68% when compared with the diagnostic results for at least one personality disorder by the Structured Clinical Interview for DSM Personality Disorders (SCID II).31,33 The SCID II interview was used to diagnose DSM-IV axis II personality disorders.31 It consists of 94 questions that cover the criteria for each of the 10 personality disorders. Four or five of

Procedure. Eleven patients were diagnosed and evaluated by a psychiatrist in residence. For those of her patients who scored positively on the IOWA, the principal investigator (A.I.) performed a SCID II interview for personality disorders. The rest of the patients were diagnosed and evaluated by the principal investigator. Pretreatment evaluation comprised diagnostic interview and the self-rating scales. Assessment was done at four occasions: one to five weeks before treatment, one to two weeks after treatment, three months after treatment, and 12 months after treatment. The wait-list group had a fifth assessment point, as there was an additional evaluation before the waiting period.

TFT as a Treatment for Anxiety

TFT intervention. In ordinary TFT practice, patients are offered a treatment package of five hours of individual therapy. The patients in the current study received only two sessions of indi-

EXPLORE November/December 2012, Vol. 8, No. 6 333

Table 1. Demographic Data

were calculated using pooled standard deviation for admission and follow-up. Statistics were performed using SPSS version 16.0.

Group Characteristic Female Age at study start, years Duration of symptoms, years Number of anxiety diagnoses, mean Number of other psychiatric diagnoses, mean Number of patients diagnosed with one or more personality disorders Number of patients using psychotropic drugs, all types Number of patients using benzodiazepines Number of patients using antidepressives Number of patients receiving additional psychotherapy

TFT (n ⫽ 23)

Wait list (n ⫽ 22)

17 36.6 20.7 2.2 2.4

16 37.4 16.2 1.8 2.0

14

7

17

17

9 11 9

6 13 6

Note. Relevant tests revealed no significant (P ⬎ .05) differences between the two groups, though for personality disorder the P value was 0.051.

vidual therapy, the first one lasting up to 50 minutes, the other up to 25 minutes. The reason for giving such a short treatment was that the TFT therapist did not have time for a full treatment package, because he practices in another part of the country. Clinical experience has shown that such a short treatment may be sufficient. Additional sessions were provided by the first author, when judged as necessary. In the 12-month follow-up, 11 patients received no additional treatment by the first author, eight patients received one session, 14 patients received from two to five sessions, six patients received from six to 10 sessions, and six patients received from 11 to 23 additional sessions. The treatment was based on the TFT guidelines1 and was conducted by an experienced TFT therapist. The treatment sessions were observed by the first author in order to ensure that the treatment given was TFT and that the treatment was provided in accordance with the study protocol, but no treatment fidelity instrument was applied. Statistics To compare groups, independent sample t tests or chi-square tests were used. To compare change of symptoms in the two groups, a time (pre-treatment and posttreatment/end of waiting period) ⫻ treatment condition multivariate analysis of variance (MANOVA) was conducted. If the MANOVA yielded a significant (P ⬍ .05) interaction, the individual measures were analyzed with repeated-measures ANOVA.34 In the follow-up investigations, a repeated-measures MANOVA was performed on all the six outcome measures across pretreatment, posttreatment, and three and twelve months’ follow-up. If there was a significant time effect, a repeated-measures ANOVA would be performed on the individual measures. Because we were interested in the stability of the improvements occurring from pre-to posttreatment, we used the simple contrast. Effect sizes (Cohen’s d)

334 EXPLORE November/December 2012, Vol. 8, No. 6

Ethics The regional medical ethics committee approved the study before the start of the study. Written consent was obtained from each patient at the first interview.

