Veterans Self-Reported Reasons for Non-Attendance in Psychotherapy for Posttraumatic Stress Disorder

Conflicts in Iraq and Afghanistan have resulted in an increased focus on posttraumatic stress disorder (PTSD) among U.S. veterans. PTSD is a distressing and interfering mental health condition associated with higher rates of unhealthy alcohol use, suicidality, and poorer physical health (Ramsawh et al., 2014; Schnurr, Wachen, Green, & Kaltman, 2014; Thomas et al., 2010). Approximately 11% of U.S. veterans enrolled in the Veterans Health Administration (VHA) have been diagnosed with PTSD, with prevalence estimated to be more than twice as high among veterans returning from Iraq and Afghanistan (26.7%; Greenberg & Hoff, 2016; Harpez-Rotem & Hoff, 2015).

In an effort to better address the treatment needs of individuals diagnosed with PTSD, organizations such as the Department of Veterans Affairs (VA) and Department of Defense (DoD) have identified trauma-focused psychotherapies as front-line interventions for PTSD (Institute of Medicine, 2008; Veterans Health Administration, Department of Defense, 2017). Trauma-focused psychotherapies include components of exposure (recounting and processing trauma-related memories and exposure to trauma-related situations/reminders) and/or cognitive restructuring (identification and modification of trauma-related beliefs and meanings). Two of these trauma-focused cognitive–behavioral therapies, cognitive processing therapy (CPT) and prolonged exposure (PE), have been the focus of extensive dissemination and implementation efforts by the VHA (Karlin & Cross, 2014; Karlin et al., 2010).

Despite evidence for the effectiveness of CPT and PE in veteran samples (Alvarez et al., 2011; Eftekhari et al., 2013; Monson et al., 2006; Tuerk et al., 2011), concerns have been raised about low rates of initiation (Grubbs et al., 2015; Kehle-Forbes, Meis, Spoont, & Polusny, 2016; Mott et al., 2014), inconsistent attendance (Tarrier, Sommerfield, Pilgrim, & Faragher, 2000), and early treatment discontinuation (i.e., discontinuing treatment before receiving a recommended dose of care; Kehle-Forbes et al., 2016; Mott et al., 2014; Najavits, 2015; Rutt, Oehlert, Krieshok, & Lichtenberg, 2018; Shiner et al., 2013) in these interventions. For instance, research suggests that only a small minority of veterans initiate trauma-focused treatment. In one study of more than 1,900 veterans receiving care across six VHA facilities, just 6% attended at least one trauma-focused therapy session in the first 6 months of treatment in a VHA PTSD clinic (Shiner et al., 2013). Veterans who successfully initiate a trauma-focused psychotherapy then need to remain in care long enough to receive an adequate “dose” of treatment, as fewer missed therapy sessions and treatment completion have been associated with improved treatment outcomes (e.g., Rutt et al., 2018; Tarrier et al., 2000). However, many veterans who begin CPT or PE discontinue care prior to receiving a full course of treatment (DeViva, Sheerin, et al., 2016; Imel, Laska, Jakupcak, & Simpson, 2013; Kehle-Forbes et al., 2016; Mott et al., 2014; Najavits, 2015; Steenkamp & Litz, 2013). For instance, a study of 427 male and female veterans found that half of the veterans who began CPT or PE discontinued care prior to completing treatment (Kehle-Forbes et al., 2016), while a review of PE and CPT treatment outcome studies in veteran and military populations found rates of early treatment discontinuation ranging from 12% to 39% (Steenkamp & Litz, 2013). Similar rates of discontinuation have been observed when comparing veterans participating in PTSD treatment via telemedicine to veterans attending psychotherapy in-person, suggesting telemedicine delivery is neither a consistent deterrent nor a consistent facilitator of better attendance (Turgoose, Ashwick, & Murphy, 2018).

Given the relatively low rates of treatment initiation and completion, efforts have been made to identify treatment and patient characteristics that predict non-initiation and/or early discontinuation from PTSD psychotherapy. For treatment characteristics, group modality and longer treatment protocols predict increased dropout (Imel et al., 2013). For patient characteristics, benzodiazepine use, alcohol use, younger age, lower intelligence test scores, and less education have been associated with early discontinuation in CPT and PE (Rizvi, Vogt, & Resick, 2009; van Minnen, Arntz, & Keijsers, 2002). Among veterans specifically, Iraq/Afghanistan service era, younger age, higher clinical severity (e.g., depression or PTSD symptoms), previous psychiatric inpatient hospitalization, psychiatric medication prescription, PTSD service connection, and higher levels of service connected disability have been associated with non-initiation of care and/or early discontinuation (DeViva, Bassett, Santoro, & Fenton, 2017; Grubbs et al., 2015; Harpaz-Rotem & Rosenheck, 2011; Hundt et al., 2018; Kehle-Forbes et al., 2016; Mott et al., 2014).

