Ketamine-Assisted Psychotherapy Provides Lasting and Effective Results in the Treatment of Depression, Anxiety, and Post-Traumatic Stress Disorder at 3 and 6 Months: Findings from a Large Retrospective Effectiveness Study

Abstract

Importance:

Ketamine-assisted psychotherapy (KAP) is an emerging treatment option to alleviate treatment-resistant affective disorders, but its long-term effectiveness remains unclear.

Objective:

To examine the treatment effects of KAP on anxiety, depression, and post-traumatic stress disorder (PTSD) at 1, 3, and 6 months post-treatment.

Design, Setting, and Participants:

This retrospective effectiveness study included self-reported outcomes from adults with a history of major depressive disorder, generalized anxiety disorder (GAD), or PTSD who had not responded to prior treatment interventions and received KAP administered across 11 Field Trip Health clinics in North America between March 13, 2020, and June 16, 2022. The evaluable sample sizes were 346 and 94 participants at 3 and 6 months, respectively, representing loss to follow-up rates of 82% and 95%.

Intervention:

KAP consists of 4–6 guided ketamine sessions (administered through intramuscular injection or sublingual lozenge) with psychotherapy-only integration visits after doses 1 and 2 and then after every 2 subsequent doses. Mean number of doses administered was 4, standard deviation (SD) = 3, and mean number of integration sessions was 3, SD = 2.

Main Outcomes and Measures:

Primary outcomes were changes in symptoms of depression, anxiety, and PTSD at 3 months relative to baseline, assessed, respectively, using the 9-item Patient Health Questionnaire, the 7-item GAD measure, and the 6-item PTSD checklist. Secondary outcomes were changes at 1 and 6 months relative to baseline.

Results:

Large treatment effects were detected at 3 months (d's = 0.75–0.86) and were sustained at 6 months (d's = 0.61–0.73). Case reductions (identified based on cutoff values) ranged from 39% to 41% at 3 months and 29% to 37% at 6 months. In total, 50–75% reported a minimal clinically important difference at 3 months and 48–70% at 6 months.

Conclusions and Relevance:

KAP produced sustained reductions in anxiety, depression, and PTSD, with symptom improvement lasting well beyond the duration of dosing and integration sessions. These effects extended to as much as 5 months after the last KAP session. However, the high rates of attrition may limit validity of the results. Given the growing mental health care crises and the need for effective therapies and models of care, especially for intractable psychiatric mood-related disorders, these data support the use of KAP as a viable alternative. Further prospective clinical research should be undertaken to provide evidence on the safety and effectiveness of ketamine within a psychotherapeutic context.

Key Points

Question: What are the lasting effects of ketamine-assisted psychotherapy (KAP) on psychological distress?

Findings: In this retrospective effectiveness study, the sample sizes were 346 and 94 participants at 3 and 6 months, respectively, representing loss to follow-up rates of 82% and 95%. Large effect sizes were found at 3 months on depression, anxiety, and post-traumatic stress (d's = 0.75–0.86) that were sustained at 6 months.

Meaning: These findings suggest that KAP is an effective treatment option with substantial clinical benefits detected for up to half a year.

Introduction

Over the past two decades, ketamine has demonstrated the potential to produce rapid and sustained antidepressant effects and therapeutic outcomes for several psychiatric conditions.¹ Most studies to date have looked at the administration of ketamine through intravenous administration^2,3; however, application of ketamine has evolved to include psychotherapeutic practices to reduce its unwanted dissociative effects while bolstering positive antidepressant, anxiolytic, antistress, and other positive mood effects.⁴ Ketamine-assisted psychotherapy (KAP) has borrowed from earlier research on psychedelic substances such as lysergic acid diethylamide to facilitate deep and rapid introspective work within an organized psychotherapeutic framework.⁵

Currently, ketamine is the only legal psychedelic medicine to treat persistent and severe psychological distress in North America, making it suitable for wider psychotherapeutic implementation. Preliminary clinical trial data are increasingly supportive of the safety and efficacy of KAP to alleviate anxiety, depression, and post-traumatic stress disorder (PTSD),^6,7 with more trials underway worldwide. In a recent study of patients with moderate-to-severe depression or anxiety, ketamine-assisted therapy was found to have immediate effects on depression and anxiety; outcomes showed persistent efficacy to 2 and 4 weeks in most treated subjects.⁸ However, the impact on post-treatment quality of life is understudied, and there remains a need to clarify the long-term effectiveness of KAP for a variety of mood-related disorders.