RESULTS The means, SDs, statistics, and effect sizes for each of the six scales for symptoms and functioning for the period, comparing TFT with wait-list are shown in Table 2. Group ⫻ time MANOVA for the six scales taken together demonstrated a significant time effect, F(1,43) ⫽ 4.06, P ⬍ .05, and a group X time effect, F(1,43) ⫽ 2.65, P ⬍ .05. Follow-up repeated measures ANOVAs of the individual scales revealed significant group ⫻ time interaction effects for scores on SCL-90R GSI, HAD anxiety subscale, and Sheehan disability scale on social life and leisure activities. The HAD depression subscale, Sheehan disability subscales on work and on family and domestic work, showed no significant effect. Inspection of the means revealed that the patients in the TFT group improved more than the patients in the wait-list group on all scales. The SCL-90R PST showed the same tendency as the GSI. In the TFT group the mean number of positive scores in the SCL 90-R was reduced by 17.0, whereas in the wait-list group the mean reduction was 2.3, P ⫽ .001. This reduction was maintained at the 12 months’ follow-up. Table 3 shows the means, SDs, and Cohen’s d for all the time intervals for all patients in the study, and F for the difference between measures before and 12 months after treatment. At 12 months’ follow-up, all outcome measures showed a significant change. All effect sizes were based on change from pretreatment values and showed a sharp decrease in terms of both primary and secondary outcome measures immediately after treatment, as indicated by small to large effect sizes (range, 0.38-0.82). At three months’ follow-up, no additional effect were detected in terms of effect sizes (range, 0.33-0.86). At 12 months’ follow-up, effect sizes were still small to large (0.44-0.97). This pattern is illustrated by the scores on the SCL-90 R GSI in Figure 2; the figures are quite similar for the five other measures. We have used criterions from Tingey and al35 to evaluate changes in general symptomatic distress (SCL-90 R GSI) from just before initiation of therapy to 12 months after completion. They define mean GSI scores according to a normative continuum as up to 0.19 (SD ⫽ 0.16) for an asymptomatic sample, 0.31 (SD ⫽ 0.31) for the mildly symptomatic, 0.79 (SD ⫽ 0.45) for the moderately symptomatic, and 1.30 (SD ⫽ 0.82) for the severely symptomatic. According to these criteria, 6 of our patients where asymptomatic 12 months after treatment, with a GSI lower than 0.19, and five were classified as mildly symptomatic, with a GSI between 0.19 and 0.31. All these 11 (24%) patients had experienced a more than 50% reduction of their symptoms. In addition, 10 more patients experienced a more than 50% reduction of their symptoms, although they had a GSI score higher than 0.31 on the 12 months’ follow-up.

TFT as a Treatment for Anxiety

Table 2. Comparison of TFT Treatment and 2.5 Months Wait-List, Means and Standard Deviations, and Effect Sizes (Cohen’s d) for the Difference in Effect Between the Treatment Group and The Wait-List Group Group TFT (n ⫽ 23) Measure SCL-90R, GSI Pretreatment Posttreatment HAD, anxiety Pretreatment Posttreatment HAD, depression Pretreatment Posttreatment Sheehan, work Pretreatment Posttreatment Sheehan, social Pretreatment Posttreatment Sheehan, family Pretreatment Post-treatment

Wait List (n ⫽ 22)

F (1,43)

F (1,43)

M

SD

M

SD

Time

Time ⫻ Group

d

1.57 1.06

0.67 0.84

1.31 1.22

0.66 0.69

20.6b

10.00b

0.96

13.0 9.43

3.63 5.23

4.61 4.79

11.4b

8.84b

0.89

7.43 5.52

3.17 4.27

6.91 6.59

3.89 4.98

4.73a

2.42

0.46

7.52 5.96

3.18 4.19

7.50 6.77

2.96 3.80

5.32a

0.71

0.25

6.83 4.74

2.06 3.36

4.95 5.95

3.09 2.73

1.78

14.3b

1.13

5.26 3.74

2.24 2.88

4.45 4.05

2.72 3.15

6.24a

2.07

0.43

12.4 12.1

Note. a P ⬍ .05. b P ⬍ .01.

The number of additional sessions during the 12 months follow-up period varied from none to 23 with a median of two and a mean of 4.51 (SD 5.97). A significant negative correlation was found between the effect size and the number of sessions on the three scales that show significant positive change when TFT was compared to wait-list condition. This varied from ⫺0.34 on the Sheehan disability scale on social life and leisure activities to ⫺0.41 on the SCL-90R GSI, P ⬍ .05. Harms or unintended effects were not systematically registered. One patient reported worsening of traumatic memories after treatment.

