Search tips
Search criteria 


Logo of nihpaAbout Author manuscriptsSubmit a manuscriptHHS Public Access; Author Manuscript; Accepted for publication in peer reviewed journal;
J Clin Psychol. Author manuscript; available in PMC 2013 December 25.
Published in final edited form as:
PMCID: PMC3872485

Family Treatment for Bipolar Disorder and Substance Abuse in Late Adolescence


The initial onset of bipolar disorder occurs in childhood or adolescence in about 50% of patients. Early-onset forms of the disorder have a poorer prognosis than adult-onset forms and are frequently characterized by comorbid substance abuse. Clinical trials research suggests that family psychoeducational approaches are effective adjuncts to medication in stabilizing the symptoms of bipolar disorder in adults and youth, although their efficacy in patients with comorbid substance use disorders has not been systematically investigated. This article describes the family-focused treatment (FFT) of a late adolescent with bipolar disorder and polysubstance dependence. The treatment of this patient and family required adapting FFT to consider the family’s structure, dysfunctional alliance patterns, and unresolved conflicts from early in the family’s history. The case illustrates the importance of conducting manual-based behavioral family treatments with a psychotherapeutic attitude, including addressing unstated emotional conflicts and resistances that may impede progress.

Keywords: expressed emotion, family-focused treatment, drug abuse, lithium, psychoeducation, structural family therapy

Bipolar disorder (BD) often has its first onset in late childhood or adolescence. Although there is substantial evidence for the effectiveness of pharmacotherapy in early-onset BD, many children and adolescents have multiple recurrences and psychosocial impairment, as well as substantial medication side effects. There is increasing evidence for the efficacy of adjunctive psychosocial treatments for both adult and pediatric bipolar disorder (Miklowitz, 2008).

Approximately 50%–67% of adults with BD report onset of their first episode before the age of 18 years and between 15% and 28% before the age of 13 years (Perlis et al., 2004). Patients with early onset (before 18 years of age) have more severe courses of illness, more switches of mood polarity, longer episodes, more comorbid disorders, and more mixed episodes than adult-onset BD patients. Each of these clinical features bodes poorly for long-term outcome (Post et al., 2010).

There has been a dramatic increase in the community diagnosis of early-onset BD in the past 10 years, leading some to question whether the diagnosis is a fad. There is indeed a fair amount of slippage in the use of the Diagnostic and Statistical Manual, Fourth Edition (DSM-IV) diagnostic criteria in the community. However, contrary to popular belief, clinical studies find that pediatric-onset BD is equally prevalent across countries and is not elevated in frequency in the United States, with a worldwide prevalence of 1.8% (Van Meter, Moreira, & Youngstrom, 2011).

Adolescents with BD are approximately five times more likely to also develop substance abuse disorders (SUDs) than healthy controls. Those with both BD and SUD are at an increased risk for suicide attempts, more frequent relapses of mood disorder, and poor psychosocial functioning (Goldstein et al., 2008). In some cases, the SUD comes before the initial onset of mood symptoms, and in other cases (such as the present case), the reverse. It can be quite difficult to determine whether an adolescent’s symptoms are the result of mood disorder, drugs, or both: It appears that each disorder worsens the other.

This article focuses on family-focused treatment (FFT) with an adolescent suffering from both BD and substance abuse disorder. The treatment was conducted in conjoint sessions with the patient, his parents, and siblings. As in all cases of FFT, the structure involved sessions of psychoeducation about BD, communication enhancement training, and problem-solving skills training. To effectively administer these components, the clinicians had to be flexible and sensitive to a number of underlying emotional conflicts in this family, which, they believed, stood in the way of real change.

Family-Focused Treatment

The treatment of pediatric-onset BD parallels the treatment of adults, although medication dosages (typically, mood stabilizers and/or atypical antipsychotics with adjunctive antidepressants or anxiolytics) are adjusted for age and body weight. FFT, which is provided in conjunction with pharmacotherapy, aims to teach patients and family members about the causes, prognosis, and self-management of bipolar disorder; how to communicate more effectively and learn skills to solve family problems; how to accept the necessity of ongoing treatment; and how to maximize the patient’s likelihood of success in reengaging with the community, school, or workplace.

FFT is conducted in four stages: an engagement phase, in which the objective is to connect with the patient and parents (and, where possible, siblings) and relay information about the treatment’s structure and expectations; psychoeducation, in which therapists lead the family in discussions of the nature, causes, and management of BD symptoms; communication enhancement training, in which patients and family members rehearse effective speaking and listening skills (e.g., how to give praise and constructive criticism and how to listen actively); and problem-solving skills training, in which patients and family members define, generate and evaluate, and implement solutions to problems in the family’s or the patient’s life. The treatment is given in 21 sessions (12 weekly, 6 biweekly, 3 monthly) over 9 months. When practiced in the community, clinicians and families sometimes opt for shorter versions or longer intervals between sessions.

