Predicting post treatment client satisfaction between behavioural activation for depression delivered either in-person or via home-based telehealth

Abstract

Introduction

Treatment engagement, adherence, cancellations and other patient-centric data are important predictors of treatment outcome. But often these data are only examined retrospectively. In this investigation, we analysed data from a clinical trial focused on innovative delivery of depression treatment to identify which patients are likely to prefer either in-home or in-person treatment based on pre-treatment characteristics.

Methods

Patient satisfaction was assessed in a trial of individuals with depression treated using identical behavioural activation therapy protocols in person or through videoconferencing to the home (N = 87 at post treatment: 42 in-person and 45 in-home participants). The Client Satisfaction Questionnaire was administered at the end of the treatment. A Tobit regression model was used to assess moderation using treatment assignment. Regression lines were generated to model treatment satisfaction as a function of treatment assignment and to identify whether and where the groups intersected. We examined the distributions of the contributing moderators to the subsets of participants above and below the intersection point to identify differences.

Results

While no significant differences in patient satisfaction were observed between the two groups, or between patients receiving treatment by different providers, baseline characteristics of the sample could be used to differentiate those with a preference for traditional, in-office care from those preferring in-home care.

Discussion

Participants who were more likely to prefer in-home care were characterized by larger proportions of veterans and lower-ranked enlisted service members. They also had more severe symptoms at baseline and less formal education. Understanding client reactions when selecting treatment modality may allow for a more satisfying patient experience.

Keywords

Home telecare telehealth telepsychiatry

Introduction

Physicians and other health care providers often assess their patients’ participation – responsiveness to treatment; adherence to therapy; tardiness or cancellations; and other patient-centric information – as a part of standard health care practice. Patients’ satisfaction with received care should be an integral part of the health care process, ensuring a dynamic, responsive relationship between a patient and a provider. Mental health services are particularly affected by patient satisfaction, since therapeutic outcome largely depends on the quality of interactions between a provider and a patient, as well as the degree of therapeutic alliance.^1–7 Ample evidence indicates that patients’ participation and satisfaction with the interaction, the treatment, and the provider are vital for a successful therapeutic relationship.^8–10 It is, therefore, not surprising that attempts to measure patient satisfaction, starting in the early 1950s, involved mental health services.^11–13

Patient satisfaction represents a complex outcome measure that combines a mixture of potential predisposing elements, including a patient’s characteristics, expectations, involvement and attitudes; the disease or illness itself and comorbid conditions; and provider-, environment- and therapeutic approach-associated factors.^3,14–23 Among the factors that affect patient satisfaction is the presence of clinical depression.¹⁴ Depression is associated with lower satisfaction with life, distorted views of others and self, and negative or destructive communication patterns with strangers, acquaintances and, particularly, with those closest to patients with depression.^24–38 Lower levels of life satisfaction were observed in patients with major depressive disorder even during periods of diagnostic remission,³⁴ and there is a strong association between depression severity and life satisfaction.²⁴ Patients with depression may also experience difficulty interacting with others, problems with social skills, reduced motivation, and lower ability to cooperate to achieve a goal.^35,39,40 These depression-related symptoms could impair patients’ ability to successfully engage in the treatment process, undermine their social support network and hinder their ability to constructively interact with health care providers. This can reduce the likelihood of a successful therapeutic outcome and, consequently, patient satisfaction in psychotherapy.

Patient satisfaction is further complicated in therapeutic modalities using telehealth options because attitudes towards technology, quality of the connection (e.g. audio and video quality) and other technology-based factors may contribute to or alter relevant elements responsible for patient’s satisfaction.^41–46 While convenient access is often mentioned as a reason for high satisfaction among patients utilizing telehealth options for mental health,^43,47–48 some patients may be best served by in-office treatment because of their individual clinical needs (e.g. loneliness, symptom severity, avoidance behaviour) and preferences (e.g. driving distance, work schedule, childcare needs).^45,48–54 In addition to barriers to successful telemental health care associated with patients’ characteristics and attitudes, providers may also be resistant to, poorly prepared for or less capable of providing mental health care using telehealth options.^55–56 Therefore, it is not surprising that some patients may not respond in the same way or receive the same benefit from services provided via telehealth or videoconferencing mental health care.^54,57,58

Because treatment satisfaction is an important part of achieving a successful treatment outcome, we attempted to identify differential treatment satisfaction between two delivery modalities of behavioural activation therapy for depression (BATD):^59,60 in-office or via videoconferencing to the patient’s home. The identification of differential treatment satisfaction and the characteristics of individuals who prefer one modality over the other will aid in the identification of the most suitable modality for delivery of BATD.

