Using Problem-Solving Therapy to Reduce Depressive Symptom Severity Among Older Adult Methadone Clients

Abstract

Objective:

Treatment of older adults in methadone maintenance treatment (MMT) with comorbid mood disorders presents many challenges. The goal of this randomized controlled trial was to assess the combined effect of problem-solving therapy (PST) with treatment as usual (TAU) compared to TAU alone on decreasing depressive symptoms in opioid-addicted older adults in MMT.

Methods:

Sixty-four methadone clinic attendees, at least 50 years old, with moderate depressive symptom were randomized to PST + TAU (n = 29) or TAU (n = 35). PST is an intervention for depressed older adults with a focus on teaching problem-solving skills.

Results:

There was not a statistically significant difference in depression scores between the two treatment groups. However, there is evidence for a clinically meaningful decrease in symptoms in the PST + TAU group 7 months post treatment (Cohen’s d = .57)

Conclusion:

Clinical implications of PST + TAU in comparison to TAU on depressive symptom seemed to have lasting benefits at 6-month post treatment.

Keywords

older adults methadone maintenance depression problem-solving therapy

Older adults in methadone treatment with comorbid mood disorders present many challenges when entering treatment. Federal law requires that methadone be distributed in a clinic setting and be coupled with counseling. Yet due to inadequate mental health training for staff and financial restrictions, treatment for mental health issues is lacking at methadone clinics (Hu, Kline, Huang, & Ziedonis, 2006; Searby, Maude, & McGrath, 2015). Effective treatments for depression exist for older adults, yet these treatment approaches are complicated by numerous service barriers when applied with older adult patients with opioid addiction. Concomitantly, retaining clients in methadone maintenance treatment (MMT) to address their addiction also poses challenges (Delgadillo et al., 2015).

Background

Adults receiving treatment for opioid use are aging; individuals aged 50–59 years comprised the largest age cohort in opioid treatment in New York City (Han et al., 2015). The aging opioid-addicted cohort has begun to alter the demographic characteristics of individuals in need of services for opioid addiction (Kuerbis, Sacco, Blazer, & Moore, 2014; Levy & Anderson, 2005). For example, depressive disorders are common among older adults who use opioids. Studies of older adults participating in MMT have documented over a third of the participants meet the criteria for major depression (Lofwall, Brooner, Bigelow, Kindbom, & Strain, 2005; Rosen, Smith, & Reynolds, 2008). Adverse consequences related to depression among older adults include impaired social and occupational functioning that interferes with compliance to treatment modalities (American Psychiatric Association, 2013; Stein & Anderson, 2003) and adverse consequences on well-being and ability to refrain from illegal drug use (Stein & Anderson, 2003; Wu & Blazer, 2014). As opioid-addicted individuals age, these mental health problems may be exacerbated by age-related physical health problems or social isolation and may require specialized medical approaches and psychosocial support (Choi, DiNitto, & Marti, 2014).

Clinicians have begun to view comorbid depression and substance use as interactive and mutually sustaining (Hendrie et al., 2013; Volkow, 2004). Yet the treatment of depression in older adults who are addicted to opioids has not been addressed in a systematic fashion (Choi et al., 2014). Understanding the interconnectedness of psychiatric and substance use disorders is critical to providing appropriate services to an aging population (Sahker, Schultz, & Arndt, 2015). Thus, a relatively brief standardized approach that focuses on current problems and solutions is needed for use with the aging and substance using population within substance use treatment facilities.

Study Objectives

The purpose of this study was to compare the efficacy of the combination of problem-solving therapy and treatment as usual (PST + TAU) to TAU in decreasing rates of depressive symptoms for adults over the age of 50 in MMT. It was hypothesized that depressive symptoms would be less for recipients of PST + TAU in comparison to recipients receiving only TAU.

Method

A university-based institutional review board reviewed and approved the protocol throughout the duration of the study. Participants provided written informed consent prior to their enrollment in the study.

Research Design

This study used a randomized control treatment design to assess the effect of PST + TAU to TAU on depressive symptoms.

Participants

Participants were eligible to be enrolled in the study if they were (a) at least 50 years old, (b) attended the methadone clinic from 3 to 7 days a week, (c) had a depressive symptom severity score on the Patient Health Questionairre - 9 (PHQ-9) of 10 or greater in the past 2 weeks or a score between 6 and 9 for two consecutive months, and (d) had a score of 23 or higher on the mini-mental state examination. Respondents who had a PHQ-9 score between 6 and 9 were screened a second time, 1 month after the baseline assessment. These respondents were eligible for the study if the second score was 6 or higher thus establishing persistent low-level depressive symptoms.