FFT has been recognized as an empirically supported treatment for BD by the American Psychological Association’s Society for Clinical Psychology, and it is listed as the only one of five therapies to have strong research support for the treatment of bipolar depression ( It has also been included in a shortlist of exemplary empirically supported treatments (Baker, McFall, & Shoham, 2008) and is cited as a preferred psychosocial intervention in most of the psychiatric treatment guidelines for BD in adults or children. There are four randomized trials supporting the effects of FFT in combination with pharmacotherapy in preventing recurrence or stabilizing symptoms (Miklowitz & Scott, 2009) after an episode of BD I or II. One large-scale community effectiveness study indicated the benefits of adjunctive FFT, interpersonal and social rhythm therapy, and cognitive-behavioral therapy in hastening time to recovery and elongating periods of wellness in adults with bipolar I or II depression relative to brief care (Miklowitz et al., 2007). A two-site trial of adolescents with BD indicated that FFT and pharmacotherapy were more effective than brief psychoeducation and pharmacotherapy in reducing time to recovery from depressive symptoms, the amount of time in depressive episodes, and the severity of inter-episode depressive symptoms over 2 years (Miklowitz et al., 2008).

Theoretical Basis of FFT

The symptoms of BD affect the emotional responses of caregiving family members, and these emotional responses affect the reactions of the bipolar family member and ultimately the course of the illness. As in other disorders, high expressed emotion (EE) attitudes (high levels of criticism, presence of hostility, or high levels of emotional overinvolvement) in parents of patients with BD are associated with higher rates of recurrence and more severe symptoms compared with low EE attitudes in parents (Miklowitz, 2007).

High-EE attitudes among parents of bipolar offspring may have their roots in childhood dyadic interactions. As shown in Figure 1, people who later develop BD, especially those who grow up with a parent with BD I or II, were often irritable, aggressive, and emotionally reactive as children, and they may have been paired with parents with whom there was a poor temperamental fit. The caregiving parent may or may not have the disorder, but usually finds it increasingly difficult to deal with the child’s temper outbursts, inability to self-soothe, violence, poor school performance, and low social competence. Even though these childhood features may reflect the early expression of a genetic vulnerability, the dyadic interactions between parents and the at-risk child may become increasingly negative over time, especially as the child reaches adolescence and strives for more autonomy. When he or she first develops mood disorder symptoms, or even difficulty with emotional self-regulation, the stage has been set for increasingly volatile family interactions that contribute to the onset and subsequent cyclic course of BD.

Figure 1
Developmental pathways to improvement versus adverse outcomes among youth at risk for bipolar disorder.

We postulate a similar bidirectional relationship between parental EE and patients’ bipolar mood symptoms after the onset of the illness (Miklowitz, 2007). The pathway begins with postepisode residual symptoms in the patient, such as depression, withdrawal, impulsiveness, low functioning, or irritability, all of which contribute to the intensity of the patient’s reactions to caregivers who are attempting to help the patient manage the illness. Escalating negative family interactions reduce the threshold for caregivers to react with fear, frustration, and hopelessness; recall and exaggerate negative experiences from prior illness episodes; and make attributions of controllability and negative predictions about the future (e.g., “He’s doing this to hurt me”; “I’ll always have to take care of her”). This cognitive reactivity of the parent may fuel his or her expression of high-EE attitudes toward the patient, which creates an aversive, stressful environment for the patient and contributes to temporary exacerbations of mood symptoms and a worsening pattern of dyadic interaction. In combination with biological and genetic vulnerability factors, repeated exposure to negative family interactions may contribute to the patient’s overall liability to new episodes or recurrences.

As indicated in Figure 1, the optimal time to intervene in these processes may be when a child is showing early signs of the disorder. Typically, this means that the child has only had one or two episodes of mania or, more frequently, has had depression and several brief periods of mania or hypomania that are impairing or at least noticeable to others but do not meet the duration criteria for full episodes. Intervening before the onset of the disorder or even after the first or second episode may reduce the frequency of subsequent episodes and their severity.

FFT assists the parents, child, and available siblings in how to identify early warning signs of new episodes and develop strategies for preventing the episode or limiting its effects on functioning (psychoeducation). It also includes skill training to improve day-to-day interactions and solve problems in the current family environment. These skill-based components may lead to the greater use of emotion regulation strategies in the patient.

To encourage acceptance of the disorder and foster effective emotional communication, clinicians administering FFT must adopt a psychotherapeutic attitude, a sensitivity to underlying emotional conflicts. This aspect of treatment was apparent to us from the beginning, when we discovered that the rule-governed, skill-training tasks typical of behavioral treatments for schizophrenia were unwelcome to bipolar patients and their families. Interactions in these families were highly emotionally charged and members seemed to thrive on chaos, humor, and unpredictability. Early on, we hypothesized that highly emotional family interactions, especially when members were discussing seemingly minor problems, signaled underlying conflicts that needed to be brought into the open and discussed. This case below illustrates a family with one such emotional conflict: anger about paternal neglect in childhood.