While the existing research literature has contributed to our understanding of treatment and patient characteristics that may identify veterans at greater risk for non-initiation or early discontinuation from PTSD interventions, these findings have not provided information about the veteran perspective or reasons for not initiating, regularly attending, or completing treatment. A limited amount of research with active duty service members has begun to examine patient identified reasons for early treatment discontinuation from either PTSD treatment or general mental health care (Hoge et al., 2014; Jennings, Zinzow, Britt, Cheung, & Pury, 2016). The most common reasons identified for early treatment discontinuation have included self-reliance (i.e., service members feeling that they could take care of problems on their own), being too busy with work, stigma (e.g., a concern that unit members or leaders might treat them differently or lose confidence in them), and a belief that treatment was not working (Hoge et al., 2014; Jennings et al., 2016). Unfortunately, it is not known how generalizable such findings are to veteran populations. Given the unique help seeking challenges associated with military service as well as those associated with transitioning to civilian life after service, veterans may or may not have the same concerns as active duty soldiers.

To our knowledge, just two qualitative studies examining veterans’ beliefs about initiating treatment have been completed. Sayer and colleagues (2009) examined determinants of PTSD treatment initiation in a sample of treatment seeking and nontreatment seeking veterans who recently submitted disability claims for military-related PTSD (n = 44). Identified barriers to treatment included avoidance of trauma-related emotions and memories, values or priorities that conflict with seeking care (e.g., self-reliance), beliefs that discourage treatment seeking (e.g., treatment is not helpful), concerns regarding the health care system (e.g., VHA and VHA providers cannot be trusted), knowledge-related barriers (e.g., little information/understanding regarding PTSD symptoms or available services), access to care (e.g., time constraints, expense of treatment) and barriers stemming from an invalidating sociocultural environment following trauma (e.g., negative homecoming experiences). More recently, Hundt and colleagues (2018) examined veterans’ reasons for choosing not to initiate trauma-focused psychotherapies for PTSD (n = 24). The most commonly endorsed barriers were related to the VHA system, including inefficiencies and delays (e.g., lack of follow-up, “red tape”), negative experiences with VHA staff and providers, discomfort with the VHA environment (e.g., experience of the VHA as a “nontherapeutic/unsafe environment”), difficulty navigating the VHA system and lack of evening/weekend appointment availability. It remains unknown if identified reasons for non-initiation are just as relevant for understanding inconsistent attendance or early discontinuation in trauma-focused therapies. Furthermore, reasons for non-attendance have not been compared directly between trauma-focused and non-trauma-focused therapies. Trauma-focused psychotherapies appear to have higher rates of early discontinuation compared to other forms of therapy (e.g., CBT for anxiety [DeViva, 2014], present-centered therapy [Imel et al., 2013]), which suggests there may be unique barriers that are therapy specific.

Understanding veterans’ perspectives on why they choose not to initiate, regularly attend, or complete trauma-focused psychotherapies, and whether such reasons differ from other forms of individual and group psychotherapy offered to veterans with PTSD, is critical for optimizing engagement in recommended evidence-based PTSD treatments. Questions remain about whether CPT or PE need to be adapted for greater acceptability among veterans or whether engagement would be enhanced by adjunctive or preparatory interventions, such as those with additional focus on motivation or coping skills (Cook, Simiola, Hamblen, Bernardy, & Schnurr, 2017; Hamblen et al., 2015; Landes, Garovoy, & Burkman, 2013; Zubkoff, Carpenter-Song, Shiner, Ronconi, & Watts, 2016). To our knowledge, no published studies have explored veterans’ self-reported reasons for choosing not to initiate, regularly attend, or complete trauma-focused psychotherapy as compared to nontrauma-focused psychotherapies, which has limited our understanding of whether modification is needed. The present study begins this work by characterizing veterans’ self-reported reasons for non-initiation, inconsistent attendance, and early discontinuation in CPT and other forms of available PTSD-focused individual and group psychotherapy in veterans participating in the Telemedicine Outreach for PTSD (TOP) Study (Fortney et al., 2015). The aims were to (a) explore rates of initiation, attendance, early discontinuation, and completion in CPT and other forms of available PTSD-focused individual and group psychotherapy and (b) characterize veterans’ self-reported reasons for non-initiation, inconsistent attendance, and early discontinuation in scheduled CPT sessions as compared to scheduled sessions in these other forms of available PTSD-focused individual and group psychotherapy.

Method

TOP Study

The TOP study, a pragmatic randomized effectiveness trial, was designed to evaluate a telemedicine-based collaborative care intervention for PTSD as compared to usual care in 11 VHA community-based outpatient clinics (CBOCs) located in the south central and western United States (Fortney et al., 2015). Veterans were recruited via provider or self-referral and opt-out cards. To be included, veterans had to meet diagnostic criteria for PTSD according to the Clinician-Administered PTSD Scale (Blake et al., 1995). Veterans already receiving specialty PTSD treatment at a VHA medical center, those diagnosed with schizophrenia, bipolar disorder, substance dependence with active use, hearing impairment, or a life-threatening illness, individuals without a telephone, and/or those lacking the capacity to consent were excluded from the trial. A total of 265 eligible veteran participants enrolled in care across the 11 participating CBOCs completed baseline assessments and were randomized to the TOP study intervention or usual care. The TOP collaborative care intervention was designed to improve access to and engagement in CPT and pharmacotherapy for rural veterans using telemedicine technologies (e.g., telephone, interactive video, and shared electronic medical records) to support treatment delivered by medical providers located at VHA CBOCs. Veterans in both the TOP intervention and usual care conditions were eligible to receive any services offered at their local CBOC or distant VHA medical center. As a result, both CPT and other forms of PTSD-focused individual and group psychotherapy were available to and used by veterans in the both the TOP collaborative care intervention and the usual care group as is described below. For veterans in the intervention group, a care manager encouraged adherence to counseling and medications. Veterans in the usual care condition were not referred to any particular treatment by the research team. For a list of all interventions available to TOP intervention and usual care participants see Table 1 in the online supplemental materials. For additional details on the TOP study see Fortney et al., 2015.