Methods

Participants

Active duty participants were recruited at a large regional military treatment facility located at Joint Base Lewis-McChord (JBLM) in Washington State, and veterans at the VA Health Care System (VAPHCS) in Portland, Oregon, USA. Participants were included in the study if they met Diagnostic and Statistical Manual of Mental Disorders (DSM-IV)⁶¹ diagnostic criteria for major depressive disorder or minor depressive disorder based on the Structured Clinical Interview for the DSM-IV Axis I Disorders, Research Version, Patient Edition (SCID-I/P⁶²). A complete description of the trial methodology is available in a previous publication,⁶³ as are the primary clinical outcomes associated with this trial.⁶⁴

Procedures

A study research coordinator conducted initial eligibility criteria screening via telephone. Eligible participants were scheduled to complete informed consent procedures and an in-person SCID-I/P with the study team’s independent clinical assessor. A total of 92 active duty and 29 veteran participants met all inclusion criteria and were randomized into two treatment groups, one receiving treatment in person at providers’ offices, and the other receiving treatment via videoconference in their own homes. Participants were randomized to treatment condition in a 1:1 ratio using block randomization of size 10, stratified by baseline major or minor depressive disorder diagnosis using Stata 12.1.⁶⁵ Participants randomized to the in-home condition were provided with a Dell M6500 laptop computer and Tandberg Precision High Definition webcam. The laptops were pre-loaded with Jabber video software.

Treatment protocol

An eight-session BATD protocol was used as the study’s intervention.⁶⁶ Both groups received the same BATD treatment for 50–60 min sessions every week for eight weeks with clinical assessments conducted at baseline, four-week midpoint, eight-week treatment completion and three-month follow-up. There were five doctoral level mental health treatment providers (four at JBLM and one at VAPHCS), who received training from a BATD expert and consultant. The fidelity reviews confirmed an overall adherence of 98.19% for all clinicians providing treatment.

Measures

Client Satisfaction Questionnaire (CSQ)⁶⁷

The CSQ-8 is an eight-item self-report measure of general satisfaction with psychotherapeutic treatment. Participants are asked to rate satisfaction on a four-point scale, with a possible range of 8–32, with higher scores indicating greater satisfaction. Internal consistency and construct validity have been established and the measure is widely used in research. The CSQ was administered at post treatment. Coefficient alpha for the measure was 0.91.

Other measures

Participants provided demographic information including occupation/work status/income/living situation, branch of service/highest rank, and medication use.

The Beck Hopelessness Scale (coefficient alpha = 0.91)⁶⁸ and the Beck Depression Inventory II (BDI-II; coefficient alpha = 0.95)⁶⁹ were the primary outcome measures for the original trial. Anxiety was assessed by the Beck Anxiety Inventory (coefficient alpha = 0.91),⁷⁰ posttraumatic stress syndrome (PTSD) by PTSD Checklist – Military version (coefficient alpha = 0.95),⁷¹ loneliness using the De Jong Gierveld loneliness scale (coefficient alpha = 0.76)⁷² and stigma using the Inventory of Attitudes toward Seeking Mental Health Services (coefficient alpha = 0.86).⁷³

Statistical analyses

We used the technique outlined by Kraemer⁷⁴ to identify moderation of treatment satisfaction. Unlike traditional testing of moderation using product terms for each moderator in a regression model, Kraemer’s technique derives a composite moderator of two or more underlying potential moderators. This approach improves the likelihood of identifying moderation with modest sample sizes that are underpowered to detect individual moderators in the traditional approach.

In the first phase of the analysis, the data were restructured to generate a dataset with all possible pairs of participants in one treatment group with those in the other. Differences scores were calculated for the satisfaction measure. For each of the candidate moderator variables, we calculated both a difference score and an average score. The difference score accounted for the main effect of the candidate moderator on the outcome; the average score was the effect of any moderation. The difference and average scores for each candidate moderator were used in separate regression models with the difference score for satisfaction as the outcome. From these models, we calculated the effect size of moderation for each candidate moderator. This effect size was the correlation between the satisfaction difference score and the candidate moderator average score. We used the 2.5 and 97.5 percentile effect size values of bootstrap resamples of the data to create the 95% confidence interval (CI) for each effect size estimate.