DISCUSSION An overall test, across the six scales of function and general symptoms, indicates that the patients who received TFT changed more than the patients in the wait-list group. It is noteworthy that the initial positive effect demonstrated on all six scales immediately after treatment was retained at the follow-ups after three and 12 months. This finding supports the hypothesis that TFT may be a viable treatment for anxiety symptoms. In their meta-analytic review of the efficacy of CBT across the anxiety disorders, Norton and Price36 described overall effect

Table 3. Means and SDs for all 45 Patients Taken Together, Measured (1) Before Treatment, (2) After Treatment, (3) Three Months After Treatment, and (4) 12 Months After Treatment (1) Before Treatment SCL-90R, GSI HAD, anxiety HAD, depression Sheehan, work Sheehan, social Sheehan, family

(2) Immediately After Treatment

(3) Three Months After Treatment

(4) 12 Months After Treatment

M

SD

M

SD

d

M

SD

d

M

SD

d

F (1,45)

1.40 12.6 7.02 7.16 6.40 4.67

0.70 4.21 4.13 3.48 2.43 2.76

0.98 9.42 5.67 5.84 4.80 3.71

0.78 4.80 4.42 3.52 3.38 2.90

0.76 0.82 0.43 0.49 0.63 0.38

0.98 9.38 5.71 5.89 4.58 3.69

0.78 5.30 4.46 3.97 3.30 2.71

0.86 0.81 0.41 0.43 0.70 0.33

0.95 9.31 5.33 5.38 4.13 3.18

0.83 5.21 4.70 3.86 3.17 2.78

0.93 0.86 0.44 0.51 0.97 0.53

21.5a 20.1a 4.46b 6.62a 15.2a 5.79b

Note. Effect sizes (Cohen’s d) are estimated in relation to before treatment values. F values are computed as time effect across the assessments. a P ⬍ .001. b P ⬍ .01.

TFT as a Treatment for Anxiety

EXPLORE November/December 2012, Vol. 8, No. 6 335

Figure 2. Mean GSI scores for all 45 patients across the assessments before and after treatment.

sizes of CBT ranging from 1.14 to 1.98. The effect sizes found in our study are substantially less, ranging from 0.25 to 1.13 when comparing TFT to wait-list conditions, and from 0.44 to 0.97 on the 12 months follow-up. Nonetheless, these results support the study hypothesis. Study frames gave opportunities for notably fewer sessions with TFT than what is the usual treatment package, one and a half session versus five sessions. Additional sessions were given by the first author only when it was obvious that patients needed more treatment than the initial one and a half sessions. This is reflected in the negative correlation between positive changes on the SCL 90R GSI and the number of additional sessions. Twenty-one (47%) of the 45 patients experienced a more than 50% reduction on the SCL 90R GSI from just before treatment to the 12 months follow-up. Six patients were symptom-free at this last follow up. None of these six patients received treatment from outside therapists during the study period. Two of them got one brief therapeutic intervention from the first author; one immediately following TFT treatment and one in the three months follow-up period. Three of these six patients had experienced symptoms of social phobia (n ⫽ 2) and generalized anxiety disorder (n ⫽ 1) as long as they could remember. For the other three, symptom durations were 3.5, 6, and 12 years. These findings underscore the possibility that TFT may have positive effects for some patients. Results on the symptom scales show that TFT had an immediate effect on symptoms of anxiety but not on symptoms of depression. This is demonstrated by a nonsignificant change in depression symptoms when compared with wait-list, and a smaller Cohen’s d in the follow-up part of the study. This fits in with the study’s aim of treating primarily anxiety symptoms. Likewise, the significant effect on the Sheehan score on social

336 EXPLORE November/December 2012, Vol. 8, No. 6

life and leisure activities, in opposition to the two other function scales, may be attributable to TFT targeting symptoms of social phobia in this study. One challenge with TFT is its theoretical foundation, which is controversial because it is based on theories of energy meridians derived from traditional Chinese medicine. However, the techniques used in TFT are also based on the exposure effect: when thinking of a problem, feelings associated with this problem get triggered, and when triggered, they are accessible to reduction by the use of TFT. Exposure is an effective and common intervention used in many psychotherapies.37 It is possible that the observed effect of TFT is mediated by reciprocal inhibition, as in applied relaxation,38 but it is beyond the scope of this study to dismantle the TFT techniques in order to investigate possible underlying mechanisms of effect. It is problematic that the randomization list was open to the primary investigator. Because all potential patients were included consecutively, the open randomization list did not represent a source of bias. Placebo effects may well be the reason for the observed differences between the wait-list group and the TFT group that is described in Table 2. It is less likely that placebo effects alone can explain the long-standing positive results three and 12 months after primary treatment with TFT, as it was the patients who received few or no additional treatments who had the best lasting results. We have not found studies where prolonged placebo has been compared to psychotherapy. To be able to include enough patients within our time constraints, we accepted the use of any kind of medication or additional therapy. This limits the ability to draw a conclusion from this study. There was a negative correlation between symptom-