Present Study

The present study used baseline and 6-month follow-up data collected as part of the TOP Study (Fortney et al., 2015). Follow-up telephone interview assessments were completed for 86.8% of participants at 6 months (n = 230). Among the 230 veterans with a completed 6-month follow-up, 176 (76.5%) had a scheduled appointment for CPT and/or PTSD-focused individual or group psychotherapy and were included in the present study. In the present study participants were categorized into groups based on their participation in available forms of psychotherapy (i.e., CPT and/or other PTSD-focused individual and/or group psychotherapy) regardless of their randomization status in the TOP Study (see Procedures). For example, veterans participating in CPT were included in the CPT group regardless of whether the veteran participated in CPT as a recipient of the TOP study intervention or engaged in this treatment while participating in the usual care condition.

Interventions Available to TOP Trial Participants

Cognitive processing therapy (CPT)

CPT is a manualized psychotherapy that has been shown to significantly reduce PTSD symptoms among veterans (Monson et al., 2006). It typically consists of 12 sessions focused on providing psychoeducation about PTSD, using written exposure about the traumatic event, challenging problematic beliefs (“stuck points”), and examining beliefs across a number of areas that could be negatively impacted by the experience of trauma (e.g., safety, trust, self-esteem). In the first 6 months of this study, individual CPT was delivered via interactive video by clinical psychologists located at a parent VHA medical center to 51 veterans located at a CBOC in the context of the TOP collaborative care intervention. CPT was also delivered in-person to nine veterans in the usual care group through the CBOC or parent VHA medical center during the same time period (Grubbs et al., 2017). Thus, while the majority of veterans were scheduled to receive CPT via telehealth as part of the TOP study intervention (85%), a portion of individuals were scheduled for in-person CPT appointments as part of usual care (15%). A veteran could elect to begin CPT at any time after the TOP trial baseline assessment. Group CPT was also available, but because so few veterans scheduled an appointment or attended a session (n = 2), we did not examine reasons for non-attendance in CPT groups.

Other PTSD-focused individual and group psychotherapy

Other forms of PTSD-focused individual and group psychotherapy were available and offered in-person to veterans in the both the intervention and usual care groups through a CBOC or the parent VHA medical center. A previously published article reporting the results of a chart review of care received by TOP trial participants in the usual care condition found that, outside of CPT, no other evidence-based treatments for PTSD (PE or eye movement desensitization and reprocessing [EMDR]) were received by participants (Grubbs et al., 2017). However, many veterans received other forms of PTSD-focused individual and group psychotherapy that consisted of a variety of structured and unstructured interventions with at least some PTSD and/or trauma-related content (e.g., PTSD psychoeducation, process group focused on trauma). In the usual care arm, patients had an average of 7.7 PTSD-focused psychotherapy sessions a year, suggesting somewhere between a monthly and bimonthly schedule on average (Grubbs et al., 2017). Veterans may have initiated these other forms of PTSD-focused psychotherapy prior to joining the TOP trial. As a result, the first session of other PTSD-focused individual and group psychotherapy appointments may have occurred before or after the TOP baseline assessment.

Procedures

Research assistants, blinded to study intervention condition, conducted telephone interviews at baseline and 6-month follow-up using an interview instrument specifically designed for the TOP study. Research assistants collected demographic data, queried about treatment history and acceptability, assessed PTSD symptom severity and psychiatric comorbidities, and inquired about health-related quality of life and social support. At 6-month follow-up, research assistants also collected information regarding self-reported reasons for not attending and/or discontinuing treatment when relevant. Specifically, prior to conducting the telephone-based interview, a study investigator (TJH) determined if a veteran had been scheduled for individual CPT or another form of PTSD-focused individual or group psychotherapy during the 6-month follow-up period using an electronic medical record review method developed for the TOP study. Determinations regarding type of visit were made based on provider name, clinic name and, if available (i.e., veteran attended scheduled visit), review of available psychotherapy notes. Only a limited number of providers delivered CPT at the CBOCs included in the TOP study and special clinic names were often used to denote CPT appointments. Study staff maintained a list of provider and clinic names throughout the study to facilitate identification of these appointments.