In this study, we conducted a retrospective analysis of outcomes from patients who were treated with KAP at Field Trip Health centers across North America and agreed to participate in an open label evaluation of outcomes up to 6 months after the intervention. Ketamine was administered through intramuscular injection or sublingual lozenge. Primary outcomes were changes in depression, anxiety, and post-traumatic stress at 3 months from baseline. Secondary outcomes were changes at 1 and 6 months from baseline.

Methods

Ethics and design

Ethics approval for this observation of KAP study was granted by Veritas Independent Review Board (No. 2022-3067-11240-5). This was a retrospective effectiveness study of KAP involving chart review of patient self- reported mental health outcomes assessed at baseline (T0), 1 month (T1), 3 months (T3), and 6 months (T6).

Participants and procedure

Data were collected from clients treated across 11 Field Trip Health clinics in North America: (1) Toronto, ON, Canada; (2) Vancouver, BC, Canada; (3) Fredericton, NB, Canada; (4) New York City, NY, USA; (5) Atlanta, GA, USA; (6) Chicago, IL, USA; (7) Houston, TX, USA; (8) Seattle, WA, USA; (9) Santa Monica, CA, USA; (10) La Jolla, CA, USA; and (11) Washington, DC, USA, between March 13, 2020, and June 16, 2022.

Participants included in the analysis had a documented history of depression or anxiety that showed lack of adequate response to previous treatment(s) or presented with PTSD as assessed by psychiatric clinicians.⁹ Patient assessments were collected remotely using a proprietary digital platform (PortalTM) and through the electronic medical records system at 1, 3, and 6 months. The completion of measures was voluntary after the collection of signed written informed consent.

Treatment

Prospective clients were either self-referred in the United States or referred for treatment by a health care provider to one of the three Canadian clinic locations. All clients were assessed by a psychiatrist or psychiatric nurse practitioner to determine the appropriateness for treatment. Inclusion criteria included signed written informed consent; being over the age of 18 years; and having a documented, prior diagnosis of one or more of major depressive disorder (MDD), bipolar depression, generalized anxiety disorder (GAD), obsessive compulsive disorder, eating disorder, or a significant history of trauma and/or a formal diagnosis of PTSD as per the Diagnostic and Statistical Manual of Mental Disorders, Fifth Edition (DSM-5) (American Psychiatric Association, 2013).¹⁰

Exclusion criteria included pregnant women and nursing mothers, although postpartum depression was considered on a case-by-case basis in consultation with the national medical director; a relative (not absolute) contraindication for individuals with a body mass index >35 kg/m²; any individual who has met DSM 5 criteria for a substance use disorder in the past 3 months and had been unable to exhibit a reduction in use; psychosis or psychotic symptoms; active mania: bipolar 1 (chronic nondisruptive hypomania is an exception at the discretion of the treatment team); borderline personality disorder; uncontrolled medical disorders or physical conditions with negative interaction with ketamine; individuals with symptomatic acute brain injury within 90 days of serious injury; individuals diagnosed with moderate-to-severe sleep apnea; and individuals who were unable to identify a person or service to assure their safe transport to home post-treatment.

Upon medical approval, clients met with a licensed therapist to discuss preparation for KAP and to initiate a therapeutic relationship. Dosing with ketamine was completed through intramuscular injection in the United States and Vancouver and through sublingual lozenge in Toronto and Fredericton. Clients were offered personalized treatment recommendations consisting of 4–6 guided ketamine sessions with psychotherapy-only integration visits after doses 1 and 2 and then after every 2 subsequent doses. Patients with a main diagnosis of PTSD were encouraged to engage in additional preparatory work and integration sessions.