Study Setting

Older adult methadone clients were screened for depression at an urban, freestanding MMT clinic. During the course of the study, the clinic served approximately 640 clients per day with 65 staff members, including drug and alcohol counselors, outreach workers, medical personnel, and other support staff. Similar to national trends, the clinic was serving increasing numbers of older adults, and at the time of the study about a third of the clinic’s patients were 50 years of age or older.

Interventions

PST

PST is an empirically tested, relatively brief, intervention for older adults with late life mood disorders; it is designed for health professionals with a focus on teaching problem-solving skills (Areán, Hegel, & Reynolds, 2001) and often used by social workers (Gellis & Kenaley, 2008). The advantages of PST relative to other cognitive behavioral therapies is that it is less cognitively demanding for an older adult population with mood and substance use disorders (Mackin & Arean, 2005; Rosen, Hunsaker, Albert, Cornelius, & Reynolds, 2011), and it is a nonpharmacological treatment that can be applied in settings where there are staffing and resource limitations (Kiosses & Alexopoulos, 2014).

PST guides recipients to a more realistic view of their problems and encourages them to articulate strategies to address these issues (Alexopoulos, Raue, & Areán, 2003). PST treatment consists of seven stages that address psychosocial problems. These stages are (1) selecting and defining the problem, (2) establishing realistic and achievable goals, (3) generating alternative solutions, (4) implementing decision making guidelines, (5) evaluating and choosing solutions, (6) implementing the preferred solution, and (7) evaluating the outcome.

TAU

TAU’s intensity and modality was not standardized in the MMT clinic, rather treatment varied depending on the patient and the clinic counselor’s background and training. Individual and group counseling were the primary therapeutic approaches utilized by the counselors at the MMT clinic. In individualized sessions, counselors used a variety of therapeutic approaches including gestalt therapy, person-centered therapy, and existential therapy. Group sessions focused on issues related to drug use, abstinence skills, relapse prevention, and recovery support services. TAU consisted of 3–4 counseling sessions per month.

Experimental condition: PST and TAU (PST + TAU)

The adaptation of PST for older adult opiate addicted populations (Rosen et al., 2011) focuses on the specific circumstances of participants in MMT. These circumstances include (1) being aware of other treatments, including both individual treatment histories and clinic-level issues, (2) managing neurocognitive deficits, and (3) recognizing the distinctive psychosocial issues of this population. Two master’s level clinically trained therapists certified in PST provided the treatment. The experimental condition consisted of PST in combination with the customary TAU provided by the methadone clinic staff. This was done in order to ensure that the methadone clinic was in compliance with federal-level regulations ensuring a minimum amount of counseling that each client at a methadone clinic must receive. Participants receiving PST + TAU received minimal paper handouts due to literacy issues and received individualized PST counseling in a private setting in a separate wing of the treatment facility. The initial PST session lasted an average of 1 hr with subsequent sessions lasting approximately 30 min. Treatment completion was based on the PST therapist’s assessment of the participant’s mastery of the seven stages of PST. This constituted the acute phase of the PST treatment.

To assure fidelity of the PST intervention, all sessions were taped and reviewed by a certified PST trainer. The PST trainer met weekly with the two master’s trained therapists providing the intervention to ensure that the intervention was being employed accurately. In addition, the PST therapists, PST trainer, and principal investigator of the study participated in a data safety monitoring review meeting where cases were reviewed by a larger panel of psychiatrists and therapists.

Outcome: Depressive Symptom Severity

The PHQ-9 is a 9-item scale used to identify a depressive disorder or to summarize the severity of depressive symptoms. The 9 items reflect the criteria for depression provided in the Diagnostic and Statistical Manual of Mental Disorders, fourth edition. Participants rate the frequency of each item on a 4-point scale (0 = not at all, 1 = several days, 2 = more than half the days, and 3 = nearly every day). Depressive symptom severity reflects the sum of the 9 items with a range of 0–27. A score of 10 or higher suggests major depression. The PHQ-9 has demonstrated reliability, validity, and good measures of sensitivity and specificity with a variety of different clinical populations (Kroenke, Spitzer, & Williams, 2001; Meader, Moe-Byrne, Llewellyn, & Mitchell, 2013). The PHQ-9 has demonstrated internal consistency and convergent validity with outpatients who abuse substances (Dum, Pickren, Sobell, & Sobell, 2008) and identified mild depression in residential clients with substance abuse diagnoses (Hepner, Hunter, Edelen, Zhou, & Watkins, 2009).

The PHQ-9 was used for all participant assessments during the acute phase of the treatment which includes pretreatment baseline, every 4 weeks postbaseline until the treatment period was complete, and a final acute stage assessment which was conducted 1 month after the completion of the treatment period. The subsequent durability phase included an assessment 2 months later (identified as 3-month assessment), and a second assessment was conducted 3 months later (identified as the 6-month assessment).