For the purposes of the present study, each participant was classified into one of five categories related to his or her attendance in CPT and/or another form of PTSD-focused individual or group psychotherapy and his or her responses to interview questions: (a) Completed, (b) Still Attending, (c) Inconsistent Treatment Attendance, (d) Stopped Attending, (e) Did Not Attend. More specifically, if review of the electronic medical recorded indicated the veteran participant had been scheduled for a psychotherapy visit during the past 6 months, the veteran was asked if they were still attending those appointments. Participants indicating they had attended appointments in the past 6 months but had since completed psychotherapy were categorized in the Completed group. Participants reporting continued psychotherapy attendance were asked how many appointments were scheduled in the past 4 weeks and how many sessions they attended. Those attending ≥75% of their treatment sessions were categorized in the Still Attending group. Those attending <75% of sessions were categorized in the Inconsistent Attendance group. Participants reporting previous but not current psychotherapy attendance were categorized in the Stopped Attending group (i.e., early treatment discontinuation). Participants reporting they did not attend their scheduled psychotherapy appointment or any other scheduled appointment in the past 6 months were categorized in the Did Not Attend group.

Measures

Reasons for non-attendance

Participants classified into the Inconsistent Attendance, Stopped Attending, and Did Not Attend groups were read a list of possible reasons for non-attendance informed by clinical experience (see Figures 4-6). Participants were allowed to endorse multiple reasons. Following review of this list, participants were asked if there were any other reasons they decided not to attend care and their response was recorded and categorized.

Figure 4.

Veterans’ self-reported reasons for non-attendance in individual CPT (n = 23)

*Veteran provided reason.

Figure 6.

Veterans self-reported reasons for non-attendance in other PTSD-focused group psychotherapy (n = 14)

*Veteran provided reason.

Additional measures used to characterize the sample

Research assistants collected demographic data including: age, gender, race/ethnicity, income, marital status, education, employment, military service era, combat trauma history, and service connection status. Information regarding participants’ treatment history and acceptability of pharmacotherapy and psychotherapy were also collected. PTSD diagnostic status and symptom severity was assessed using the Clinician-Administered PTSD Scale and the PTSD Diagnostic Scale (Blake et al., 1995; Weathers, Ruscio, & Keane, 1999). The presence of comorbid depression, panic disorder and generalized anxiety disorder was evaluated using the Mini-International Neuropsychiatric Interview (Sheehan et al., 1998). The Alcohol Use Disorder Identification Test was used to measure alcohol misuse (Babor, Higgins-Biddle, Saunders, & Monteiro, 2001). Health-related quality of life was measured using the Mental Component Summary and Physical Component Summary scores of the 12-item Short Form Health Survey for veterans (Kazis et al., 1998). Social support was assessed using the Medical Outcomes Study Social Support Scale (Sherbourne & Stewart, 1991).

Analysis

Descriptive statistics were used to characterize the sample and to examine the frequency of self-reported reasons for non-attendance. Quantitative differences between veterans scheduled for CPT and other forms of PTSD-focused individual or group psychotherapy were examined to identify potentially clinically meaningful differences. Statistical tests were not calculated given that groups were not mutually exclusive.

Results

Participants

Table 1 presents sociodemographic and baseline clinical characteristics of the full TOP trial sample and veterans scheduled for individual CPT and/or other forms of PTSD-focused individual or group psychotherapy during the 6-month follow-up period. Potentially clinically meaningful differences were observed between groups. For instance, compared to veterans scheduled for CPT, veterans scheduled for other forms of PTSD-focused individual therapy tended to have a greater percentage of participants who identified as Caucasian, combat exposed, and service connected, as well as a greater percentage reporting prior use of psychotherapy, medication, and/or PTSD-specific treatment. Veterans scheduled for other forms of PTSD-focused individual psychotherapy also tended to have a greater percentage of veterans who met criteria for current major depression. Conversely, veterans scheduled for CPT tended to have higher rates of reported household income greater than $20,000 when compared to veterans scheduled for other forms of PTSD-focused individual psychotherapy.

Table 1.