Initial dosing through intramuscular injection was 25–50 mg and clinicians in shared decision making with patients could increase beyond that if clinically indicated. Subsequent visit doses were often titrated higher into a typical range of 50–100 mg. The initial dose for lozenges was 200 mg with the option to increase by 50–100 mg per visit up to 500 mg. Ketamine doses were not scheduled on consecutive days and could be interspersed by 1 week or more. Some participants completed additional sessions beyond dose 6.

Integration sessions were based on a clinical protocol that consolidated motivational interviewing¹¹ and behavioral activation,¹² and trauma-informed therapeutic techniques. Therapists, most of whom were hired with skillsets in basic cognitive behavioral therapy, motivational interviewing, and trauma-informed care, had the option to integrate other psychotherapeutic modalities depending on their assessment of the client's needs, unique goals, and established clinical competencies.

Ketamine dosing was completed in Field Trip Health clinics using an approach consistent with psychedelic studies in a setting designed to be aesthetically and functionally conducive to a state of relaxation. Clients received ketamine while seated in a comfortable reclining chair, wearing an eye shade, and listening to curated music playlists. Therapists were present to support clients during dosing sessions while medical staff monitored heart rate, blood pressure, respiration rate, and oxygen saturation throughout the session.

Outcome measures

Symptoms of depression were assessed by the 9-item Patient Health Questionnaire (PHQ-9).¹³ PHQ scores may range from 0 to 27 with higher scores indicating more severe depressive symptoms. A PHQ cutoff score of 15 has been validated to identify cases with at least moderately severe depression.¹³ A change of 3 points is considered a minimal clinically important difference (MCID).^14,15

The PHQ also contains an item assessing suicidal ideation (SUI) that can range from 0 to 3, with higher scores indicating greater ideation. A cutoff value of 2 may be used to indicate cases of at least moderate severity.

Symptoms of anxiety were assessed by the 7-item GAD measure.¹⁶ GAD scores may range from 0 to 21 with higher scores indicating more severe anxiety symptoms. A GAD cutoff score of 10 has been validated to identify cases with at least moderate anxiety.¹⁶ A change of 3 points has been considered an MCID.^14,15

Symptoms of post-traumatic stress were assessed using the 6-item PTSD checklist (PCL-6).¹⁷ PCL scores may range from 6 to 30, with higher scores indicating more severe stress symptoms. A PCL cutoff score of 14 has been validated to identify cases of PTSD.¹⁸ A change of 5 points may be considered an MCID.¹⁸

Statistical analysis

We report descriptive statistics for the sample and analyze the extent of loss to follow-up. The primary analysis was by intention to treat. We used linear mixed modeling to fit growth curves describing the normative patient trajectory on each outcome.¹⁹ The main analysis involved fitting linear and curvilinear trends over time and estimating mean differences at each endpoint compared with the baseline. Cohen's d was reported as a standardized measure of effect size for mean differences (d = 0.2 is a small effect, 0.5 a medium effect, and 0.8 a large effect).

We also report secondary analyses on the effect of doses administered, controlling for age, gender, and site differences; case reductions in depression, anxiety, and PTSD based on cutoffs; and proportions of treatment responders based on MCIDs. As a sensitivity analysis, we used the expectation–maximization and Markov Chain Monte Carlo algorithms to multiply imputed missing follow-up data.²⁰ We simulated 1000 data sets per outcome measure, calculated the effects per imputation, and combined the findings across imputations to achieve best estimates for comparison with initial estimates.

Results

Descriptive statistics

In total, 1806 participants entered treatment. Overall, the mean age was 42 years, standard deviation (SD) = 12, and 52% of participants were female. Most individuals had a primary diagnosis of MDD (24%), GAD (28%), or PTSD (25%). The mean number of assessments completed per participant was 2.26, SD = 2.01. In total, 18% of baseline participants provided a 3-month assessment, and 5% provided a 6-month assessment (see Fig. 1 for patient flow diagram).

FIG. 1.

Patient flow diagram.

The mean number of doses was 4, SD = 3, with 12% (210/1806) receiving >6 doses and 24% (440/1806) receiving 1 dose. Ketamine was administered sublingually to 32% (579/1806) of the sample and intramuscularly to 68% (1227/1806). The mean number of (psychotherapeutic) integration sessions was 3, SD = 2, with 20% (365/1806) receiving >4 sessions.