Candidate moderators were selected for inclusion in the composite moderator based on substantive considerations, individual effect size, and independent contribution to the composite. For the latter, we used a principal component analysis to identify orthogonal sets among the candidate moderator variables. This allowed us to create a better composite by minimizing shared variance and potential collinearity between the individual variables included.⁷⁵ After identifying the set of candidate moderators to be included in the composite moderator, we estimated a regression model in the all-possible-pairs dataset with the difference and average terms for each of the candidate variables included simultaneously. The coefficients for the average terms became the weights for generating the composite moderator.

We used a Tobit regression model to assess moderation using a product term of treatment assignment and the combined moderator in the original post-treatment dataset. We selected a Tobit model given the pronounced ceiling effect observed in the data. We generated regression lines of treatment satisfaction as a function of treatment assignment and composite moderator score to identify where the lines intersected. For descriptive purposes, we examined the distributions of the contributing moderators to the subsets of participants above and below the intersection point to identify possible differences for consideration in future research.

All models were estimated using Stata 13.⁶⁵ Categorical candidate moderator variables were effect coded (–.5, .5). All variables were standardized using the total sample distribution at post treatment. This facilitated direct comparison of the resulting effect size estimates. Bootstrap 95% CIs were generated using 1000 resamples of the data.

Results

Of the 121 participants randomized to the two treatment groups, a total of 87 (71.9%) completed the post-treatment outcome assessment, which included the measure of satisfaction with the assigned treatment. While there was no difference between treatment groups in the rate of study withdrawal,⁶⁴ which lessens concerns about internal validity, external validity to the original study population may be limited. Characteristics of the participants included in the current study are presented in Table 1. These characteristics are similar to those for the total sample randomized.

Table 1.

Characteristics of participants who finished the study and provided information on client satisfaction, by assigned treatment group.

	In-office (n = 42)	In-home (n = 45)
	M (SD)	M (SD)	p-value
Age	36.5 (12.06)	34.33 (11.82)	.390
Number of deployments	2.00 (2.20)	1.98 (1.86)	.960
Baseline BDI-II	29.69 (11.74)	26.96 (11.25)	.270
Baseline BHS	11.00 (6.27)	8.73 (5.22)	.070
Baseline BAI	17.07 (12.69)	14.71 (10.12)	.339
Baseline PCL	44.29 (15.49)	42.89 (13.99)	.660
Baseline stigma	31.12 (14.28)	34.56 (14.72)	.273
Baseline loneliness	13.55 (3.10)	12.53 (2.87)	.117
	%	%
Female	23.81	15.56	.332
Rank/grade E1–E4	47.62	44.44	.767
White, not Hispanic	73.81	75.56	.851
Veteran	28.57	24.44	.663
Two years of college or more	35.71	31.11	.649
Married	54.76	44.44	.336

BAI: Beck Anxiety Inventory; BDI-II: Beck Depression Inventory II; BHS: Beck Hopelessness Scale; PCL: Posttraumatic stress syndrome Checklist.

There were 14 variables identified as potential moderators (Table 2). Highest attained rank/grade, being a veteran as opposed to an active-duty service member, and baseline symptom severity on the BDI-II had the largest effect size estimates. We identified a total of four components in the principal component analysis that had eigenvalues greater than 1. A highest attained rank/grade of E1–E4, being a veteran as opposed to an active-duty service member, and baseline symptom severity as measured by the BDI-II were the potential moderators with the largest effect sizes. These three variables all loaded onto different components. Not surprisingly, the other measures of symptom severity at baseline all loaded onto the same component as baseline BDI-II score. The fourth and final contributing moderator was education (attaining a two-year college degree or more).

Table 2.

Effect size and bootstrapped 95% confidence intervals of all potential moderators and the regression weights of moderators included in the composite moderator.

Potential moderator	Effect size	95% CI	Weight
Rank/grade E1–E4	0.08	0.05, 0.11	0.51
Veteran	0.07	0.04, 0.10	0.98
Baseline BDI-II	0.07	0.04, 0.10	0.85
Baseline loneliness	0.06	0.03, 0.09
Baseline PCL	0.06	0.03, 0.10
Two-year college degree or more	−0.06	−0.09, –0.03	−0.92
White, not Hispanic	−0.06	−0.09, –0.02
Baseline stigma	−0.04	−0.07, –0.01
Number of deployments	−0.03	−0.06, 0.00
Age	0.03	0.00, 0.06
Female	0.02	−0.01, 0.05
Baseline BAI	0.02	−0.01, 0.05
Baseline BHS	0.01	−0.02, 0.04
Married	−0.01	−0.04, 0.02

BAI: Beck Anxiety Inventory; BDI-II: Beck Depression Inventory II; BHS: Beck Hopelessness Scale; CI: confidence interval; PCL: Posttraumatic stress syndrome Checklist.