Sociodemographic and clinical characteristics

	All (N=265)	Individual CPT (N=60)	PTSD Individual (N=89)	PTSD Group (N=58)
Variable	N (%)	N (%)	N (%)	N (%)
Age, mean (SD)	52.18 (13.79)	49.22 (14.89)	49.21 (13.73)	56.29 (11.26)
Male sex	238 (89.81%)	51 (85.00%)	77 (86.52)	55 (94.83%)
Race
White	169 (63.77%)	35 (58.33%)	56 (62.92%)	43 (74.14%)
African American	52 (19.62%)	17 (28.33%)	20 (22.47%)	6 (10.34%)
Hispanic	20 (7.55%)	5 (8.33%)	8 (8.99%)	3 (5.17%)
Other	24 (9.06%)	3 (5.00%)	5 (5.62%)	6 (10.34%)
Annual household income <$20,000	67 (25.28%)	18 (30.00%)	15 (16.85%)	10 (17.24%)
Married	185 (69.81%)	39 (65.00%)	60 (67.42%)	47 (81.03%)
High school graduate	249 (93.96%)	57 (97.00%	87 (97.75%)	55 (94.83%)
Employed	68 (25.66%)	17 (28.33%)	23 (25.84%)	13 (22.41%)
Period of wartime service
OEF/OIF	77 (29.06%)	22 (36.67%)	32 (35.96%)	28 (48.28%)
Other	188 (70.94%)	38 (63.33%)	57 (64.04%)	30 (51.72%)
Combat	131 (49.43%)	27 (45.00%)	48 (53.93%)	28 (48.28%)
Service connection
Never Applied	16 (6.04%)	4 (6.67%)	3 (3.37%)	1 (1.72%)
Applied, denied	16 (6.04%)	2 (3.33%)	6 (6.74%)	3 (5.17%)
Applied, pending	28 (10.57%)	9 (15.00%)	9 (10.11%)	5 (8.62%)
Approved	201 (75.85%)	44 (73.33%)	70 (78.65%)	49 (84.48%)
PTSD service connection
Never Applied	41 (15.77%)	10 (16.67%)	6 (6.82%)	7 (12.07%)
Applied, denied	25 (9.62%)	5 (8.33%)	7 (7.95%)	3 (5.17%)
Applied, pending	60 (23.08%)	22 (36.67%)	15 (17.05%)	11 (18.97%)
Approved	132 (50.77%)	23 (38.33%)	59 (67.05%)	34 (58.62%)
Psychotropic mediations acceptable
Definitely	119 (44.91%)	34 (56.67%)	51 (57.30%)	36 (62.07%)
Probably	91 (34.34%)	20 (33.33%)	24 (26.97%)	11 (18.97%)
Probably not	24 (9.06%)	4 (6.67%)	9 (10.11%)	8 (13.79%)
Definitely not	31 (11.70%)	2 (3.33%)	5 (5.62%)	3 (5.17%)
Individual psychotherapy acceptable
Definitely	167 (63.74%)	39 (65.00%)	62 (69.66%)	42 (72.41%)
Probably	73 (27.86%)	18 (30.00%)	23 (25.84%)	13 (22.41%)
Probably not	20 (7.63%)	3 (5.00%)	3 (3.37%)	3 (5.17%)
Definitely not	2 (.76%)	--	1 (1.12%)	--
Prior use of psychotherapy	241 (90.9%)	53 (88.33%)	82 (92.13%)	52 (89.66%)
Prior use of medication	238 (89.81%)	50 (83.33%)	83 (93.26%)	49 (84.48%)
Prior PTSD-specific treatment	207 (78.11%)	45 (75.00%)	72 (80.90%)	47 (81.03%)
Current major depressive disorder	209 (78.87%)	46 (76.67%)	74 (83.15%)	42 (72.41%)
Current panic disorder	117 (44.15%)	31 (51.67%)	46 (51.69%)	28 (48.28%)
Current generalized anxiety disorder	178 (67.17%)	39 (65.00%)	59 (66.29%)	37 (63.79%)
AUDIT treatment recommendation
Alcohol education (1)	202 (77.69%)	43 (75.44%)	64 (72.73%)	45 (78.95%)
Simple Advice (2)	31 (11.92%)	9 (15.79%)	15 (17.05%)	7 (12.28%)
Brief counseling and continued monitoring (3)	10 (3.85%)	2 (3.51%)	2 (2.27%)	3 (5.26%)
Referral to specialist (4)	17 (6.54%)	3 (5.26%)	7 (7.95%)	2 (3.51%)
Social support score, mean (SD)	3.50 (1.00)	3.46 (1.02)	3.48 (.94)	3.56 (.97)
PTSD Severity
CAPS score, mean (SD)	74.98 (12.70)	74.7 (11.56)	73.85 (13.76)	58.00 (74.93)
PDS score, mean (SD)	34.23 (8.09)	34.75 (7.70)	34.33 (7.78)	33.24 (7.74)
PCS score, mean (SD)	35.02 (12.80)	37.08 (13.72)	35.18 (13.40)	35.21 (12.37)
MCS score, mean (SD)	32.75 (10.34)	32.97 (11.73)	33.06 (10.28)	33.61 (9.44)
No. of chronic physical illnesses, mean (SD)	4.31 (2.34)	3.72 (2.39)	4.09 (2.35)	4.29 (2.62)

Note. AUDIT = Alcohol Use Disorder Identification Test; CAPS = Clinician Administered PTSD Scale; PCS = Physical Component Summary (PCS) score of the 12-Item Short Form Health Survey for Veterans; PDS = Posttraumatic Diagnostic Scale; MCS = Mental Component Summary (MCS) score of the Short Form Health Survey for Veterans.

Sociodemographic and Clinical Characteristics

Compared to veterans scheduled for CPT, veterans scheduled for other forms of PTSD-focused group psychotherapy tended to be older and tended to have a greater percentage of veterans who identified as male, Caucasian, married, and/or as an Operation Enduring Freedom/Operation Iraqi Freedom era service member. Veterans scheduled for other forms of PTSD-focused group psychotherapy also tended to have higher rates of service connection and prior use of medication or PTSD specific treatment. Conversely, veterans scheduled for CPT tended to have higher rates of employment, reported household income greater than $20,000, and had greater Clinician-Administered PTSD Scale scores when compared to veterans scheduled for other forms of PTSD-focused group psychotherapy.

Treatment Attendance

Over 90% of veterans reported attending at least one psychotherapy appointment across treatments examined. Sixty participants were scheduled for a CPT session according to the electronic medical record, 89 participants were scheduled for another form of PTSD-focused individual psychotherapy, and 58 participants were scheduled for another form of PTSD-focused group psychotherapy. For respective rates of attendance, see Figures 1-3.