To address these family processes, we have incorporated several structural therapy strategies (Minuchin & Fishman, 1981) into FFT, such as strengthening the alliances between the child and the more disengaged parent or emphasizing the boundaries between the parental and child subsystems. We also incorporate the use of reframing and relabeling. For example, we often label the aversive behavior of the child as uncontrollable when parents believe it to be controllable, or the high level of parent-offspring conflict as normative when a close relationship is disrupted by the challenges of the postepisode recovery period.

The version of FFT used in the case illustration also reflected adaptations suggested by our colleague, Ben Goldstein, M.D., of the University of Toronto, who has re-manualized FFT for dual-diagnoses adolescents with BD (Goldstein et al., 2008). The goals of the FFT-SUD approach are as follows: (a) promoting substance-free homes in which drug or alcohol use by parents is addressed as a risk factor; (b) challenging the notion that substance abuse in BD is self-medication and framing it instead as a health-compromising behavior; and (c) using problem solving to identify high-risk situations in which the adolescent may be exposed to substances and develop strategies for avoiding these situations or managing the resulting cravings.

Case Illustration

Presenting Problem and Client Description

Drew, an 18 year old, lived with his mother Angie, age 42, sister Maddy, 20, and older brother Daniel, 22, in a suburban, middle-class community. He was referred to a pediatric mood disorders clinic for severe drug abuse that his therapist had said might be masking BD. The original referral, made just as the school year was ending, requested only a diagnostic evaluation, although the family made it clear from the outset that they wanted treatment. In the prior 3 months, Drew had been on “lockdown” at his mother’s house pending verification of several months of sobriety, as revealed by home-based urinary assays. The event that stimulated the referral was a relapse of his marijuana use and increased verbalization of suicidal thoughts. His biological father Aaron, age 42, was a musician and lighting specialist who lived in another city and was only intermittently involved in his life. Aaron attended the intake sessions.

In separate individual interviews, Drew and his parents described a highly destructive pattern of drug abuse involving oxycontin, cocaine, methamphetamine, and marijuana. He began smoking marijuana and using methamphetamine at age 14, although his worst patterns of drug abuse had occurred in the year prior to intake. Before his lockdown, he had been disappearing late at night to look for fixes; his urinary drug screens were consistently positive for opiates and marijuana.

Drew presented as an insightful and agreeable but depressed and tearful young man. He mourned the damage he had done to his life and expressed shame and guilt for hurting his mother, father, and siblings. His parents agreed that Drew had always been moody, but that his moodiness had been much worse in the past 6 months. He had received minimal treatment. Individual sessions with a psychotherapist at a city clinic, he said, had been “OK, helpful at times.” He had attended a few Narcotics Anonymous meetings but complained that he could not relate to them; no one was his age and “the ‘addiction is an illness’ thing doesn’t really fit me.” He had never taken a psychiatric medication.

The diagnostic assessment included the Kiddie-Schedule for Affective Disorders and Schizophrenia-Present and Lifetime Version (KSADS-PL), administered by a psychologist and social worker that separately interviewed Drew and each of his parents. Drew described a 6-month period of depression that had begun the previous fall and included intensely sad mood, social withdrawal, suicide preoccupations, fatigue, hypersomnia, loss of appetite, and feelings of worthlessness. During the majority of this interval, he had been using marijuana but no harder drugs. According to his mother, his mood and energy level had shifted in the spring, during which he experienced a 1-month period of high mood, rapid thinking and speech, increased energy, decreased need for sleep, and grandiose thinking (consumed with plans concerning elaborate money-making schemes for nuclear waste storage). He acknowledged this manic period as out of the ordinary for him and denied taking any drugs at first, although he had begun taking benzodiazepines (obtained on the street) after going several weeks with minimal sleep. After multiple truancies during this period, he was expelled from school in the second semester of his junior year and began a home-schooling program.

A pharmacological evaluation by a staff psychiatrist revealed a similar clinical picture, although the psychiatrist was particularly struck by his current level of depression, which did not appear related to any drug abuse. His urine toxicology screen came out negative on the day of the interview. The psychiatrist agreed that the manic episode in the spring did not appear to be precipitated by drugs, although it may have been worsened by his subsequent drug abuse.

Drew’s father felt that many of Drew’s problems stemmed from his relationship with his girlfriend, Vida, a Latino young woman 3 years older than him. Vida had her own problems with drugs and alcohol and came from a small, fractured, inner-city family. Angie, Drew’s mother, was less convinced of Vida’s role, and was more prone to blaming his biological father for not being accessible enough. Drew agreed that his father had been inaccessible but did not feel this was related to his drug abuse or mood swings. When asked if he thought he had bipolar disorder, Drew shrugged and said “Probably. Doesn’t everybody?”

Aaron warily described Drew’s relationship with his mother as “close but full of tension.” Angie, who had recently closed her jewelry business and was pursuing a career in real estate, acknowledged that she’d felt less effective as a parent of late. She described Drew’s father as being “helpful but distant … he always keeps Drew at arm’s length.” Aaron agreed with this characterization: He had put his emphasis in recent years on his new marriage and his musical career.