The final Tobit regression model identified a small interaction that was not statistically significant in that model at the conventional .05 level (b = 0.30, SE = –0.18, p = .111); however, the interaction effect size based on the all-possible-pairs dataset was 0.08 (95% CI = 0.06, 0.10). This effect size was comparable to that of the strongest individual contributing moderator. The composite moderator effect size did have a marginal improvement in precision over the individual moderator effect sizes. Figure 1 provides a graphical interpretation of the possible moderation effect. For those in the in-office treatment group, treatment satisfaction scores decreased with increasing scores on the moderator (b = −0.21, 95% CI = −0.48, 0.06); for those in the in-home group, there was a modest increase in treatment satisfaction scores (b = 0.09, 95% CI = −0.16, 0.33) as the moderator score increased. The two curves intersected at the mediator score value of 0.95. Participants with moderator scores below the intersection point had higher satisfaction scores for the in-office treatment modality. Those with moderator scores above the intersection point had higher satisfaction scores for the in-home modality.

Figure 1.

Plot of satisfaction as a function of the composite moderator score, by treatment group.

Table 3 presents the descriptive values of the four contributing moderator variables to the composite moderator for participants below and at or above the intersection point in Figure 1. Participants below the intersection point had higher satisfaction for the in-office version of the treatment. They had lower baseline scores on the BDI-II and more formal education than those above the intersection point. Participants with composite moderator scores at, or above, the intersection point tended to favour the in-home version of therapy over the in-office option. They had a higher percentage of veterans and individuals from the lower enlisted ranks.

Table 3.

Values of individual moderator variables included in the composite moderator (M) for individuals below and above the intersection point of the treatment group regression lines.

Moderator variable	Below intersection point	Above intersection point
Moderator variable	M < 0.95 (n = 61)	M ≥ 0.95 (n = 26)
Rank/grade E1–E4, %	32.79	76.92
Veteran, %	16.39	50.00
Baseline BDI, M (SD)	25.33 (10.97)	35.19 (9.78)
Two-year college degree or more, %	44.26	7.69

BDI: Beck Depression Inventory

Discussion

Patient satisfaction plays an important role in treatment outcome. Thus, it is not surprising that both clinical practitioners and researchers often include patient satisfaction with the treatment, the provider and the therapy results as one of the outcome measures.^76,77 Importantly, O’Donoghue and colleagues⁷⁸ have shown that patient satisfaction data are comparable between clinical settings and research studies. The present study evaluated patient satisfaction in a group of current and former military personnel with depression who were participating in a randomized clinical trial using BATD delivered in-office and by videoconferencing to the home.^63,64

In the present study, patient satisfaction was very high, with no significant differences between the treatment modalities. These findings are consistent with previous studies comparing in-person and telehealth treatments of mental health disorders.^79–81 There were no significant differences in patient satisfaction between treatment sites or between different treatment providers, indicative of a comparable quality of care and a strong therapeutic allegiance developed between the participants and the providers, irrespective of the treatment modality. The two treatment modalities have different strengths and weaknesses. In-person BATD provides a traditional, in-office treatment with a richer communication environment (both verbal and non-verbal) than in-home treatment through telehealth. On the other hand, the in-home treatment modality may offset some of the potential missing aspects of the in-person treatment by enabling greater comfort due to the participants’ ability to relax in a familiar environment. This balance between the positive and negative aspects of the two modalities may explain the similarity in the levels of patient satisfaction observed in our study.

What is unique about the current investigation, however, is that rather than assessing change in satisfaction over the course of a standardized, eight session treatment protocol, this study attempted to identify which patients are likely to prefer either in-home or in-person treatment based on baseline characteristics. Contrary to popular beliefs about tech-savvy youth, higher end-of-treatment satisfaction for in-person care was most commonly associated with younger age and more junior military status. Conversely, higher end-of-treatment satisfaction for in-home care was more commonly associated with older, more senior, and more symptomatic service members. Given this distinction, the key question becomes: why? Plausible explanations for this profile difference include stigma, symptom severity and convenience.

Stigma. We typically assume that stigma prevents treatment-seeking behaviour, but what if it could also account for what treatment modalities patients express a preference for? Those who experience stigma about mental health services, but who are able to seek treatment regardless, may show an increased preference for in-home services, such as those in this investigation who represented an older, more senior demographic that may have experienced high degrees of stigma in the past prior to the military’s efforts to actively combat mental health stigma.