See Table 1 for a comparison of demographic and clinical variables between those who did or did not have follow-up data on the PHQ. Those who had follow-up data received more treatment (ketamine dose and integration sessions). They were also more likely to have had sublingual administration, a primary diagnosis of MDD, and to be White.

Table 1.

Comparison of Participants Who Did or Did Not Have Follow-Up Data on the Patient Health Questionnaire

Demographic and clinical variables	Had follow-up data ( N  = 616)		Did not have follow-up data ( N  = 1190)		p
	Mean	SD	Mean	SD
Age	42.87	12.77	41.73	10.94	0.06
Frequency of ketamine doses administered	5.74	2.62	3.15	2.29	<0.0001
Frequency of integration sessions delivered	4.37	2.11	2.56	1.67	<0.0001

	Frequency	Percent	Frequency	Percent
Gender					0.41
Women	310	50.32	638	53.61
Men	303	49.19	547	45.97
Nonbinary	3	0.49	5	0.42
Ethnicity					0.0003
White	483	78.41	826	69.41
Hispanic	17	2.76	57	4.79
Black	12	1.95	57	4.79
East Asian	20	3.25	35	2.94
South Asian	10	1.62	36	3.03
Middle Eastern	5	0.81	24	2.02
Southeast Asian	10	1.62	12	1.01
Other/missing	59	9.58	143	12.02
Primary diagnosis					<0.0001
Anxiety	157	26.21	331	28.88
Bipolar disorder	5	0.83	11	0.96
Depression	177	29.55	241	21.03
Eating disorder	12	2	15	1.31
Obsessive-compulsive disorder	17	2.84	30	2.62
PTSD	156	26.04	286	24.96
Substance use disorder	27	4.51	58	5.06
Other	48	8.01	174	15.18
Route of administration					<0.0001
Intramuscular	334	54.22	893	75.04
Sublingual	282	45.78	297	24.96

T0 = baseline; T1 = 1-month assessment; T3 = 3-month assessment; T6 = 6-month assessment.

PTSD, post-traumatic stress disorder; SD, standard deviation.

Loss to follow-up analyses

Loss to follow-up refers to individuals who did not provide a follow-up assessment and is separate from treatment status. Loss to follow-up was associated with less pretreatment psychological distress.

Compared with those who provided a 3-month assessment, those who did not had lower baseline scores on PHQ, M_Diff = −1.83, CI_.95 (−1.09 to −2.58), p < 0.0001, d = 0.28, and PCL, M_Diff = −1.56, CI_.95 (−0.86 to −2.25), p < 0.0001, d = 0.28. Compared with those who provided a 6-month assessment, those who did not had lower baseline scores on GAD, M_Diff = −1.58, CI_.95 (−0.26 to −2.90), p = 0.02, d = 0.28, and PCL, M_Diff = −1.86, CI_.95 (−0.51 to −3.22), p = 0.007, d = 0.33. Overall, the frequency of missing assessments tended to be negatively correlated with baseline scores on the PHQ, r = −0.13, p < 0.0001; the GAD, r = −0.06, p = 0.02; and the PCL, r = −0.10, p < 0.0001.

Primary and secondary outcomes

The primary outcomes were changes at 3 months from baseline on self-reported PHQ, GAD, and PCL measures and secondary outcomes were changes at 1 and 6 months from baseline on these measures. There was a significant reduction on the PHQ at 1 month, d = 0.50, which was amplified at 3 months, d = 0.85, and remained detectable at 6 months, d = 0.73. There was a reduction on the GAD at 1 month, d = 0.47, which was amplified at 3 months, d = 0.86, and remained detectable at 6 months, d = 0.73. There was a reduction on the PCL at 1 month, d = 0.38, which was amplified at 3 months, d = 0.75, and remained detectable at 6 months, d = 0.61. See Table 2 for details.

Table 2.