Figure 1.

Attendance in individual Cognitive Processing Therapy (CPT) six months after study enrollment

Figure 3.

Attendance in other PTSD-focused group psychotherapy six months after study enrollment

Veteran Self-Reported Reasons for Non-Attendance

CPT

Overall, the most common reasons given for non-attendance in individual CPT were (a) having CPT appointment(s) that conflicted with work, school, or other medical care (35%); (b) thinking that CPT would not or did not help (26%); and (3) not having reliable transportation (17%; see Figure 4). One (4.3%) veteran in the intervention group receiving CPT indicated that non-attendance was related to the interactive video technology.

Other PTSD-focused individual psychotherapy

In addition to not endorsing any reason for non-attendance (39%), the most common reasons given for non-attendance in other PTSD-focused individual psychotherapy were (a) problems with scheduling an appointment (28%) and (b) reporting that counseling took too much time (17%; see Figure 5).

Figure 5.

Veterans self-reported reasons for non-attendance in other PTSD-focused individual psychotherapy (n = 18)

*Veteran provided reason.

Other PTSD-focused group psychotherapy

Overall, the most common reasons given for non-attendance in other PTSD-focused group psychotherapy were (a) not being able to afford counseling (e.g., copays, transportation; 21%) and (b) appointment(s) conflicting with work, school, and/or other medical care (21%; see Figure 6).

Discussion

Concerns have frequently been raised regarding low rates of initiation and completion of trauma-focused psychotherapies (e.g., CPT) in veteran populations, yet little is known about why veterans decide not to initiate, consistently attend, or complete these evidence-based interventions and whether such reasons differ from reasons reported by veterans engaging in other forms of psychotherapy for PTSD. The present study examined rates of initiation, attendance, and completion in CPT, offered predominantly via interactive video to a VHA CBOC, and other forms of available PTSD-focused individual and group psychotherapy offered in person at a VHA CBOC or medical center and characterized veterans’ self-reported reasons for non-attendance in these forms of treatment. To the best of our knowledge, this is the first study to examine self-reported reasons for non-attendance in both trauma-focused psychotherapy and nontrauma-focused psychotherapy in a veteran sample.

Rates of attendance in at least one appointment during the 6-month follow-up period were high in CPT and in other forms of PTSD-focused individual and group psychotherapy, with over 90% of veterans reporting that they attended at least one appointment. The rate of CPT completion over the 6-month period (25%) was higher than completion of both PTSD-focused individual psychotherapy (4.4%) and PTSD-focused group psychotherapy (15.5%). This difference was observed despite the fact that all CPT treatment was initiated after baseline whereas PTSD-focused individual and group treatments may have been initiated prior to baseline, indicating treatment duration could have been even longer in these other forms of treatment. However, CPT typically involves weekly psychotherapy visits. PTSD-focused individual and group treatments may have been scheduled less frequently (e.g., monthly; Grubbs et al., 2017) and may have taken longer to complete.

Previously reported rates of completion of evidence-based psychotherapies for PTSD have varied greatly depending on the definition of completion used, making comparison to existing literature somewhat difficult (Szafranski, Smith, Gros, & Resick, 2017). For instance, completion rates of over 70% were reported in a sample of veterans completing PE in person or via telehealth when completion was defined as attendance of 12 sessions (Gros, Yoder, Tuerk, Lozano, & Acierno, 2011). When treatment completion also included clinician confirmation that treatment goals were achieved, completion rates in PE or cognitive therapy for PTSD were closer to 30% in a sample of Iraq and Afghanistan veterans (Garcia, Kelley, Rentz, & Lee, 2011). Lower rates of CPT completion seen in the present study may be due in part to the relatively short window in which attendance was examined (i.e., 6-month follow-up period) and the fact that CPT initiation could have ranged from shortly after baseline up to shortly before the 6-month follow-up interview.

Rates of inconsistent attendance (13.3%) and discontinuation of CPT (i.e., Stopped Attending, 18.3%) detected in the present study fall within previously reported ranges of early discontinuation from evidence-based psychotherapies for PTSD (i.e., 12% to 39%; Steenkamp & Litz, 2013), however, these rates were higher than those detected in veterans participating in other forms of PTSD-focused individual and group psychotherapy in the present study. The largest differences were seen related to inconsistent treatment attendance in CPT (13.3%) as compared to both PTSD-focused individual psychotherapy (2.2%) and PTSD-focused group psychotherapy (6.9%). Again, this may be due to the more frequent appointment scheduling for CPT (i.e., weekly) compared to PTSD-focused psychotherapy. In contrast to CPT, over 60% of veterans in other forms of PTSD-focused individual or group psychotherapy reported they were still attending treatment and some may have been attending psychotherapy for years. Thus, our sample may have included patients in ongoing PTSD-focused individual and group therapy who had been able to address (and overcome) attendance barriers in the past. In contrast, our sample of patients initiating CPT may have been facing newly emerging barriers.