Randomization

After screening for eligibility and obtaining informed consent, a list of participants’ names was sent to an off-site coordinator; the off-site coordinator generated an identification number and randomly assigned participants to PST + TAU or to TAU. Random assignment was communicated by the off-site coordinator to the principal investigator who, in turn, provided a PST-trained study therapist with the participants’ names assigned to the PST + TAU group; the study therapist then contacted the randomized participant directly. This process ensured that the on-site project coordinator, responsible for screening assessments for both groups, was blinded from the assignment.

Statistical Analysis

A series of t-tests and χ² tests were conducted to test the success of the randomization to treatment group. To test the hypothesis that participants receiving PST + TAU would report lower depression symptom severity scores than participants receiving only TAU, a profile analysis was performed on the scores of the completers as a function of time and randomized group assignment. Profile analysis is the multivariate equivalent of a mixed analysis of variance and can be used when comparing the several dependent variables between groups over time (Tabachnick & Fidell, 2013). Profile analysis also produces a plot of the data which allows for a visual comparison of the patterns of effects. Three questions framed as null hypotheses can be asked of patterns resulting from a profile analysis: Do the combined PHQ-9 scores remain constant across time? Ignoring time, are the average PHQ-9 scores the same in both groups? Are the patterns of change in the PHQ-9 scores over time the same in the two groups? Significant effects result when any of these questions are rejected. For this study, all statistical tests were performed at the .05 level of significance. All analyses were conducted using general linear model repeated measures analysis in SPSS version 23.

Results

Participant Flow

Figure 1 presents the study flow diagram from prescreening contact, enrollment, and retention. There were 132 potential participants screened for the study. Among the 132 participants, 64 met the eligibility criteria: 41 had PHQ-9 scores 10 or higher and 23 had a score between 6 and 9 for two consecutive months. All 64 eligible participants agreed to participate in the study; 29 participants were randomized to PST + TAU and 35 participants were assigned to the TAU group.

Figure 1.
Study flow diagram: screening, enrollment, and retention.

Thirteen participants (20.3%) dropped out of the study following the completion of their baseline assessment and before the start of the acute phase of treatment: 11 from the PST + TAU group and 2 from the TAU group. Reasons for participant dropout prior to the imitation of treatment included nine who refused PST + TAU treatment, two who discontinued PST + TAU treatment, and two who discontinued TAU treatment. The acute phase (week 4 to 1-month posttreatment) began with 51 participants. During this phase, one participant dropped out of PST + TAU, two participants dropped out of TAU, and one participant in the TAU group died. Forty-seven participants were available during 3-month and 6-month durability phase posttests.

Participant Recruitment

Starting in December 2009 and ending in May 2012, the MMT clinic generated a list of clients age 50 years or older every 6 months. MMT clinic clients were initially contacted by a letter describing the study; the letter was also given to each potential participant by clinic staff as they arrived to the clinic. Clients who failed to contact the project coordinator were then given a recruitment card by the staff; the card provided a brief summary of the study and a reminder that they were eligible. The project coordinator would then attempt to contact the potential participant.

Upon contact with a potential participant, the on-site project coordinator scheduled an appointment and subsequently met with the potential participant for a baseline interview to determine eligibility. The master’s-level, blinded, on-site project coordinator received additional training in working with at-risk populations and supervision by qualified staff with training in social work and psychiatry. The on-site project coordinator consented all subjects and conducted all interviews with participants to collect assessment data throughout the duration of the study.

The MMT clinic provided separate offices on different floors to the project coordinator and the counselors providing PST to reduce barriers to participation in the study. The offices were located in a private area away from where the general clinic treatment was conducted.

A total of 47 subjects comprised the analytic sample. Seventeen participants comprised the PST + TAU group and 30 participants comprised the TAU group. Analysis was based on complete cases and the one missing value from a TAU participant was replaced using mean imputation.

Comparability of Randomization

Baseline and demographic characteristics of each group can be found on Tables 1 and 2. The PST + TAU and TAU groups were similar on depressive symptom severity scores and most demographics at baseline (see Tables 1 and 2). Participants in the PST + TAU group were more likely to live alone than those participants assigned to TAU (51.7% vs. 20%; p < .01). However, this was determined to be a weak negative association, φ = −.33. The average depressive symptom severity scores for both groups exceeded 10 suggesting that the two groups were both statistically and clinically similar on this outcome variable.

Table 1.
Sample Characteristics and χ² Difference Tests Between Randomized Groups.