Estimated Mean Differences from Linear Mixed Model Analyses of Available Cases and Multiple Imputation

Outcome	M1 Timepoint	M2 Timepoint	M1	M2	diffM1-M2	Lower CL.95	Upper CL.95	d	n	Imputed diffM1-M2	Imputed lower CL.95	Imputed upper CL.95	Imputed d
PHQ	T1	T0	10.38	13.62	−3.23***	−3.57	−2.90	0.50	330	−3.88***	−4.49	−3.28	0.60
	T3	T0	8.06	13.62	−5.56***	−5.96	−5.15	0.85	315	−5.96***	−6.57	−5.34	0.92
	T6	T0	8.88	13.62	−4.74***	−5.68	−3.80	0.73	92	−4.79***	−5.83	−3.75	0.74
GAD	T1	T0	9.24	11.89	−2.65***	−2.95	−2.34	0.47	333	−3.16***	−3.71	−2.61	0.56
	T3	T0	7.09	11.89	−4.80***	−5.17	−4.44	0.86	305	−5.27***	−5.80	−4.74	0.94
	T6	T0	7.78	11.89	−4.10***	−5.02	−3.18	0.73	72	−4.04***	−5.24	−2.84	0.72
PCL	T1	T0	15.85	17.99	−2.14***	−2.44	−1.83	0.38	327	−2.72***	−3.26	−2.18	0.48
	T3	T0	13.75	17.99	−4.24***	−4.60	−3.87	0.75	299	−4.53***	−5.08	−3.98	0.80
	T6	T0	14.55	17.99	−3.44***	−4.38	−2.50	0.61	69	−3.57***	−4.71	−2.42	0.63
SUI	T1	T0	0.34	0.58	−0.24***	−0.28	−0.20	0.28	334	−0.22***	−0.29	−0.16	0.26
	T3	T0	0.22	0.58	−0.36***	−0.41	−0.32	0.42	334	−0.36***	−0.42	−0.30	0.42
	T6	T0	0.26	0.58	−0.33***	−0.42	−0.23	0.39	105	−0.31***	−0.40	−0.22	0.36

T0 = baseline; T1 = 1-month assessment; T3 = 3-month assessment; T6 = 6-month assessment.

^*** p < 0.0001.

GAD, generalized anxiety disorder; PCL, PTSD checklist; PHQ, Patient Health Questionnaire; SUI, suicidal ideation.

See Supplementary Table S1 for an analysis of PCL, GAD, and PHQ outcomes in their respective clinical subgroups, that is, PHQ change in people with primary diagnosis of depression, PCL change in PTSD, etc. Effect sizes ranged from 0.55 to 1.20 and were generally larger than estimates from the overall sample.

See Supplementary Table S2 for a comparison of effects in those with versus without experience of psychiatric medications (past or present use). Effects were generally larger in those with experience of psychiatric medications (d's ranging from 0.43 to 0.98) than those without (d's ranging from 0.01 to 0.72).

Dose and other predictive factors

We entered dose as a predictive factor, controlling for age, gender, and site. Gender and route of administration (sublingual vs. intramuscular) were not significantly predictive of outcomes. Dose referred to the number of doses administered at the time of assessment and was a within-subjects time-varying factor. There was a significant effect of dose on PHQ, b = −1.20, standard error (se) = 0.061, p < 0.0001, on GAD, b = −0.97, se = 0.055, p < 0.0001, and on PCL, b = −0.92, se = 0.055, p < 0.0001. These coefficients indicate that per administered dose, there was a reduction of 1.20 points on the PHQ and ∼1 point reductions on the GAD and PCL.

There were medium-to-large site differences due to the requirement for clinician treatment referral in Canada. Overall, Canadian sites treated cases with more intense symptomatology than American sites (which permitted self-referral): d = 0.80, M_Diff = 5.19, CI_.95 (4.19 to 6.20) on PHQ; d = 0.43, M_Diff = 2.43, CI_.95 (1.58 to 3.28) on GAD; d = 0.58, M_Diff = 3.27, CI_.95 (2.41 to 4.14) on PCL; all p's < 0.0001.