In the separate interviews, Aaron described his own history of bipolar I disorder, marked by repeated and lengthy periods of depression and intermittent periods of mania or hypomania, usually sparked by new musical interests and often involved buying, repairing, and attempting to re-sell expensive instruments. Surprisingly, he was being treated only with Effexor, an antidepressant, without a mood stabilizer. He admitted having had problems with marijuana in the past, which he said was “part and parcel of being a musician,” although he denied consistent drug and alcohol abuse. Angie described herself as having periods of severe depression alternating with long-term periods of dysthymia.

Drew’s brother and sister seemed quite protective of him. Daniel said that his parents had “turned a blind eye” to Drew’s increasing drug abuse, and that he had talked to Drew about it on a number of occasions, to little effect. His sister Maddy, who had herself been arrested twice for marijuana possession, saw Drew as “trying hard to fool himself.”

Case Formulation

Drew received two diagnoses based in part on the KSADS-PL: bipolar I disorder (currently depressed) and polysubstance dependence disorder. The rationale for the bipolar diagnosis was that he had had both a full manic and a depressive episode in the absence of significant drug abuse; he also described periods of depression during his teen and preteen years, even when he wasn’t using hard drugs. Although it is possible that these episodes were accompanied by substance abuse to which he was not admitting, it is unlikely that they would have lasted as long as they did (in the most recent depressive episode, 6 months).

The initial case formulation took account of both biological and psychosocial factors. Drew had a strong genetic loading for mood disorder and substance abuse on both sides of the family. The family dynamics associated with Drew’s problems became an early focus of treatment. The clinicians observed that neither parent seemed clear on what BD was, although they had both read about it. Understandably, neither Drew nor his parents knew how depressive and manic episodes were different, how to recognize the escalation of new episodes, or what treatments were available to them. More salient was the enmeshed relationship between Drew and his mother and his stated desire to become closer to his father.

An initial hypothesis was that he was abusing drugs to distance himself from his mother. An alternative hypothesis, not mutually exclusive, was the possibility that his drug abuse and depression brought him in closer contact with his biological father who, in his mother’s words, “always swooped in to clean up a mess, and then swooped out again.” Drew’s history was rife with examples of problems that had caused his father to immediately come to his aid, after which Aaron disappeared again.

The clinicians offered the family a course of FFT (weekly and then biweekly), with the stated goal of helping Drew stay substance-free, maintain a stable mood, and finish high school. His psychiatrist recommended a course of lithium, with a target dosage of 1500 mg. The treatment regimen also required regular serum lithium assays and urinary toxicology screens to supplement those being conducted at home.

Course of Treatment

In the beginning of FFT, the two clinicians scheduled dyadic sessions separately with Drew and his mother and with Drew and his father. The initial goal was similar in both dyads: to educate the family about BD and substance abuse and develop a relapse prevention plan. Each dyad developed a list of Drew’s symptoms of depression (becoming more socially withdrawn, letting his appearance go, and looking “angry, lonely and tired”), mania (decreased need for sleep, an “electric feeling” throughout his body, more conflict with family members), and substance abuse (being “glib” about how he was spending his time, “talking ‘round and ‘round about things,” impulsive laughter, and behaving like a “caged animal” [mother’s words]). They were asked to identify which of these symptoms appeared in muted form before the onset of a full episode. For example, irritability appeared to presage the onset of his major depressive episode, and behaving like a caged animal antedated an increase in his marijuana use.

In the second step of relapse prevention planning, they developed a list of potential psychosocial triggers (conflicts with his brother Daniel, excessive contact with Vida, feeling “boxed in” by his mother, contact with certain friends with whom he had used drugs before). In the third step, they listed a series of prevention and health-promoting strategies when early warning signs and triggers were present (talking over his distressed feelings with his father, talking to Vida by phone, getting his lithium level tested, arranging an emergency therapy session, using relaxation techniques he had learned previously). Drew’s primary concern in these sessions was getting his parents to clarify when and under what conditions they would take him off lockdown and allow him out of the house. Neither parent was willing to offer such clarifications.

During the father-son sessions, Aaron was encouraged to share as much of his own history as he could with Drew, including how he had coped with his own mood disorder and intermittent substance abuse. Drew was instructed in active listening, a communication skill involving nonverbal attentiveness, asking clarifying questions, and paraphrasing.

Drew met his own diagnosis and the associated psychoeducational material with relative equanimity, saying it explained a lot. He felt validated by his father’s admission of his own disorder. He began to read online about BD and was intrigued by how neural circuits worked and how medications might alter preexisting neural imbalances. The clinicians explained the ways in which BD and substance abuse can negatively interact, and the importance of keeping a substance-free home. Indeed, Drew and his father bonded in educating each other about the disorder and discussing their individual problems with substance use, including how to avoid situations that would lead to use (e.g., parties, certain friends’ houses) and how to manage cravings (e.g., distraction to delay automatic responses). This increased understanding was accompanied by spending more time together, often doing low-key manual projects (e.g., building musical instruments, fixing sprinkler systems).