Variable All Participants Attrition Group Analytic Group

TAU (%) PST + TAU (%) χ²/φ TAU (%) PST + TAU (%) χ² TAU (%) PST + TAU (%) χ²

(n = 35) (n = 29) (n = 5) (n = 12) (n = 30) (n = 17)

Race 0.16 1.0 0.01

Black 57.1 (20) 62.1 (18) 40.0 (2) 66.7 (8) 60.0 (18) 58.8 (10)

White 42.9 (15) 37.9 (11) 60.0 (3) 33.3 (4) 40.0 (12) 41.2 (7)

Gender 0.07 0.73 0.08

Male 68.6 (24) 65.5 (19) 80.0 (4) 58.3 (7) 66.7 (20) 70.6 (12)

Female 31.4 (11) 34.5 (10) 20.0 (1) 41.7 (5) 33.3 (10) 29.4 (5)

Employment status 0.23 2.55 0.16

Unemployed 85.7 (30) 89.7 (26) 80.0 (4) 100 (12) 86.7 (26) 82.4 (14)

Employed 14.3 (5) 10.3 (3) 20.0 (1) 0 (0) 13.3 (4) 17.6 (3)

Education 0.16 0.46 0.14

<High school grad 28.6 (10) 24.1 (7) 20.0 (1) 8.3 (1) 30.0 (9) 35.3 (6)

High school grad or more 71.4 (25) 75.9 (22) 80.0 (4) 91.7 (11) 70.0 (21) 64.7 (11)

Income 0.16 3.19 0.21

<$10,000 71.4 (25) 75.9 (22) 40.0 (2) 83.3 (10) 76.7 (23) 70.6 (12)

$10,000+ 28.6 (10) 24.1 (7) 60.0 (3) 16.7 (2) 23.3 (7) 29.4 (6)

Living arrangement 7.08/−.33 2.08 3.80

Alone 20.0 (7) 51.7 (15) 20.0 (1) 58.3 (7) 20.0 (6) 47.1 (8)

With other 80.0 (28) 48.3 (14) 80.0 (4) 41.7 (5) 80.0 (24) 52.9 (9)

Note.* TAU = treatment as usual; PST = problem-solving therapy.

p < .01.

Table 2.
Continuous Variables and Randomization.

Variable All Participants Attrition Group Analytic Group

TAU (%) PST + TAU (%) t(df) TAU (%) PST + TAU (%) t(df) TAU PST + TAU (%) t(df)

(n = 35) (n = 29) (n = 5) (n = 12) (n = 30) (n = 17)

M (SD) M (SD) M (SD) M (SD) M (SD) M (SD)

PHQ-9 total score 12.20 (3.7) 11.59 (3.1) .72(62) 12.00 (1.2) 11.67 (3.6) .20(15) 12.23 (3.9) 11.53 (2.8) .65(45)

Age 56.37 (5.4) 56.00 (4.89) .29(62) 54.60 (5.0) 54.67 (5.5) −.03(15) 56.67 (5.5) 56.94 (5.6) −.16(45)

Note. SD =* standard deviation; TAU = treatment as usual; PST = problem-solving therapy.

Attrition

The average PHQ-9 score for all participants was lower than the overall TAU group baseline score. Among all participants, the PST + TAU group consisted of persons who were Black, male, unemployed, with more than a high school education, annual incomes less than US$10,000, and lived alone. In total, about 30% (17) of the participants withdrew from the study with over two thirds of the attrition coming from the PST + TAU group (Table 1). While the PST + TAU group lost 12 participants (41.4%) and the TAU group lost 5 participants (14.3%) post randomization, there was no difference in attrition once participants agreed to initiate treatment (see Figure 1). Despite these patterns of attrition, the analytic groups’ composition did not differ statistically on either the demographic differences (Table 1) or the depressive symptom severity score (Table 2).

Assessment of PST + TAU in Comparison to TAU

A profile analysis was performed on depression severity scores over time as a function of group assignment (PST + TAU or TAU). Depression scores were assessed at four time points (baseline, 1-month posttreatment, 3-month durability, and 6-month durability). The assumption of homogeneity of covariance matrices was met, Box’s M = 9.586, F(10, 5168.23) = 0.854, p = .576. All other assumptions were met.

There was a significant decrease in average depression scores across time (Pillai’s trace = .317), F(3, 43) = 6.639, p = .001, η² _p = .317, η² = .13. This significant decrease in the combined averaged PST + TAU and TAU PHQ-9 scores over time can be seen in Figure 2 and Table 3. Results of the equal levels test demonstrated no significant difference between group main effects. That is, ignoring time there was no difference in average PHQ-9 scores between the two groups, F(1, 45) = 1.467, p = .232, η² _p = .032, η² < .004. Results of the profile analysis demonstrated no statistically significant interaction pattern, that is, depression scores in each segment of time were not statistically significantly different (Pillai’s trace = .07, F(3, 43) = 1.014, p = .396, η² _p = .07, η² = .02). Inspection of Figure 2, however, suggests that at 7-month posttreatment average PHQ-9 scores in the PST + TAU group continued to decline, whereas in the TAU group scores are increasing, returning to baseline levels. Though not statistically significant, this pattern is evident from the changes in the Cohen’s effect size values between PST + TAU and TAU at each assessment point: baseline, d = .21, 1-month posttreatment, d = .11, 3-month durability, d = .26, and 6-month durability, d = .57. Thus, following the acute phase there is an increasing effect size between PHQ-9 scores reaching a moderate level at the last assessment.