There were some small age effects. Compared with younger clients (estimated at 30 years or mean age – 1 SD), older clients (estimated at 54 years or mean age +1 SD) had lower overall PHQ scores, d = 0.23, M_Diff = −1.53, CI_.95 (−2.09 to −0.98), p < 0.0001, lower overall GAD scores, d = 0.31, M_Diff = −1.73, CI_.95 (−2.20 to −1.25), p < 0.0001, and lower overall PCL scores, d = 0.14, M_Diff = −0.78, CI_.95 (−1.27 to −0.30), p = 0.002.

Caseness

There were statistically significant reductions in the proportions of depressed, anxious, and PTSD cases as identified by scoring above or equal to clinical cutoff values on each outcome measure (Table 3). Case reductions showed similar patterns of change across outcomes as in the growth curve models, with the largest case reductions at the primary 3-month endpoint.

Table 3.

Proportions Scoring Above Cutoff By Outcome and Timepoint

Outcome	Pr1 Timepoint	Pr2 Timepoint	Pr1	Pr2	diffPr1-Pr2	Lower CL.95	Upper CL.95	n
PHQ ≥15	T0	T1	0.52	0.27	0.25***	0.20	0.31	330
	T0	T3	0.56	0.17	0.39***	0.33	0.45	315
	T0	T6	0.51	0.16	0.35***	0.24	0.46	92
GAD ≥10	T0	T1	0.64	0.41	0.23***	0.17	0.29	333
	T0	T3	0.67	0.26	0.41***	0.35	0.47	305
	T0	T6	0.69	0.32	0.37***	0.23	0.52	72
PCL ≥14	T0	T1	0.79	0.59	0.20***	0.15	0.25	327
	T0	T3	0.84	0.45	0.39***	0.33	0.45	299
	T0	T6	0.87	0.58	0.29***	0.17	0.41	69
SUI ≥2	T0	T1	0.13	0.07	0.06**	0.03	0.10	334
	T0	T3	0.15	0.03	0.12***	0.08	0.16	334
	T0	T6	0.17	0.04	0.13**	0.06	0.20	105

T0 = baseline; T1 = 1-month assessment; T3 = 3-month assessment; T6 = 6-month assessment.

^** p < 0.001, ^***p < 0.0001.

In the 315 patients who provided a 3-month PHQ assessment, the depression case rate dropped from 56% at baseline to 17% at 3 months. In the 305 patients who provided a 3-month GAD assessment, the anxiety case rate dropped from 67% at baseline to 26% at 3 months. In the 299 patients who provided a 3-month PCL assessment, the PTSD case rate dropped from 84% at baseline to 45% at 3 months.

In the 92 patients who provided a 6-month PHQ assessment, the depression case rate dropped from 51% at baseline to 16% at 6 months. In the 72 patients who provided a 6-month GAD assessment, the anxiety case rate dropped from 69% at baseline to 32% at 6 months. In the 69 patients who provided a 6-month PCL assessment, the PTSD case rate dropped from 87% at baseline to 58% at 6 months.

Treatment responders

See Table 4. In the 315 patients who provided a 3-month PHQ assessment, 75% reported a 3-point MCID compared with baseline. In the 305 patients who provided a 3-month GAD assessment, 68% reported a 3-point MCID compared with baseline. In the 299 patients who provided a 3-month PCL assessment, 50% reported a 5-point MCID compared with baseline.

Table 4.

Proportions Reporting Minimal Clinically Important Differences at Endpoint Compared with Baseline

Outcome	Endpoint	Pr	Lower CL.95	Upper CL.95	n
PHQ	T1	0.53***	0.47	0.58	330
	T3	0.75***	0.70	0.79	315
	T6	0.70***	0.60	0.79	92
GAD	T1	0.50***	0.44	0.55	333
	T3	0.68***	0.62	0.73	305
	T6	0.65***	0.54	0.76	72
PCL	T1	0.32***	0.26	0.37	327
	T3	0.50***	0.44	0.56	299
	T6	0.48***	0.36	0.60	69

MCID = 3 points for PHQ and GAD. MCID = 5 points for PCL. T1 = 1-month assessment; T3 = 3-month assessment; T6 = 6-month assessment.

^*** p < 0.0001.

In the 92 patients who provided a 6-month PHQ assessment, 70% reported a 3-point MCID compared with baseline. In the 72 patients who provided a 6-month GAD assessment, 65% reported a 3-point MCID compared with baseline. In the 69 patients who provided a 6-month PCL assessment, 48% reported a 5-point MCID compared with baseline.