About five sessions into the FFT, Drew requested that the sessions be focused exclusively on his relationship with his father. Although his mother and he had developed a useful relapse prevention plan and had begun practicing the communication skills with one another, he felt that those sessions had taken on a stereotypical quality, in which his mother lectured him on what it would take to trust him again, and he gave her hollow assurances. In his words, the sessions were “like hearing the same joke over and over again.” The clinicians decided to honor his request, and his mother, who felt that Drew would be hurt by returning to the distant relationship he used to have with his father, agreed.

After a brief period of communication enhancement training for Drew and his father, the two began to spontaneously discuss their relationship, especially events from the past. The clinicians had reached a decision point: Should they continue with skill training in the step-by-step fashion outlined in the FFT protocol, or depart from the protocol to address historical events and the associated feelings? Their decision to do the latter reflected two conclusions. First, Drew seemed stable enough to tackle more personal conflicts and seemed to want to do so. Second, neither member of the dyad was likely to adopt the new communication skills until these issues were addressed.

Drew held considerable resentment toward Aaron for having “abandoned” [his words] the family when Drew was 6 years old. Aaron’s response was apologetic, but he also used the opportunity to clarify how complex his life had been early on. As he described it to Drew in a communication exercise, he had been “young, dumb, 25 years old with no job, three kids, and a failing marriage.” At the time, he believed that separating from the family and giving his ex-wife maximal control over the children was better for them, given how confused he felt about his own future. Drew and his father spent several sessions discussing events from Drew’s childhood and early adolescence. His sister and older brother attended several of these sessions and seemed equally motivated to discuss key events in the family’s history. Although anger and tears were often generated by these discussions, all agreed that the format allowed them to talk about things they had never discussed before.

About 3 months into treatment, Drew got an afternoon and weekend job at a dairy farm, which got him out of the house and gave him plenty of exercise. His mood gradually improved and he became euthymic by the Christmas holidays. Throughout the fall, his urinalyses were consistently negative, and his lithium levels were within the therapeutic range. He complained of lithium’s side effects, including increased urination, a hand tremor, and a blunting of his emotions. He did admit, however, that his mood felt more stable. He remained adherent to treatment, completed a home-schooling curriculum, and did quite well despite taking demanding math and science classes.

Increasingly, Drew wondered whether his prior drug abuse reflected “self-medication” of his depressive and manic symptoms. This possibility, he felt, was bolstered by his positive response to lithium, which had reduced his cravings significantly. The clinicians warned against this view, noting that substance use is likely to aggravate rather than control mood swings and often temporally precedes mood disorder recurrences (Strakowski, Delbello, Fleck, & Arndt, 2000). Nonetheless, as the treatment transitioned into the problem-solving phase at approximately 6 months, it was easy for everyone, including the clinicians, to believe that Drew’s drug abuse and mood swings were now a thing of the past.


After the holidays, and after a 2-week break from FFT, Drew’s mother decided to resume his urine assays because “his behavior was getting strange again.” She had lifted the lockdown approximately 3 months earlier, under the condition that he stay sober and continue to take his lithium. His first urine assay came back positive for marijuana and opiates. His psychiatrist checked his lithium level, which was 0.4 mmol/L, well below the therapeutic level. During a family session which included Drew, his parents, and both siblings, he admitted that he had been gradually cutting back on the lithium over the prior month and was now down to one 300 mg pill per day. His mood and behavior were mildly hypomanic, with increased irritability and decreased sleep over the prior 2 weeks; his thinking had taken on a grandiose quality. He had continued working and completed his fall school assignments, although with difficulty.

Drew explained that he did not fully believe he had BD, and that he wanted to test this belief by going off of his lithium. He wanted to know “what my mind is capable of without drugs in my system.” He was less clear on why he had begun smoking marijuana and using opiates again, except to say that they relaxed him, made him feel more social, and made him less combative with his family. His mother and father were quite upset; his mother wondered aloud if he needed residential substance abuse treatment. Daniel, who had held the view that nothing was wrong with Drew except stubbornness and immaturity, expressed anger that “he’s got all of us on a string.”

The clinicians conducted a behavioral analysis to determine what sequence of events had led to Drew’s drug use and lithium discontinuation. They asked him to recreate the details of his most recent use, his thoughts and feelings at the time, and the interpersonal events or stressors that had precipitated and followed the use. His marijuana use was largely related to wanting to reconnect with friends, believing that these friendships depended on marijuana use. He denied that his girlfriend Vida had been involved in his decision to use, and in fact she was quite angry with him about it.

Drew’s decision to stop the lithium had been brewing for some time, although it was hastened by a comment one month earlier from Daniel, who had said in the heat of an argument, “Your brain is wired all wrong. Maybe you should take more of that damn medication.” He rationalized his decision to stop lithium with, “My mind isn’t working anymore. Now everything I do feels like it’s because of the medications I take. I want to clear my mind of all the toxins I’ve put in it.”