Symptom severity. Perhaps those with more severe depression scores prefer in-home treatment because it does not force them to engage in the naturalistic behavioural activation required for in-office care. In this scenario, in-home treatment options are a double edged sword. One the one hand, they result in a stepped care approach where those who are unwilling to engage in in-office care may be willing to engage in home-based care. On the other hand, however, home-based care may facilitate avoidance and disconnection from the naturalistic reinforcement of functional behaviour that occurs when an individual takes the activating steps required to attend an office appointment (e.g. engaging in personal hygiene routine, grooming one’s self, getting dressed, driving, interacting with others, etc.).

Convenience. Convenience has long been cited as a potential benefit for telehealth. For example, retired military veterans who reside several hours away from the nearest VA Hospital may be especially interested in remote care. But what about active duty service members who typically live on or very near the military installation to which they are assigned? Convenience becomes more than travel time or cost. It has to do with time off/away from work, arrangement of child care services, managing a rigid schedule, et cetera. Given this, it is not surprising that the veterans included in this study were more satisfied with in-home services as they may have appealed to their needs and challenges. Conversely, those still in service may be navigating a different set of demands that were better suited to traditional care models.

While we cannot say which, or if any, of these issues accounts for the differences observed, this study stresses the importance of knowing what preferences are likely to be influencing a patient’s treatment-seeking behaviour and experience. In this case, basic demographic characteristics may influence the satisfaction that a patient may experience when receiving either home based or in-office care.

There are some key limitations to consider in this study. First, the power to detect a statistical interaction was low. This was reflected in the lack of traditional statistical significance of the interaction product term in the regression model. However, the use of bootstrapping for the overall effect size did provide a CI that did not cover the null value. At best, this suggests weak evidence for an interaction that would need to be studied more in depth in a larger study. Second, the generation of a linear combination moderator variable improved our ability to identify a possible interaction; however, this came at the cost of identifying which particular variable(s) ultimately produce the observed differential response in treatment satisfaction. Another consideration associated with the linear combination is that the composition of the linear combination in terms of both variables and the associated weights was determined from the dataset used in the final analysis. This means that what was observed in these data may not replicate in an independent dataset. Finally, the description of groups based on the intersection point of the interaction is not inferential. This description can inform hypothesis generation for additional study, but it is not in and of itself strong evidence that individuals with these characteristics will exhibit a strong preference for one modality over another.

The construct of satisfaction is complex, and when applied to clinical intervention, that complexity is amplified. However, as technology allows the application of more diverse and unfamiliar treatment modalities, clinicians need to be aware of how those modalities may be received by their patients. This study provides insight into how basic demographic characteristics can help target particular intervention modalities based on the satisfaction a service member or veteran may experience when seeking treatment for depression. Junior patients or those with less severe symptoms may be more satisfied with traditional care models, while more senior or more symptomatic patients may prefer home-based care.

Footnotes

Acknowledgement

The content of this information does not necessarily reflect the position or the policy of the United States Government, and no official endorsement should be inferred.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

References

Brady

Warnock-Parkes

Barker

, et al. Early in-session predictors of response to trauma-focused cognitive therapy for posttraumatic stress disorder. Behav Res Ther 2015; 75: 40–47.

Germain

Marchand

Bouchard

, et al. Assessment of the therapeutic alliance in face-to-face or videoconference treatment for posttraumatic stress disorder. Cyberpsychol Behav Soc Netw 2010; 13: 29–35.

Haertl

Behrens

Houtujec

, et al. Factors influencing satisfaction and efficacy of services at a free-standing psychiatric occupational therapy clinic. Am J Occup Ther 2009; 63: 691–700.

Hall

Fereira

Maher

, et al. The influence of the therapist-patient relationship on treatment outcome in physical rehabilitation: A systematic review. Phys Ther 2010; 90: 1099–1110.

McMillan

Lee

. A systematic review of behavioral experiments vs. exposure alone in the treatment of anxiety disorders: A case of exposure while wearing the emperor’s new clothes? Clin Psychol Rev 2010; 30: 467–478.

Sweeney

Fahmy

Nolan

, et al. A mixed-method study exploring therapeutic relationships and their association with service user satisfaction in acute psychiatric wards and crisis residential alternatives. Health Serv Delivery Res 2014; 2: 22–22.

Wagner

Brand

Schulz

, et al. Online working alliance predicts treatment outcome for posttraumatic stress symptoms in Arab war-traumatized patients. Depress Anxiety 2012; 29: 646–651.