Adverse events

There were no serious adverse events (i.e., hospitalizations and deaths). Regarding SUI, there were significant reductions over time, with d-values ranging from 0.28 to 0.42 (Table 2). Using a cutoff value of 2 to indicate at least moderate ideation, there were significant reductions in individuals scoring above cutoff compared with baseline, with reductions ranging from 6% to 13% (Table 3). Concerning whether any persons increased in SUI compared with baseline, 7.5% (25/334) did at T1, 3.6% did at T3 (12/334), and 7.6% did at T6 (8/105).

Simulations

We multiply imputed 1000 data sets per outcome measure in which missing values were filled in with simulated values, achieving relative efficiencies of 0.999. This method allowed us to evaluate what would have happened if individuals who were lost to follow-up were retained in the data analysis and was meant to mitigate the poor retention rate. For each imputation, we specified linear mixed models to directly estimate mean differences at each endpoint from baseline. These parameters were combined across imputations to achieve best estimates of mean differences (Table 2). There were no apparent differences between imputed estimates and those based on the analysis of available cases only. Many imputed estimates indicated a slightly larger effect than initially detected.

Discussion

This effectiveness study examined the largest data set to date of long-term outcomes in clients who received KAP. The intention of KAP is to alleviate symptoms of depression, anxiety, and post-traumatic stress in patients seeking treatment. Among patients who provided follow-up data, we found evidence of large to moderately large treatment effects at 3 and 6 months, with d's ranging from 0.61 to 0.86 and corresponding to 5–6 point reductions on the PHQ, 4–5 point reductions on GAD, and 3–4 point reductions on the PCL.

There was a consistent pattern of change across measures characterized by reductions in psychological distress that was most apparent at 3 months. There were ∼40% reductions in caseness for depression, anxiety, and PTSD at this primary endpoint, and treatment responders ranged from 50% to 75%. This clinical benefit was sustained at 6 months in those who provided follow-up data.

Age differences in outcomes were small indicating the broad suitability of KAP. Over 90% of individuals entered treatment within 3 months of intake, and 76% completed more than one ketamine dose. Each ketamine dose was associated with 1-point reductions on the PHQ, GAD, and PCL, in support of the current program, which recommends 4–6 ketamine sessions plus 4–6 psychotherapeutic integration sessions. This may produce an estimated 4–6 point reduction across outcome measures over time, larger than the MCID thresholds of 3–5 points for indicating treatment response.

These effects must be interpreted within the context of KAP, which involves integration sessions occurring in lock step with ketamine dosing to activate and reinforce psychological gains. Occasional side effects included nausea, vomiting, and increases in blood pressure, not requiring medical intervention nor requiring patients to be taken to a hospital. Findings may be most generalizable to populations with treatment-resistant depression, anxiety, or PTSD eligible for KAP. Given the lack of novel treatment options in the behavioral health space, KAP represents a viable alternative that should be more widely considered.

Limitations

The main limitation of this study is the substantial attrition during the 6-month follow-up period. The importance of completing the follow-up assessments was not made clear to patients and increased efforts are needed to improve questionnaire completion rates moving forward. The treatment effects demonstrated here are based on those who provided follow-up data. These individuals tended to have more pretreatment distress and may be considered a clinical priority. The estimates of treatment effect at 6 months are less reliable than at 3 months. To the extent possible, we examined these issues using multiple imputation of missing data. The estimates of treatment effect from these simulation analyses were stable and consistent with the analysis of only available cases.

Other limitations include the retrospective nature of the study and lack of a randomized control group. Although there were no serious adverse events, discrete data about side effects (e.g., nausea, vomiting) were not available and there was a lack of assessment of additional psychotherapeutic interventions and psychotropic medications. The findings are promising but a prospective clinical trial with improved data collection is needed for validation. Future study should also clarify the indications and contraindications of KAP for clinical subpopulations and the tailoring of treatment to improve individual outcomes.