Figure 2.
Estimated marginal means across time.

Table 3.
Depression Severity Scores by Treatment Group Type Over Time.

Group Assignment Time M (Standard Deviation) Cohen’s d (Problem-Solving Therapy vs. TAU) 95% Confidence Interval

Lower Bound Upper Bound

TAU Baseline 12.23 (3.93) 10.92 13.54

1-Month posttreatment 9.60 (4.52) 7.88 11.33

3-Month durability 9.93 (4.27) 8.34 11.52

6-Month durability 10.77 (5.06) 8.98 12.55

Problem-solving therapy + TAU Baseline 11.53 (2.79) .21 9.79 13.27

1-Month posttreatment 9.06 (4.98) .11 6.77 11.35

3-Month durability 8.82 (4.41) .26 6.71 10.93

6-Month durability 8.06 (4.47) .57 5.69 10.43

Note. TAU = treatment as usual.

Discussion and Applications to Practice

The objective of this study was to examine whether using PST in combination with TAU (PST + TAU) would reduce depressive symptom severity scores in older MMT participants. Thirteen percent of the group differences are accounted for by time, whereas randomization and the interaction did not significantly account for any of the variance in effects. While there was no statistically significant nonparallelism between groups over time, all participants had lower depressive symptom severity scores at the 6-month postintervention measurement than baseline (Figure 2). The depression scores were lower for both groups throughout the acute phase of the study. Yet during the durability phase, the TAU group scores began to increase toward baseline levels, whereas the PST + TAU group depression scores continued to decrease (Figure 2).

The question as to whether PST in combination with TAU contributed to a decline in depressive symptom severity scores is still left open. This study focused on the dependent variable of depressive symptoms as the outcome of interest. However, the reality of the participants’ lives was one of extensive mental and physical health challenges coupled with tremendous poverty. Depressive symptoms were just one of a myriad of challenges faced by individuals who have been addicted to drugs for decades. The PST therapists were often overwhelmed by the complexity of the participant’s life situations, and felt that the frequency that they were able to see each participant was not sufficient. Future modifications to PST for this population may need to incorporate more frequent sessions than provided in this study’s current protocol.

Additionally, the decline in depressive symptom severity scores for both groups may be due to a Hawthorne effect. Both groups participated in regular assessments for which their time was compensated with a US$25.00 gift card. It may be that this attention and compensation for the assessments played a role in reducing feelings of isolation, thereby reducing their depressive symptom severity scores at the assessments. This pattern continued up to the last data collection time point and it is perhaps at this point the true effect of PST was found.

Another possible explanation for the decrease in depressive symptom severity scores across both groups is that both groups continued to receive TAU at the MMT clinic. Federal regulations requiring counseling prohibited the research team from modifying any existing treatment that the participants were receiving. The range of treatments that participants received at the clinic was quite varied. Often, the treatment would be driven by the clinical training of a particular counselor rather than the stated needs of the individual. It was also not uncommon for a clinic client to “bounce around” from one type of treatment to another based on the time of day that they came to the clinic and the availability of a given counselor or therapeutic group. The variation in TAU made it impossible for the study team to measure its impact. Therefore, it can also not be discounted that TAU may have had some impact on clinical outcomes yet the added value of having received PST seemed to sustain a decreased depression score at the 6-month posttreatment assessment.

High attrition rates pose a challenge for conducting randomized control trials with consumers of methadone maintenance programs. While some attrition in randomized control trials is largely unavoidable due to the nature of working with human subjects, high rates of attrition can lead to bias in the findings particularly if there is differential attrition between the treatment and control groups. The overall attrition in the current study (30%) was lower than is often found in studies of people who have histories of substance use; while attrition rates vary, they can be as high as 50% in the first 30 days and 80% in the first year (Mancino et al., 2010). In our study, higher dropout rate (41%) occurred among those participants who were assigned to the PST + TAU group versus those in the TAU group (14%), but this occurred following randomization even before active treatment began. Feedback from those participants who dropped out indicated that they did not realize the extent of the effort that would be required to take part in the PST protocol. Participants were not compensated financially for attending PST sessions and the extra time required of participants often interfered with work and family obligations. Once a participant agreed to engage in the PST sessions, there was very little attrition. Future studies will need to be more explicit in the time requirement for participants to be enrolled in a PST protocol.