It is quite common for young patients with BD to test the waters by discontinuing their medications. This is part of the natural developmental process in accepting a long-term illness. In the vignette below, the clinicians framed Drew’s doubts about the bipolar diagnosis as healthy and expectable for his age. They also pointed out the contradiction between wanting to have a substance-free mind to determine his capabilities and also wanting to smoke marijuana regularly.

Drew : I wouldn’t be surprised if it turned out I didn’t even have bipolar disorder.

Clinician : It’s understandable that you’d have that question. You’ve only had the one manic and one depressive episode. Most people who’ve just been diagnosed have that same question.

Drew : Yeah, and I wonder if I would’ve even had those if I hadn’t been using before. I guess I don’t really even know what I’m like. Maybe without the lithium I’d know.

Clinician : How would stopping your lithium tell you if you were bipolar or not?

Drew : Maybe it wouldn’t. But I would like to know what it’s like to have a clear head without any drugs in it.

Clinician : I can certainly understand wanting to know that. But then why would you want to smoke weed? That’s not a very good test of what your mind is like without drugs.

Drew : Good point.

Mother : In fact, it fogs your mind and your feelings even more than lithium.

Drew (defensive) : No it doesn’t. You don’t know what it’s like inside my head.

Clinician (redirecting) : Neither do I. But I think your questions about the bipolar diagnosis are healthy. It’s good to ask questions before you accept any diagnosis. But let me ask you again—how will going off your lithium and smoking weed instead tell you whether you’re bipolar or not?

The clinicians pointed out that Drew’s marijuana and oxycontin use could increase the likelihood of a manic relapse and, at minimum, would interfere with any therapeutic effects of the lithium in his system. They challenged the notion that marijuana is a medical treatment for BD. (In fact, the research literature suggests the opposite, namely, that increased marijuana use and the worsening of mania symptoms become intertwined with time; Strakowski et al., 2000.) Drew responded thoughtfully to these points and agreed to resume the urinalyses, but made no further commitments about taking lithium or remaining sober.

Postrelapse course of treatment

It is critical to resume FFT as soon as possible after a mood or substance abuse relapse, to communicate that the treatment’s objectives can still be achieved despite setbacks. Families are often frustrated at this point with the recurrent nature of mood and substance use disorders and are eager to get on with their own lives. Sometimes, family members will interpret a relapse as meaning that the treatment did not work and that the patient should instead be treated individually. Clinicians in FFT emphasize that relapses are an expected part of the course of mood and substance use disorders. In this case, they pointed out that Drew’s relapse was less severe than it might have been without medication and FFT sessions. Although we sometimes do recommend adjunctive individual sessions to supplement family sessions, Drew’s therapists were convinced of the centrality of family dynamics in the course of his substance use disorder. In addition, Drew preferred the family format.

After a session of problem solving, in which the problem of Drew’s lithium nonadherence was addressed, Drew reluctantly agreed to go back on his medication “for a few months.” His parents reintroduced drug testing, and the FFT sessions, which had been tapered to biweekly, resumed at a weekly pace. The clinicians decided to broaden the sessions to include his siblings on a more regular basis. One session with his mother and father also helped clarify rules to be kept consistent across the two households.

In the family sessions with Drew and his siblings, Aaron began to share more of his own history of BD. At the urging of his psychiatrist (who had been in contact with the FFT clinicians), he had begun taking divalproex sodium (Depakote), a mood stabilizer, and felt that his mood was more stable than it had been in years. He described several prior manic episodes during which he had lost jobs, and compared them to Drew’s most recent relapse. Drew recalled one prior episode where his father showed up at his mother’s house late on a school night, asking Drew to drive out to the desert with him to climb on some wind turbines.

In a session that did not involve his mother, and within the context of a structured communication exercise, Drew’s older brother and two sisters expressed anger at Aaron for his role in their fractured family life. Maddy felt that she had been invisible to her father for most of her life, and that her own drug abuse had been a “cry for help” from her father. In fact, her arrests had succeeded in bringing him back into her life for short periods of time, after which she experienced feelings of abandonment that led to more drug abuse. Daniel, never big on emotional expression, said simply that “It’s like supply and demand … Dad is a limited quantity.”

After listening to his children’s pain, Aaron admitted that closeness with his children frightened him and made him worry that he was not up to the task of being a father. He recalled his feelings of helplessness when his children were born, and he did not feel he could support them financially or emotionally. Drew and his siblings were encouraged to paraphrase their father’s statements and ask him to clarify any misunderstandings they had about his role in family events. In the end of this segment of treatment, the clinicians praised the young adults for their courage in confronting their feelings about Aaron, and Aaron for his openness to hearing his children’s painful recollections. They reframed Drew’s recent relapse as having served the purpose of bringing these elements of the family’s history into focus.