Conclusion

In this retrospective effectiveness study, 346 and 94 participants completed assessments at 3 and 6 months, respectively, representing loss to follow-up rates of 82% and 95%. There were large effect sizes on outcomes at 3 months that were sustained at 6 months. The high rates of attrition may limit validity of the results. The data suggest that KAP is an effective treatment achieving clinically meaningful reductions in depression, anxiety, and post-traumatic stress for up to half a year.

Authors' Contributions

R.Y. and C.L. had full access to all of the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Concept and design were carried out by R.Y., M.V., and C.L. Acquisition, analysis, or interpretation of data was done by all authors excluding D.M. Drafting of the article was by all authors. Critical revision of the article for important intellectual content was carried out by all authors. Statistical analysis was done by V.S., J.B., R.Y., and C.L. Funding was obtained by R.Y. and C.L. Administrative, technical, or material support was by V.S., N.Z., and J.B. Supervision was done by R.Y. and C.L.

Data Sharing Statement

There is no plan to share data, although requests may be made for access.

Author Disclosure Statement

R.Y. was a cofounder of Field Trip Health and formerly served as chief clinical officer. He received a salary and stock options for his work in this regard. J.B. is the former manager of Health Platforms at Field Trip Health. He received a salary and stock options for his work in this regard. N.B. was chief scientific officer of Field Trip Discovery, formerly a division of Field Trip Health, but now known as Reunion Neuroscience, and he continues to own shares in Field Trip Health and Wellness, and further, as chief scientific officer of Reunion Neuroscience, has an on-going mutual service agreement with Field Trip Health and Wellness.

M.E. and J.A.D.L. received consultancy fees for the development of the clinical treatment manual and program for Field Trip Health but do not own shares in this company. J.A.D.L. was the former director of Clinical Therapy for Field Trip Health Toronto. He is the founder and clinical director at the Centre for Compassionate Care in Hamilton, Ontario. M.E. holds shares in Sintalica Corp. E.H. received a salary for her work. M.K. and M.M. worked at Field Trip Health and received a salary and stock options.

S.K. has received funding for consulting or speaking engagements from Abbvie, Boehringer-Ingelheim, Janssen, Lundbeck, Lundbeck Institute, Merck, Otsuka Pfizer, Sunovion, and Servier. He has received research support from Abbott, Brain Canada, CIHR (Canadian Institutes of Health Research), Janssen, Lundbeck, Neurocrine, Ontario Brain Institute, Otsuka, Pfizer, and SPOR (Canada's Strategy for Patient-Oriented Research). He has stock/stock options in Field Trip Health. S.K. has no disclosures to report.

R.M. was an advisor to Field Trip Health, and owns shares in Field Trip Health and Wellness and Reunion Neuroscience. B.M. was the medical director of Field Trip Health. He is the current medical director of Nue.Life Medical Group that provided medical and therapy services for Field Trip At Home. He received salary and stock options for all of these from both Field Trip and Nue.Life.

D.M. was formerly chief psychologist (USA) and vice president (global) of Therapeutic Growth and Innovation and later advisor for Field Trip Health; she previously received salary and owned stock/stock options in Field Trip Health and owns stock in Reunion Neuroscience and Numinus. She is currently in paid contract consulting and/or teaching/training roles with Journey Clinical, Fluence, Psychedelics Today, CIIS, and The MIND Foundation.

M.M. was medical director of Field Trip Health's Atlanta clinic and received a salary for his work there. He also owns stock options in Field Trip Health. S.P. is the former director of Psychotherapy Services at Field Trip Health and Wellness. She received a salary and stock options for her work in this regard. V.S. was the director of data at Field Trip Health. He received a salary and stock options for his work in this regard.

M.V. was the former medical director of Field Trip Health and Wellness and he received a salary and stock options for his work in this regard. E.W. is the former vice president of Clinical Services at Field Trip Health and Wellness. She received a salary and stock options for her work in this regard. N.Z. was a clinical consultant at Field Trip Health and received a salary for his work. C.L. received an initial consultation fee from Field Trip Health.

Funding Information

Employees of Field Trip Health were involved in the design and conduct of the study; collection, management, analysis, and interpretation of the data; preparation, review, or approval of the article; and decision to submit the article for publication.