Whether inconsistent attendance and early discontinuation differentially impacts treatment outcomes in different types of PTSD treatment (e.g., trauma-focused vs. other forms of PTSD focused psychotherapy) or treatment modalities (e.g., group vs. individual) remains an empirical question. For instance, in individual psychotherapy, missing a session generally results in the content of that session being delayed until the next appointment. In contrast, missing a group psychotherapy session often results in the patient missing the content of that session, which may differentially impact an individual’s decision to continue treatment.

Of note, it was difficult to ascertain the degree to which continued attendance in other forms of individual and group PTSD-focused psychotherapy offered clinical benefit. As previously noted, outside of CPT, no other evidence-based treatments for PTSD (PE or EMDR) were received by patients seeking care in the usual care group (Grubbs et al., 2017). Furthermore, less than 5% of usual care participants received another form of evidence-based psychotherapy, either seeking safety or acceptance and commitment therapy. Similarly, it was difficult to determine whether initial attendance in this form of treatment was due to veteran choice or to limited access to trauma-focused treatments, particularly for participants in the usual care condition. While many VHA clinicians are trained in these modalities, barriers to routinely offering such treatments have been identified (e.g., lack of sufficient staffing, schedule or workload challenges; Chard, Ricksecker, Healy, Karlin, & Resick, 2012; Finley et al., 2015) and may be particularly prominent in smaller clinical settings with limited mental health staff such as within CBOCs.

A range of reasons were endorsed for non-attendance by veterans attending CPT and other forms of PTSD-focused individual and group treatment. Only one veteran in the CPT group indicated that non-attendance was related to the interactive video technology, which matches prior studies that have found comparable treatment satisfaction for in person and clinical video teleconferencing (Morland et al., 2014; Yuen et al., 2015). Thus, our results do not appear to be driven by differences in method of treatment delivery. Within each form of psychotherapy examined, issues with scheduling appointments (i.e., having difficulty scheduling the appointment and/or appointments conflicting with other commitments) was one of the most frequently reported reasons for non-attendance (>20%). Additional logistical barriers, mainly having reliable transportation (CPT), therapy taking too much time (PTSD-focused individual psychotherapy) and not being able to afford counseling (PTSD-focused group psychotherapy), were also commonly cited (i.e., >15%) reasons for non-attendance, though specific logistical barriers reported varied across treatment groups examined. Those scheduled to attend CPT also frequently cited believing that treatment would not or did not help them as a common reason for non-attendance (26%). In contrast, only 11% of veterans scheduled to attend other PTSD-focused individual psychotherapy cited concerns about treatment efficacy as a reason for non-attendance.

The reasons for non-attendance identified in the present study were similar to those reported in the two previous studies investigating active duty samples as well as in the two qualitative studies examining veterans samples (e.g., difficulty scheduling appointments, appointments conflicting with other commitments, beliefs about treatment efficacy), however, concerns related to stigma (e.g., concerns about what others may think) appeared to be less prevalent in the current sample (Hoge et al., 2014; Hundt et al., 2018; Jennings et al., 2016; Sayer et al., 2009).

It should be noted that lists of possible reasons for non-attendance informed by clinical experience and previous research were used in both of the previous active duty studies as well as in the current veteran-focused study, with the two previous active duty studies utilizing largely the same list. While veterans in the present study were asked to generate additional reasons they may have had for non-attendance following review of the provided list, it is possible that such a method biases or limits results. For instance, participants may have been more likely to endorse reasons provided by interviewers than to provide alternative reasons not listed.

Our study has additional noteworthy limitations. Careful chart review was conducted to determine if a veteran had been scheduled for an initial CPT appointment and/or an appointment for another form of individual or group PTSD-focused psychotherapy. However, veterans were categorized into attendance groups based on self-reported attendance. Thus, if a participant misremembered details surrounding their treatment participation, it is possible that he or she could have been misclassified. Additionally, within our framework veterans who had gaps in care that did not result in missing more than 25% of scheduled appointments were classified in the Still Attending group. It is possible that such individuals are more similar to veterans in the Inconsistent Attendance group. Furthermore, a portion of veterans scheduled for other PTSD-focused individual treatment did not provide a reason for non-attendance. It is possible the response options for non-attendance in this group were less relevant.

Comparing non-attendance between a protocol-driven, time-limited psychotherapy, such as CPT, and unstructured individual or group therapy is important but imperfect. Treatment completion is more well-defined in a protocol-driven, time-limited therapy, and non-attendance is more likely to occur in a psychotherapy with weekly sessions compared with unstructured psychotherapy, which may be scheduled at more infrequent intervals (e.g., monthly).

Our sample size was also quite small, particularly once treatment groups were separated into categories of non-attendance. Replication in larger samples is needed. Generalizability of study findings is also limited by the fact that the majority of individuals participating in CPT received this care via telemedicine-based collaborative care as part of a randomized pragmatic effectiveness trial. While receipt of care via telemedicine was only reported as a reason for non-attendance by one veteran and a portion of veterans completed CPT in-person, it is possible that results may not generalize to veterans attending CPT in-person. Likewise, veterans in the TOP intervention group had care management, though the care manager encouraged engagement in all types of psychotherapy. Trauma focused therapies, such as CPT and PE, are not routinely provided in the context of care management and our findings may not generalize to those receiving CPT in the absence of care management. Similarly, veterans in the present study received care at rural CBOCs and the present study excluded some participants who may participate in these forms of treatment in usual care (e.g., veterans with substance use disorders with active use). Findings may not generalize to veterans receiving care at other facilities, such as larger VHA medical centers typically found in urban areas or to veteran subgroups who were excluded from the larger TOP trial. Finally, the majority (over 70%) of the study sample had previous PTSD treatment. It is possible that attendance rates and reasons for non-attendance may differ in veterans entering first episodes of care.