Attkisson

Zwick

. The client satisfaction questionnaire. Psychometric properties and correlations with service utilization and psychotherapy outcome. Eval Program Plann 1982; 5: 233–237.

Morris

Gannon

. Drug misuse treatment services in Scotland: Predicting outcomes. Int J Qual Health Care 2008; 20: 271–276.

10.

Willer

Miller

. On the relationship of client satisfaction to client characteristics and outcome of treatment. J Clin Psychol 1987; 34: 157–160.

11.

Ahmed

Koltuv

. Evaluation of an inpatient service by consumer feedback. Aust N Z J Psychiatry 1976; 10: 263–268.

12.

Gelso

. Effect of audiorecording and videorecording on client satisfaction and self-expression. J Consult Clin Psychol 1973; 40: 455–461.

13.

Heilbrun

Jr . Effects of briefing upon client satisfaction with the initial counseling contact. J Consult Clin Psychol 1972; 38: 50–56.

14.

Calsyn

Borse

Klinkenberg

, et al. Moderators and mediators of client satisfaction in case management programs for clients with severe mental illness. Ment Health Serv Res 2002; 4: 267–275.

15.

Hopkins

Loeb

Fick

. Beyond satisfaction, what service users expect of inpatients mental health care: A literature review. J Psychiatr Ment Health Nurs 2009; 16: 927–937.

16.

Jones

Hanusa

Appelt

, et al. Racial differences in veterans’ satisfaction with addiction treatment services. J Addict Med 2015; 9: 383–390.

17.

LeVois

Nguyen

Attkisson

. Artifact in client satisfaction assessment: Experience in community mental health settings. Eval Program Plann 1981; 4: 139–150.

18.

Locker

Dunt

. Theoretical and methodological issues in sociological studies of consumer satisfaction with medical care. Soc Sci Med 1978; 12: 283–292.

19.

Meyer

Zane

. The influence of race and ethnicity in clients’ experiences of mental health treatment. J Community Psychol 2013; 41: 884–901.

20.

Perreault

Katerelos

Tardif

, et al. Patients’ perspectives on information received in outpatient psychiatry. J Psychiatr Ment Health Nurs 2006; 13: 110–116.

21.

Tanner

. Factors influencing client satisfaction with mental health services: A review of quantitative research. Eval Program Plann 1981; 4: 279–286.

22.

Verhaeghe

Bracke

. Stigma and trust among mental health service users. Arch Psychiatr Nurs 2011; 25: 294–302.

23.

Viering

Jäger

Nordt

, et al.

Does “individual placement and support” satisfy the users’ needs?

Front Public Health 2015; 3: 160–160.

24.

Endicott

Nee

Harrison

, et al. Quality of Life Enjoyment and Satisfaction Questionnaire: A new measure. Psychopharmacol Bull 1993; 29: 321–326.

25.

Gotlib

. Perception and recall of interpersonal feedback: Negative bias in depression. Cogn Ther Res 1983; 7: 399–412.

26.

Gotlib

Robinson

. Responses to depressed individuals: Discrepancies between self-report and observer-rated behavior. J Abnorm Psychol 1982; 91: 231–240.

27.

Grabow

Burkhart

. Social skill and depression: A test of cognitive and behavioral hypotheses. J Clin Psychol 1986; 42: 21–27.

28.

Hautzinger

Linden

Hoffman

. Distressed couples with and without a depressed partner: An analysis of their verbal interaction. J Behav Ther Exp Psychiatry 1982; 13: 307–314.

29.

Jacobson

Anderson

. Interpersonal skill and depression in college students: An analysis of the timing of self-disclosures. Behav Ther 1982; 13: 271–282.

30.

Juengst

Adams

Bogner

, et al. Trajectories of life satisfaction after traumatic brain injury: Influence of life roles, age, cognitive disability, and depressive symptoms. Rehabil Psychol 2015; 60: 353–364.

31.

Kahn

Coyne

Margolin

. Depression and marital disagreement: The social construction of despair. J Soc Pers Relat 1985; 2: 447–461.

32.

Lemogne

Piolino

Friszer

, et al. Episodic autobiographical memory in depression: Specificity, autonoetic consciousness, and self-perspective. Conscious Cogn 2006; 15: 258–268.

33.

Raab

MacIntosh

Gros

, et al. Impact of comorbid depression on quality of life in male combat Veterans with posttraumatic stress disorders. J Rehabil Res Dev 2015; 52: 563–576.

34.

Schechter

Endicott

Nee

. Quality of life of “normal” controls: Association with lifetime history of mental illness. Psychiatry Res 2007; 152: 45–54.