TFT as a Treatment for Anxiety

improvement and receiving additional therapy from a therapist outside the study. Some patients reported that they worked on problems other than their anxiety disorder with their therapist. For other patients, this finding may mean that those who did not experience enough effect from the TFT treatment chose to continue their original therapy.

CONCLUSIONS The results of this randomized study with wait-list controls and 12-month follow-up suggest that TFT is effective for reducing symptoms of anxiety in patients with anxiety disorders. However, lack of appropriate controls does not allow for any conclusion regarding the cause of this change. It is possible that the observed change is due to therapy content-specific effects, but may also be unspecific effects due to contact with an empathic therapist. Properly designed RCTs with active control treatment arm and placebo are needed to validate TFT as an effective treatment for symptoms of anxiety. Acknowledgments The authors want to thank Eva Cecilie Orvin for her help with diagnostic and initial evaluation, Mats Uldal for the TFT therapy, and Kathrine Oveland for her help with coding and writing.

REFERENCES 1. Callahan RJ, Trubo R. Tapping the Healer Within: Using Thought Field Therapy to Instantly Conquer Your Fears, Anxieties, and Emotional Distress. Chicago: Contemporary Books; 2001. 2. Andrade J, Feinstein D. Energy psychology: Theory, indications, evidence. In Feinstein, D. Energy psychology interactive (Appendix, pp. 199-214). Ashland, OR: Innersource; 2004 3. Feinstein D. Rapid treatment of PTSD: why psychological exposure with acupoint tapping may be effective. Psychotherapy (Chic). 2010; 47:385-402. 4. Lane JR. The neurochemistry of counterconditioning: acupressure desensitization in psychotherapy. Energ Psychol. 2009;1:31-44. 5. Hoss RJ, Hoss LM. The dream to freedom technique, a methodology for integrating the complementary therapies of energy psychology and dreamwork. Energy. 2010;2:1. 6. Feinstein D, Church D. Modulating gene expression through psychotherapy: the contribution of noninvasive somatic interventions. Rev Gen Psychol. 2010;14:283-295. 7. Church D, Yount G, Brooks A. The effect of emotional freedom techniques (EFT) on stress biochemistry: A randomized controlled trial. J Nerv Ment Dis. in press. 8. Lambrou P, Pratt G, Chevalier G. Physiological and psychological effects of a mind/body therapy on claustrophobia. Subtle energies and energy. Medicine. 2003;14:239-251. 9. Swingle PG, Pulos L, Swingle MK. Neurophysiological indicators of successful EFT treatment of posttraumatic stress. Subtle energies and energy. Medicine. 2005;15:75-86. 10. Johnson C, Shala M, Sejdijaj X, Odell R, Dabishevci K. Thought Field therapy—soothing the bad moments of Kosovo. J Clin Psychol. 2001;57:1237-1240. 11. Sakai C, Paperny D, Mathews M, et al. Thought field therapy clinical applications: utilization in an HMO in behavioral medicine and behavioral health services. J Clin Psychol. 2001;57:1215-1227. 12. Folkes CE. Thought field therapy and trauma recovery. Int J Emerg Ment Health. 2002;4:99-104.