While study results should be considered preliminary, our findings appear to have important clinical and system-level implications for care delivery. Clinician and/or administrative understanding of potential barriers detected in the present study and efforts to address such issues could result in improved engagement and retention across mental health care settings offering PTSD focused treatment to our nations’ veterans. For instance, veterans frequently reported difficulties related to scheduling appointments and appointments conflicting with other responsibilities (e.g., work, school) as barriers to treatment attendance regardless of whether they were attending CPT or other PTSD focused psychotherapy. Such findings indicate a need to better understand barriers related to appointment scheduling, the possible need for a greater range of appointment times, and highlight an opportunity to address logistical barriers via the use of technology (e.g., secure video conference into the home). Cost associated with care (e.g., copays, transportation) and access to reliable transportation (e.g., access to transportation, cost of transportation) were also identified as barriers, which is noteworthy given that care was free or required only a minimal copayment in over 50% of the sample who were service-connected for PTSD.

Acknowledging potential barriers at the individual patient-level at the outset of treatment, including identifying difficulties with appointment scheduling procedures, exploring future appointment conflicts, and understanding the implications of copays and costs incurred while attending care may also help to facilitate engagement and retention. Revisiting these issues throughout the course of care may also be needed given the variability seen in reasons given for ultimately discontinuing treatment. Such an individual patient-centered approach is consistent with VHA clinical practice guidelines (Veterans Health Administration, Department of Defense, 2017) and may be particularly important given that a single enhancement to care or one size fits all solution to improve retention appears unlikely.

Study results are somewhat reassuring in the context of previously reported concerns regarding trauma-focused treatment (e.g., trauma-focused treatment is too hard; Steenkamp & Litz, 2013). While beliefs about treatment were endorsed as potential barriers, few veterans going to CPT reported their reason for non-attendance was that they “didn’t like the therapist,” “didn’t want to talk about my feelings with a therapist,” or “didn’t like treatment,” whereas none volunteered that trauma-focused therapy was too difficult or triggering. Instead study findings indicate that general concerns about treatment efficacy and beliefs about handling one’s problems may play a role in non-attendance. Previous research has found that both negative beliefs about treatment and importance of relying on one’s own ability to handle symptoms are associated with lower likelihood of mental health care utilization (Sayer et al., 2009; Stecker, Fortney, Hamilton, & Ajzen, 2007; Vogt, 2011). Thus, better understanding and addressing such beliefs at the outset of treatment and during the course of care may be important when engaging veterans in PTSD focused interventions.

It is noteworthy that a quarter of those receiving CPT reported beliefs that the intervention was not helpful. It is not uncommon for individuals participating in trauma-focused PTSD treatment to experience an initial increase in symptoms likely due to reduced avoidance of internal and external trauma-reminders (Larsen, Wiltsey Stirman, Smith, & Resick, 2016). Whether this experience resulted in veterans perceiving CPT to be less helpful than veterans participating in other forms of PTSD-focused individual and group psychotherapy represents an important empirical question. Furthermore, whether certain components of treatment were specifically found to be unhelpful, such as discussing one’s trauma, was not examined in the present study and represents another important area of future study.

To date, efforts to identify and characterize veterans who chose not to initiate or complete evidence-based psychotherapies for PTSD have largely focused on identifying patient characteristics associated with lack of treatment initiation and/or completion using data available within the electronic medical record or secondary analysis of survey data collected through intervention trials. The findings of the present study build upon this literature by examining rates of non-attendance in CPT and other forms of PTSD-focused individual and group psychotherapy and conducting the first examination of veterans self-reported reasons for non-initiation, inconsistent attendance and early treatment discontinuation in each of these forms of psychotherapy. Rate of treatment completion was higher for veterans attending CPT than for those attending other forms of PTSD-focused psychotherapy. However, rates of inconsistent attendance and early discontinuation were also higher among those attending CPT. Study findings suggest logistical barriers, particularly issues with scheduling convenient appointments, and beliefs about treatment efficacy may be important to address when engaging veterans in a full course of psychotherapy for PTSD. Replication of study findings in larger samples of veterans receiving evidence-based psychotherapies for PTSD via commonly offered delivery methods (e.g., in-person, telehealth without care management) in a wider range of clinics (e.g., VHA medical centers, urban CBOCs) and including subgroups excluded from this trial (e.g., veterans with substance use disorders with active use) is needed. Future research efforts in this area may also benefit from further exploring group differences between veterans who initiate and engage in different forms of PTSD treatment and from utilizing various methods to examine veterans self-reported reasons for non-attendance (e.g., self-report measures, open-ended questions prior to querying about known reasons for treatment non-attendance, interview lists, qualitative interviews).

Figure 2.

Attendance in other PTSD-focused individual psychotherapy six months after study enrollment