35.

Segrin

. Social skills deficits associated with depression. Clin Psychol Rev 2000; 20: 379–403.

36.

Segrin

Fitzpatrick

. Depression and verbal aggressiveness in different marital couple types. Commun Stud 1992; 43: 79–91.

37.

Sperduti

Martinelli

Kalenzaga

, et al. Don’t be too strict with yourself! Rigid negative self-representation in healthy subjects mimics the neurocognitive profile of depression for autobiographical memory. Front Behav Neurosci 2013; 7: 41–41.

38.

Yang

Burr

. Combat exposure, social relationships, and subjective well-being among middle-aged and older Veterans. Aging Ment Health 2015; 30: 1–10.

39.

Brendan Clark

Thorne

Hardy

, et al. Cooperation and depressive symptoms. J Affect Disord 2013; 150: 1184–1187.

40.

Lam

Schuck

Smith

, et al. Response style, interpersonal difficulties and social functioning in major depressive disorder. J Affect Disord 2003; 75: 279–283.

41.

Chou

Comer

Turvey

, et al. Technological considerations for the delivery of real–time child telemental healthcare. J Child Adolesc Psychopharmacol 2016; 26: 192–197.

42.

Jenkins–Guarnieri

Pruitt

Luxton

, et al. Patient perceptions of telemental health: Systematic review of direct comparisons to in-person psychotherapeutic treatments. Telemed J E Health 2015; 21: 652–660.

43.

Knaevelsrud

Maercker

. Does the quality of the working alliance predict treatment outcome in online psychotherapy for traumatized patients? J Med Internet Res 2006; 8: e31–e31.

44.

Urness

Wass

Gordon

, et al. Client acceptability and quality of life – telepsychiatry compared to in-person consultation. J Telemed Telecare 2006; 12: 251–254.

45.

Young

. An empirical examination of client attitudes towards online counseling. Cyberpsychol Behav 2005; 8: 172–177.

46.

Luxton

Pruitt

O’Brien

, et al. An evaluation of the feasibility and safety of a home-based telemental health treatment for post-traumatic stress in the US military. Telemed J E Health 2015; 21: 880–886.

47.

King

Stoller

Kidorf

, et al. Assessing the effectiveness of an Internet based videoconferencing platform for delivering intensified substance abuse counseling. J Subst Abuse Treat 2009; 36: 331–338.

48.

Brenes

Ingram

Danhauer

. Telephone-delivered psychotherapy for late-life anxiety. Psychol Serv 2012; 9: 219–220.

49.

Elford

White

St John

, et al. A prospective satisfaction study and cost analysis of a pilot child telepsychiatry service in Newfoundland. J Telemed Telecare 2001; 7: 73–81.

50.

Grant

Rockwood

Stennes

. Client satisfaction with telehealth in assisted living and homecare. Telemed J E Health 2015; 21: 987–991.

51.

Murphey

Mitchell

Hallett

. A comparison of client characteristics in cyber and in-person counseling. Stud Health Technol Inform 2011; 167: 149–153.

52.

Simpson

Doze

Urness

, et al. Telepsychiatry as a routine service – the perspective of the patient. J Telemed Telecare 2001; 7: 155–160.

53.

Luxton

Sirotin

Mishkind

. Safety of telemental healthcare delivered to clinically unsupervised settings: A systematic review. Telemed J E Health 2010; 16: 705–711.

54.

Smolenski

Pruitt

Vuletic

, et al. Unobserved heterogeneity in response to treatment for depression through videoconference. Psychiatr Rehabil 2017; 40: 303–308.

55.

Brooks

Turvey

Augusterfer

. Provider barriers to telemental health: obstacles overcome, obstacles remaining. Telemed J E Health 2013; 19: 433–437.

56.

Jameson

Farmer

Head

, et al. VA community mental health service providers’ utilization of and attitudes toward telemental health care: The gatekeeper’s perspective. J Rural Health 2011; 27: 425–432.

57.

Hailey

Roine

Ohinmaa

. Systematic review of evidence for the benefits of telemedicine. J Telemed Telecare 2002; 8: 1–30.

58.

Hailey

Roine

Ohinmaa

. The effectiveness of telemental health applications: A review. Can J Psychiatry 2008; 53: 769–778.

59.

Lejuez

Hopko

. A brief behavioral activation treatment for depression: Treatment manual. Behav Modif 2001; 25: 255–286.

60.

Lejuez

Hopko

LePage

, et al. A brief behavioral activation treatment for depression. Cogn Behav Practice 2001; 8: 164–175.