Much of the attrition in the experimental group occurred prior to the initiation of the actual PST treatment; specifically, of the 12 participants in the experimental group who attrited during the course of the study, only one attrited once the intervention began. In other words, of the 18 participants who began the PST, only 1 dropped out. Theoretically, we had hoped that the use of PST could allow the participant to identify the problems to address, and thus would have reduced attrition. Interestingly, once treatment started the attrition rate was the same in the two groups. It may have also been the case that the small financial incentives given for completing the assessments (not the PST sessions) played a role in retaining individuals in treatment. Broadly, this suggests that patients receiving MMT are willing to participate in additional treatment once the treatment commences.

There were several limitations that should be considered. First, the attained sample could have been too small to find statistically significant differences between the two groups even though the differences were clinically meaningful. Second, the sample size, particularly after attrition, limited our ability to examine within group differences. It will be important for future studies to have sufficient sample sizes to address these issues. It is also important to note that the study population was drawn from a clinic sample. Our sample represented individuals who were ready to engage in treatment. The effectiveness of this intervention on older adults with opioid addiction problems who are not engaged in treatment is unknown. Next, this study used a completer analysis instead of an intention-to-treat analysis. Intention-to-treat analysis is possible when complete outcome data are available for all randomized participants. In our study, no outcomes were assessed for participants who did not initiate treatment after randomization. Finally, there was no consistency about the nature of TAU at the MMT clinic and no information about the nature of what services were actually provided to either group with regard to this treatment.

Despite these limitations, in the long term, the PST intervention may be able to reduce depressive symptom scores with older adults patients in MMT. The purpose of PST is to teach the problem-solving technique to the patient to use on his/her own, not to use it as an open-ended, ongoing therapy. Thus, the gap seen in depression scores at 6-month posttreatment may demonstrate the effectiveness in teaching participants skills that are cognitively based, focused on the individuals’ sense of their problems, and that can be used independently. A larger, multisite study is warranted based on these initial findings.

Variable	All Participants	Attrition Group	Analytic Group
Race			0.16			1.0			0.01
Black	57.1 (20)	62.1 (18)		40.0 (2)	66.7 (8)		60.0 (18)	58.8 (10)
White	42.9 (15)	37.9 (11)		60.0 (3)	33.3 (4)		40.0 (12)	41.2 (7)
Gender			0.07			0.73			0.08
Male	68.6 (24)	65.5 (19)		80.0 (4)	58.3 (7)		66.7 (20)	70.6 (12)
Female	31.4 (11)	34.5 (10)		20.0 (1)	41.7 (5)		33.3 (10)	29.4 (5)
Employment status			0.23			2.55			0.16
Unemployed	85.7 (30)	89.7 (26)		80.0 (4)	100 (12)		86.7 (26)	82.4 (14)
Employed	14.3 (5)	10.3 (3)		20.0 (1)	0 (0)		13.3 (4)	17.6 (3)
Education			0.16			0.46			0.14
<High school grad	28.6 (10)	24.1 (7)		20.0 (1)	8.3 (1)		30.0 (9)	35.3 (6)
High school grad or more	71.4 (25)	75.9 (22)		80.0 (4)	91.7 (11)		70.0 (21)	64.7 (11)
Income			0.16			3.19			0.21
<$10,000	71.4 (25)	75.9 (22)		40.0 (2)	83.3 (10)		76.7 (23)	70.6 (12)
$10,000+	28.6 (10)	24.1 (7)		60.0 (3)	16.7 (2)		23.3 (7)	29.4 (6)
Living arrangement			7.08*/−.33			2.08			3.80
Alone	20.0 (7)	51.7 (15)		20.0 (1)	58.3 (7)		20.0 (6)	47.1 (8)
With other	80.0 (28)	48.3 (14)		80.0 (4)	41.7 (5)		80.0 (24)	52.9 (9)

Variable	All Participants	Attrition Group	Analytic Group
PHQ-9 total score	12.20 (3.7)	11.59 (3.1)	.72(62)	12.00 (1.2)	11.67 (3.6)	.20(15)	12.23 (3.9)	11.53 (2.8)	.65(45)
Age	56.37 (5.4)	56.00 (4.89)	.29(62)	54.60 (5.0)	54.67 (5.5)	−.03(15)	56.67 (5.5)	56.94 (5.6)	−.16(45)

Group Assignment	Time	M (Standard Deviation)	Cohen’s d (Problem-Solving Therapy vs. TAU)	95% Confidence Interval
TAU	Baseline	12.23 (3.93)		10.92	13.54
1-Month posttreatment	9.60 (4.52)		7.88	11.33
3-Month durability	9.93 (4.27)		8.34	11.52
6-Month durability	10.77 (5.06)		8.98	12.55
Problem-solving therapy + TAU	Baseline	11.53 (2.79)	.21	9.79	13.27
1-Month posttreatment	9.06 (4.98)	.11	6.77	11.35
3-Month durability	8.82 (4.41)	.26	6.71	10.93
6-Month durability	8.06 (4.47)	.57	5.69	10.43

Footnotes

Authors’ Note

Review protocol and trial registration are available through the corresponding author.