TFT as a Treatment for Anxiety

13. Darby DW. The efficacy of thought field therapy as a treatment modality for individuals diagnosed with blood-injection-injury phobia. Dissert Abstr Int B Scien Eng. 2003;64:1485. 14. Kessler RA. The differential impact of thought field therapy as a treatment modality for male perpetrators of domestic violence diagnosed with posttraumatic stress disorder. Dissert Abstr Int B Scien Eng. 2003;63:6097. 15. Pignotti M. Thought field therapy voice technology vs. random meridian point sequences: A single-blind controlled experiment. The scientific review of mental health practice: objective investigations of controversial and unorthodox claims in clinical psychology, psychiatry, and social. Work. 2005;4:38-47. 16. Upshaw R. The efficacy of thought field therapy as an adjunct treatment modality for male domestic-violence perpetrators with domestic abuse in their family of origin. Dissert Abstr Int B Scien Eng. 2006;66:4503. 17. Sakai CE, Connolly SM, Oas P. Treatment of PTSD in Rwandan child genocide survivors using thought Field therapy. Int J Emerg Ment Health. 2010;12:41-49. 18. Schoninger B, Hartung J. Changes on self-report measures of public speaking anxiety following treatment with thought field therapy. Energ Psychol. 2010;2. 19. Wells S, Polglase K, Andrews HB, Carrington P, Baker AH. Evaluation of a meridian-based intervention, emotional freedom techniques (EFT), for reducing specific phobias of small animals. J Clin Psychol. 2003;59:943-966. 20. Devilly GJ. Power therapies and possible threats to the science of psychology and psychiatry. Aust N Z J Psychiatry. 2005;39:437-445. 21. McNally RJ. Tertullian’s motto and Callahan’s method. J Clin Psychol. 2001;57:1171-1174. 22. Rosen GM, Davison GC. “Echo attributions” and other risks when publishing on novel therapies without peer review. J Clin Psychol. 2001;57:1245-1250. 23. Rosner R. Between search and research: how to find your way around? Review of the article “Thought Field Therapy—soothing the bad moments of Kosovo.” J Clin Psychol. 2001;57:1241-1244. 24. Derogatis LR. SCL-90 Administration, Scoring and Procedures Manual-I for the R (Revised) Version and Other Instruments of the Psychopathology Rating Scale Series. Baltimore: Johns Hopkins University School of Medicine; 1977. 25. Cyr JJ, McKenna-Foley JM, Peacock E. Factor structure of the SCL90-R: is there one? J Pers Assess. 1985;49:571-578. 26. Rauter UK, Leonard CE, Swett CP. SCL-90-R factor structure in an acute, involuntary, adult psychiatric inpatient sample. J Clin Psychol. 1996;52:625-629. 27. Zigmond AS, Snaith RP. The Hospital Anxiety and Depression Scale. Acta Psychiatr Scand. 1983;67:361-370. 28. Johnston M, Pollard B, Hennessey P. Construct validation of the hospital anxiety and depression scale with clinical populations. J Psychosom Res. 2000;48:579-584. 29. Leon AC, Olfson M, Portera L, Farber L, Sheehan DV. Assessing psychiatric impairment in primary care with the Sheehan Disability Scale. Int J Psychiatry Med. 1997;27:93-105. 30. Lian L, Vassend O, Andersen HT. Norwegian versions of the neuroticism, extraversion, and openness (NEO) Personality Inventory, SCL-90-revised, and Giessen subjective complaints list: II. Tidsskr Norsk Psykologforening. 1993;30:117-126. 31. Diagnostic and Statistical Manual of Mental Disorders: DSM-IV. Washington, DC: American Psychiatric Association; 1994. 32. Sheehan DV, Lecrubier Y, Sheehan K, et al. The mini-international neuropsychiatric interview (M.I.N.I): the development and validation of a structured diagnostic psychiatric interview for DSM-IV and ICD-10. J Clin Psychiatry. 1998;59:22-33.

EXPLORE November/December 2012, Vol. 8, No. 6 337

33. Langbehn DR, Pfohl BM, Reynolds S, et al. The Iowa personality disorder screen: development and preliminary validation of a brief screening interview. J Pers Disord. 1999;13:75-89. 34. Winer BJ. Statistical Principles in Experimental Design. New York: McGraw-Hill; 1971. 35. Tingey R, Lambert M, Burlingame G, Hansen N. Assessing clinical significance: proposed extensions to method. Psychother Res. 1996; 6:109-123.

338 EXPLORE November/December 2012, Vol. 8, No. 6

36. Norton PJ, Price EC. A meta-analytic review of adult cognitivebehavioral treatment outcome across the anxiety disorders. J Nerv Ment Dis. 2007;195:521-531. 37. Clark DM, Fairburn CG. Science and Practice of Cognitive Behaviour Therapy. Oxford: Oxford University Press; 1997. 38. Ost LG. Applied relaxation: description of a coping technique and review of controlled studies. Behav Res Ther. 1987;25:397409.

TFT as a Treatment for Anxiety