61.

American Psychiatric Association. Diagnostic and statistical manual of mental disorders, 4th ed. Washington, DC: American Psychiatric Association, 2000.

62.

First

Spitzer

Gibbon

, et al. Structured clinical interview for DSM-IV-TR axis I disorders, research version, patient edition. (SCID-I/P), New York: New York State Psychiatric Institute, 2002.

63.

Luxton

Pruitt

O’Brien

, et al. Design and methodology of a randomized clinical trial of home-based telemental health treatment for U.S. military personnel and veterans with depression. Contemp Clin Trials 2014; 38: 134–144.

64.

Luxton

Pruitt

Wagner

, et al. Home-based telebehavioral health for U.S. military personnel and veterans with depression: A randomized controlled trial. J Cons Clin Psychol 2016; 84: 923–934.

65.

StataCorp. Stata statistical software: Release 13. College Station, TX: StataCorp LP, 2013.

66.

Cuijpers

van Straten

Andersson

, et al. Psychotherapy for depression in adults: A meta-analysis of comparative outcome studies. J Consult Clin Psychol 2008; 76: 909–922.

67.

Nguyen

Attkisson

Stegner

. Assessment of patient satisfaction: development and refinement of a service evaluation questionnaire. Eval Program Plann 1983; 6: 299–313.

68.

Beck

Weissman

Lester

, et al. The measurement of pessimism: The hopelessness scale. J Consult Clin Psychol 1974; 42: 861–865.

69.

Beck

Steer

Brown

. Beck Depression Inventory, 2nd ed. San Antonio, TX: Psychological Corporation, 1996.

70.

Beck

Steer

. Beck Anxiety Inventory manual, San Antonio, TX: Psychological Corporation, 1990.

71.

Weathers FW, Huska JA and Keane TM. PTSD checklist – military version (PCL-M) for DSM-IV. Boston, MA: National Center for PTSD-Behavioral Science Division, 1991.

72.

De Jong Gierveld

van Tilburg

. The De Jong Gierveld short scales for emotional and social loneliness: Tested on data from 7 countries in the UN generations and gender surveys. Eur J Ageing 2010; 7: 121–130.

73.

Mackenzie

Knox

Gekoski

, et al. An adaptation and extension of the attitudes toward seeking professional psychological help scale 1. J Appl Soc Psychol 2004; 34: 2410–2433.

74.

Kraemer

. Discovering, comparing, and combining moderators of treatment on outcome after randomized clinical trials: A parametric approach. Stat Med 2013; 32: 1964–1973.

75.

Wallace

Frank

Kraemer

. A novel approach for developing and interpreting treatment moderator profiles in randomized clinical trials. JAMA Psychiatry 2013; 70: 1241–1247.

76.

Justice

McBee

. A client satisfaction survey as one element in evaluation. Community Ment Health J 1978; 14: 248–252.

77.

Sweeney

Fahmy

Nolan

, et al. The relationship between therapeutic alliance and service user satisfaction in mental health inpatient wards and crisis house alternatives: A cross-sectional study. PLoS One 2014; 9: e100153–e100153.

78.

O’Donoghue

Roche

Ranieri

, et al. Service users’ perceptions about their hospital admissions elicited by service user-researchers or by clinicians. Psychiatr Serv 2013; 64: 416–422; 416, e1-3.

79.

Bishop

O’Reilly

Maddox

, et al. Client satisfaction in a feasibility study comparing face-to-face interviews with telepsychiatry. J Telemed Telecare 2002; 8: 217–221.

80.

O’Reilly

Bishop

Maddox

, et al. Is telepsychiatry equivalent to face-to-face psychiatry? Results from a randomized controlled equivalence trial. Psychiatr Serv 2007; 58: 836–843.

81.

Stubbings

Rees

Roberts

, et al. Comparing in-person to videoconferencing cognitive behavioral therapy for mood and anxiety disorders: Randomized clinical trial. J Med Internet Res 2013; 15: e258–e258.

Predicting post treatment client satisfaction between behavioural activation for depression delivered either in-person or via home-based telehealth

Abstract

Introduction

Methods

Results

Discussion

Keywords

Introduction

Methods

Participants

Procedures

Treatment protocol

Measures

Client Satisfaction Questionnaire (CSQ) 67

Other measures

Statistical analyses

Results

Discussion

Footnotes

Acknowledgement

Declaration of Conflicting Interests

Funding

References

Client Satisfaction Questionnaire (CSQ)⁶⁷