In sessions with his mother, the clinicians encouraged Drew to examine his own role in their enmeshed relationship. He fought her protectiveness with angry, somewhat childish attempts at independence (smoking marijuana in his room with the window open and leaving the house and not telling her where he was going). The clinicians made clear the bidirectional, reciprocal relationship between these actions, his mother’s response, Drew’s attempts to further establish his independence, and her attempts to control the situation further. They rehearsed with Drew several response styles that might alter these predictable interaction patterns. For example, Drew role-played with his mother how he might reassure her regarding his whereabouts and how she might respond if she wasn’t satisfied. A single session with Drew and Daniel helped develop a common perception of what Drew’s bipolar disorder was and wasn’t and how to understand the causes of his substance abuse.

Toward the end of treatment, the clinicians encouraged Drew to integrate Vida into his growing relationship with his father. Aaron and Vida had never trusted each other, yet both realized they would have to forge a connection to keep Drew from compartmentalizing his life with Vida. Eventually, Drew and Vida invited Aaron and his wife to have dinner with them. The evening was not cozy but proved to be an important first step in bridging important gaps within this family.

Outcome and Prognosis

At a follow-up session one year after intake, Drew had discontinued his lithium but also had stopped smoking marijuana and showed no evidence of other drugs on his toxicology screens. His mood had been stable for 3 months, and he had retained his job at the dairy farm. He finished high school and was taking classes at a local community college. His relationship with his father had become closer, and with his mother, brother and sister, more civil. In the final session, the clinicians reminded Drew and Aaron that their relationship needed ongoing maintenance, much like the musical instruments they had built together. They used the problem-solving format to plan at least one casual contact per week during the upcoming school year.

Clinical Practices and Summary

Although the basic structure of FFT—psychoeducation, communication training, and problem solving—was retained in Drew’s case, the clinicians went well beyond the manual to include elements of structural family therapy. In particular, they concentrated on the restructuring of alliances, such that Drew’s enmeshed, interdependent processes with his mother were interrupted, and his tenuous, disengaged relationship with his father was bolstered. To accomplish these goals, painful events from the past (notably, Drew’s and his siblings’ resentments of their father’s limited role in their upbringing) were discussed within the format provided by communication training. Structural interventions may be especially relevant when treating children in divorced families in which disruptions to healthy patterns of intergenerational alliance (e.g., connections between a biological father and his children) contribute to a child’s presenting problems.

Had Drew been in a controlled study, the clinicians would have had less flexibility in structuring the sessions and would have been discouraged from emphasizing active listening to the neglect of other communication skills such as positive requests for change or offering positive feedback. The problem solving tended to focus on larger problems such as how to maximize contact between Drew and his father, rather than daily hassles that might have been more easily solved (e.g., his mother’s complaints about his cleanliness). Departures from a treatment manual are more problematic in controlled research studies than in actual practice, especially if they occur more in one treatment than another. We do not know whether departures from the FFT manual have a negative effect on the prognosis of patients, or whether the complexity of certain patients or family situations requires such departures.

We have been continually impressed, however, that departing from the skill-training agenda to focus on the history of conflicts between people, interrupting enmeshed relationships in certain dyads and encouraging alliances in others, or strengthening boundaries between individuals or subsystems can often unlock a family’s resistances to learning new skills or incorporating illness-related coping strategies. The clinician must decide early on whether the patient is stable enough to address problematic relational processes or interactional patterns or whether addressing them should occur later, once the patient has recovered and is capable of handling high emotional intensity. The key, as it is in other evidence-based treatments, is to balance fidelity (to the treatment manual) with fit (to the particular circumstances and family).

The major objective of our current work with FFT is to determine whether the onset of fully syndromal mood or psychotic disorders can be prevented or delayed through family skill training. Individuals with early-onset mood or thought disorders, who are variously characterized as having major depressive disorder, bipolar spectrum disorder, schizotypal personality disorder, or atypical psychosis have a relatively high likelihood (35%–45%) of converting to fully syndromal bipolar or psychotic disorders within 2–4 years, especially if there is also a family history of these disorders (Birmaher et al., 2009; Cannon et al., 2008).

In ongoing trials, we have found that FFT can be successfully adapted to these populations. For example, in prodromal psychosis youth and young adults (aged 12–25 years), the psychoeducational materials focus on negative symptoms and social withdrawal, whereas the FFT materials for youth with prodromal BD (aged 9–17 years) focus on sleep/wake regulation or high risk behaviors such as drug use or hypersexuality. Psychosis has a later age at onset than bipolar disorder, and the populations are proportionately older. As a result, FFT with these populations tends to emphasize developmental challenges common in young adults (e.g., how to define a career path). Results from controlled early intervention trials with prodromal psychosis and prodromal bipolar youth are pending.