Declaration of Conflicting Interests

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

Funding

The authors disclosed receipt of the following financial support for the research, authorship, and/or publication of this article: Dr. McCall was supported with use of facilities at the Pittsburgh, VA, and the Office of Academic Affiliations Advanced Fellowship in Women’s Health. The contents of this manuscript do not represent the views of the U.S. Department of Veterans Affairs or the United States Government. Funding for this study was provided by NIDA Grant K08 DA021570-03 (PI: Daniel Rosen). NIDA had no role in the study design, collection, analysis or interpretation of the data, writing the article, or the decision to submit the article for publication.

References

Alexopoulos

G. S.

Raue

Areán

(2003). Problem-solving therapy versus supportive therapy in geriatric major depression with executive dysfunction. The American Journal of Geriatric Psychiatry, 11, 46–52.

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (DSM-5^®). Chicago: American Psychiatric Pub.

Areán

P. A.

Hegel

M. T.

Reynolds

C. F.

(2001). Treating depression in older medical patients with psychotherapy. Journal of Clinical Geropsychology, 7, 93–104.

Choi

N. G.

DiNitto

D. M.

Marti

C. N.

(2014). Treatment use, perceived need, and barriers to seeking treatment for substance abuse and mental health problems among older adults compared to younger adults. Drug and Alcohol Dependence, 145, 113–120.

Delgadillo

Gore

Ali

Ekers

Gilbody

Gilchrist

… Hughes

(2015). Feasibility randomized controlled trial of cognitive and behavioral interventions for depression symptoms in patients accessing drug and alcohol treatment. Journal of Substance Abuse Treatment, 55, 6–14. doi:https://dx-doi-org.web.bisu.edu.cn/10.1016/j.jsat.2015.02.008

Dum

Pickren

Sobell

L. C.

Sobell

M. B.

(2008). Comparing the BDI-II and the PHQ-9 with outpatient substance abusers. Addictive Behaviors, 33, 381–387.

Gellis

Z. D.

Kenaley

(2008). Problem-solving therapy for depression in adults: A systematic review. Research on Social Work Practice, 18, 117–131.

Han

Polydorou

Ferris

Blaum

C. S.

Ross

McNeely

(2015). Demographic trends of adults in New York City opioid treatment programs—An aging population. Substance Use & Misuse, 50, 1660–1667.

Hendrie

H. C.

Lindgren

Hay

D. P.

Lane

K. A.

Gao

Purnell

,… Boustani

M. A.

(2013). Comorbidity profile and healthcare utilization in elderly patients with serious mental illnesses. The American Journal of Geriatric Psychiatry, 21, 1267–1276.

10.

Hepner

K. A.

Hunter

S. B.

Edelen

M. O.

Zhou

A. J.

Watkins

(2009). A comparison of two depressive symptomatology measures in residential substance abuse treatment clients. Journal of Substance Abuse Treatment, 37, 318–325.

11.

H. M.

Kline

Huang

F. Y.

Ziedonis

D. M.

(2006). Detection of co-occurring mental illness among adult patients in the New Jersey substance abuse treatment system. American Journal of Public Health, 96, 1785–1793. doi:10.2105/AJPH.2005.072736

12.

Kiosses

D. N.

Alexopoulos

G. S.

(2014). Problem-solving therapy in the elderly. Current Treatment Options in Psychiatry, 1, 15–26.

13.

Kroenke

Spitzer

R. L.

Williams

J. B. W.

(2001). The PHQ-9: Validity of a brief depression severity measure. Journal of General Internal Medicine, 16, 606–613.

14.

Kuerbis

Sacco

Blazer

D. G.

Moore

A. A.

(2014). Substance abuse among older adults. Clinics in Geriatric Medicine, 30, 629–654.

15.

Levy

J. A.

Anderson

(2005). The drug career of the older injector. Addiction Research & Theory, 13, 245–258. doi:10.1080/16066350500053554

16.

Lofwall

M. R.

Brooner

R. K.

Bigelow

G. E.

Kindbom

Strain

E. C.

(2005). Characteristics of older opioid maintenance patients. Journal of Substance Abuse Treatment, 28, 265–272. doi:10.1016/j.jsat.2005.01.007

17.

Mackin

R. S.

Arean

P. A.

(2005). Evidence-based psychotherapeutic interventions for geriatric depression. Psychiatric Clinics of North America, 28, 805–820.

18.

Mancino

Curran

Han

Allee

Humphreys

Booth

B. M.

(2010). Predictors of attrition from a national sample of methadone maintenance patients. The American Journal of Drug and Alcohol Abuse, 36, 155–160.

19.