Successful community implementation of treatments like FFT will require identifying the key therapeutic ingredients (fidelity components) that account for the majority of variance in outcomes. The active components of FFT are likely to include the clinicians’ skill in teaching communication and problem-solving skills and in imparting psychoeducational material in a manner that is individualized to the patients’ needs. Further, clinicians must administer FFT with a psychotherapeutic attitude. This attitude is reflected in a good balance between session structure, pacing, and didactics with alliance building, empathy, and appropriate exploration of affective responses. As the case illustration hopefully illustrates, a psychotherapeutic attitude is especially critical when the illness is complex, highly comorbid, or treatment refractory.


Support for the preparation of this article was provided by National Institute of Mental Health Grant Nos. MH073871 and MH077856.

Selected References and Recommended Readings

  • Baker TB, McFall RM, Shoham V. Current status and future prospects of clinical psychology. Psychological Science in the Public Interest. 2008;9:67–103. [PMC free article] [PubMed]
  • Birmaher B, Axelson D, Goldstein B, Strober M, Gill MK, Hunt J, Keller MB. Four-year longitudinal course of children and adolescents with bipolar spectrum disorders: The Course and Outcome of Bipolar Youth (COBY) study. American Journal of Psychiatry. 2009;166(7):795–804. doi: 10.1176/appi.ajp.2009.08101569. [PMC free article] [PubMed] [Cross Ref]
  • Cannon TD, Cadenhead K, Cornblatt B, Woods S, Addington JE, Heinssen R. Prediction of psychosis in youth at high clinical risk: A multisite longitudinal study in North America. Archives of General Psychiatry. 2008;65:28–37. doi: 10.1001/archgenpsychiatry.2007.3. [PMC free article] [PubMed] [Cross Ref]
  • Goldstein BI, Strober MA, Birmaher B, Axelson DA, Esposito-Smythers C, Goldstein TR, Keller MB, et al. Substance use disorders among adolescents with bipolar spectrum disorders. Bipolar Disorders. 2008;10(4):469–478. doi: 10.1111/j.1399-5618.2008.00584.x. [PMC free article] [PubMed] [Cross Ref]
  • Miklowitz DJ. The role of the family in the course and treatment of bipolar disorder. Current Directions in Psychological Science. 2007;16(4):192–196. doi: 10.1111/j.1467-8721.2007.00502.x. [PMC free article] [PubMed] [Cross Ref]
  • Miklowitz DJ. A family-focused treatment approach. 2. New York, NY: Guilford Press; 2008. Bipolar disorder.
  • Miklowitz DJ, Axelson DA, Birmaher B, George EL, Taylor DO, Schneck CD, Brent DA. Family-focused treatment for adolescents with bipolar disorder: Results of a 2-year randomized trial. Archives of General Psychiatry. 2008;65(9):1053–1061. doi: 10.1001/archpsyc.65.9.1053. [PMC free article] [PubMed] [Cross Ref]
  • Miklowitz DJ, Otto MW, Frank E, Reilly-Harrington NA, Kogan JN, Sachs GS, Wisniewski SR. Intensive psychosocial intervention enhances functioning in patients with bipolar depression: Results from a 9-month randomized controlled trial. American Journal of Psychiatry. 2007;164(9):1–8. doi: 10.1176/appi.ajp.2007.07020311. [PMC free article] [PubMed] [Cross Ref]
  • Miklowitz DJ, Scott J. Psychosocial treatments for bipolar disorder: Cost-effectiveness, mediating mechanisms, and future directions. Bipolar Disorders. 2009;11:110–122. doi: 10.1111/j.1399-5618.2009.00715.x. [PubMed] [Cross Ref]
  • Minuchin S, Fishman CH. Family therapy techniques. Cambridge, MA: Harvard University Press; 1981.
  • Perlis RH, Miyahara S, Marangell LB, Wisniewski SR, Ostacher M, DelBello MP. STEP-BD Investigators. Long-term implications of early onset in bipolar disorder: Data from the first 1000 participants in the Systematic Treatment Enhancement Program for Bipolar Disorder (STEP-BD) Biological Psychiatry. 2004;55:875–881. doi: 10.1016/j.biopsych.2004.01.022. [PubMed] [Cross Ref]
  • Post RM, Leverich GS, Kupka RW, Keck PE, Jr, McElroy SL, Altshuler LL, Nolen WA. Early-onset bipolar disorder and treatment delay are risk factors for poor outcome in adulthood. Journal of Clinical Psychiatry. 2010;71(7):864–872. doi: 10.4088/JCP.08m04994yel. [PubMed] [Cross Ref]
  • Strakowski SM, DelBello MP, Fleck DE, Arndt S. The impact of substance abuse on the course of bipolar disorder. Biological Psychiatry. 2000;48:477–485. doi: 10.1016/S0006-3223(00)00900-8. [PubMed] [Cross Ref]
  • Van Meter AR, Moreira AL, Youngstrom EA. Meta-analysis of epidemiologic studies of pediatric bipolar disorder. Journal of Clinical Psychiatry. 2011;72(9):1250–1256. doi: 10.4088/JCP.10m06290. [PubMed] [Cross Ref]