Meader

Moe-Byrne

Llewellyn

Mitchell

A. J.

(2014). Screening for poststroke major depression: A meta-analysis of diagnostic validity studies. Journal of Neurology, Neurosurgery & Psychiatry, 85, 198–206.

20.

Rosen

Hunsaker

Albert

S. M.

Cornelius

J. R.

Reynolds

C. F.

(2011). Characteristics and consequences of heroin use among older adults in the United States: A review of the literature, treatment implications, and recommendations for further research. Addictive Behaviors, 36, 279–285.

21.

Rosen

Smith

M. L.

Reynolds

C. F.

(2008). The prevalence of mental and physical health disorders among older methadone patients. The American Journal of Geriatric Psychiatry, 16, 488–497.

22.

Sahker

Schultz

S. K.

Arndt

(2015). Treatment of substance use disorders in older adults: Implications for care delivery. Journal of the American Geriatrics Society, 63, 2317–2323. doi:10.1111/jgs.13706

23.

Searby

Maude

McGrath

(2015). An ageing methadone population: A challenge to aged persons’ mental health services? Issues in Mental Health Nursing, 36, 927–931. doi:10.3109/01612840.2015.1081655

24.

Stein

M. D.

Anderson

(2003). Injection frequency mediates health service use among persons with a history of drug injection. Drug and Alcohol Dependence, 70, 159–168.

25.

Tabachnick

B. G.

Fidell

L. S.

(2013). Using multivariate statistics (6th ed.). Boston, MA: Pearson.

26.

Volkow

N. D.

(2004). The reality of comorbidity: Depression and drug abuse. Biological Psychiatry, 56, 714–717.

27.

L.-T.

Blazer

D. G.

(2014). Substance use disorders and psychiatric comorbidity in mid and later life: A review. International Journal of Epidemiology, 43, 304–317.

Variable	All Participants			Attrition Group			Analytic Group
	TAU (%)	PST + TAU (%)	χ²/φ	TAU (%)	PST + TAU (%)	χ²	TAU (%)	PST + TAU (%)	χ²
	(n = 35)	(n = 29)	χ²/φ	(n = 5)	(n = 12)	χ²	(n = 30)	(n = 17)	χ²
Race			0.16			1.0			0.01
Black	57.1 (20)	62.1 (18)		40.0 (2)	66.7 (8)		60.0 (18)	58.8 (10)
White	42.9 (15)	37.9 (11)		60.0 (3)	33.3 (4)		40.0 (12)	41.2 (7)
Gender			0.07			0.73			0.08
Male	68.6 (24)	65.5 (19)		80.0 (4)	58.3 (7)		66.7 (20)	70.6 (12)
Female	31.4 (11)	34.5 (10)		20.0 (1)	41.7 (5)		33.3 (10)	29.4 (5)
Employment status			0.23			2.55			0.16
Unemployed	85.7 (30)	89.7 (26)		80.0 (4)	100 (12)		86.7 (26)	82.4 (14)
Employed	14.3 (5)	10.3 (3)		20.0 (1)	0 (0)		13.3 (4)	17.6 (3)
Education			0.16			0.46			0.14
<High school grad	28.6 (10)	24.1 (7)		20.0 (1)	8.3 (1)		30.0 (9)	35.3 (6)
High school grad or more	71.4 (25)	75.9 (22)		80.0 (4)	91.7 (11)		70.0 (21)	64.7 (11)
Income			0.16			3.19			0.21
<$10,000	71.4 (25)	75.9 (22)		40.0 (2)	83.3 (10)		76.7 (23)	70.6 (12)
$10,000+	28.6 (10)	24.1 (7)		60.0 (3)	16.7 (2)		23.3 (7)	29.4 (6)
Living arrangement			7.08*/−.33			2.08			3.80
Alone	20.0 (7)	51.7 (15)		20.0 (1)	58.3 (7)		20.0 (6)	47.1 (8)
With other	80.0 (28)	48.3 (14)		80.0 (4)	41.7 (5)		80.0 (24)	52.9 (9)

Variable	All Participants			Attrition Group			Analytic Group
	TAU (%)	PST + TAU (%)	t(df)	TAU (%)	PST + TAU (%)	t(df)	TAU	PST + TAU (%)	t(df)
	(n = 35)	(n = 29)		(n = 5)	(n = 12)		(n = 30)	(n = 17)
	M (SD)	M (SD)		M (SD)	M (SD)		M (SD)	M (SD)
PHQ-9 total score	12.20 (3.7)	11.59 (3.1)	.72(62)	12.00 (1.2)	11.67 (3.6)	.20(15)	12.23 (3.9)	11.53 (2.8)	.65(45)
Age	56.37 (5.4)	56.00 (4.89)	.29(62)	54.60 (5.0)	54.67 (5.5)	−.03(15)	56.67 (5.5)	56.94 (5.6)	−.16(45)