Beyond the One-Way Mirror

A New Approach to Reviving Public Sector Psychotherapy

Scott Sells with Cynthia Franklin
Magazine Issue
July/August 2008
Beyond the One-Way Mirror

It was my first meeting as a visiting supervisor at a local community mental health center in the Southeast. To get some background, I asked the clinical staff what had been the biggest change they’d seen since they’d started at the center. Without uttering a word, John, the senior member of the group, got up and led me down a hallway with dirty floors and off-white paint peeling from the walls. He opened the door of a large room filled with boxes of old files that had a mop and a bucket standing in one corner.

“Why are you showing me a storage closet?” I asked.

“This wasn’t always a storage closet,” he said dejectedly. “This used to be our one-way mirror observation room.”

John then explained that from the 1970s through the mid-1990s, one-way-mirror supervision had been the hub of clinical discussion at the center. The treatment team had met once a week to observe and learn from one another. The phone on the wall—next to the mop and bucket—had been used regularly to call in interventions when therapists were stuck with a case. Clinicians were encouraged to bring in not just parents and siblings, but grandparents, aunts, uncles, neighbors, and friends. Medication was used, but only as a last resort, and the psychiatrist had been a working member of the treatment team.

Then, about 15 years ago, everything began to change. Live supervision was the first thing to go. “Our administrators told us to stop because we couldn’t bill for it under managed care. So our one-way mirror observation rooms were converted to storage closets,” John continued. “The next thing to go was intense group supervision. Our clinical discussions about stuck cases, usually an hour or more, were whittled down to 30-minute case-note reviews. These days, all our work is done unsupervised, behind closed doors.”

Family therapy itself was finally abandoned because new Medicaid regulations made it more practical to see individuals and conduct 15-minute medication checks. It had become easier to just take an individual child or teenager out of class and bill the same fee as for an after-hours family session.

“Money and billable hours are now driving our treatment methods,” John added. “Individual psychotherapy with meds as the quick fix is now the standard. Psychotherapy is dead at this agency. We feel we’re just cogs in a wheel. Our burnout rates are skyrocketing, and the average therapist only stays around until finding something better.”

The other therapists nodded agreement. Even though John was the only one who remembered the old days, everyone agreed that better training and more supervision were needed, but no one felt there was much likelihood of that happening.

Things Sure Have Changed

The one-way mirror observation room that’s now become a storage closet is a metaphor not only for the transformation of the profession of therapists like John in the public sector, but for many of us in private practice as well. Psychotherapy as we all knew it in the 1970s, 1980s, and 1990s—with its emphasis on quality care, innovative approaches, and patient advocacy—has gone the way of the one-way mirror. Managed care, the relentless push for briefer treatment, and exponential growth in the use of medication have placed the field in crisis.

The private practitioner who says, “Who cares what happens in the public sector? It doesn’t affect me” is like the noninvestor who asks why she should care whether the stock market crashes. The psychotherapy field isn’t neatly compartmentalized into public and private sectors. When one part is in decline, it’s only a matter of time before the ripple effects are felt in the other.

We should all be concerned that almost all the leading training centers of family therapy’s golden age from 1975 to 1990—the Family Therapy Institute of Washington, D.C.; the Philadelphia Child Guidance Center; the Mental Research Institute—have either closed or limp along as shadows of their former selves. Another cause for alarm is that, according to an American Psychiatric Association survey, mental health patients consider auto repair shops more trustworthy than their managed care mental health providers.

At the same time, recent developments within the field hold great promise for public sector therapy. There’s a growing body of empirical evidence that didn’t exist in the glory days of the community mental health movement—evidence demonstrating that a wide variety of time-limited approaches can be clinically effective. These models, such as cognitive-behavioral therapy, multisystemic therapy, and structural family therapy, are now called evidence-based practices. Unfortunately, because of depleted budgets, excessive caseloads, inadequate training and supervision, and bureaucratic inertia, therapists in the public sector typically don’t have the time, energy, or motivation to learn how to use these evidence-based models. In addition, a lot of the models lack “transportability”—meaning that it’s difficult or impossible to integrate them successfully into public or community mental health.

The CEO of a major community mental health center in the Southeast once bluntly explained to me why public sector psychotherapy isn’t feeling any particular pressure to keep up with the times: “Let’s face it: we get a constant and steady flow of referrals, whether we’re effective or not. We’re a monopoly, the only place in our local area that accepts Medicaid.” His voice turned sarcastic. “In fact, our center is so poorly run that if this were a private business, we’d have gone bankrupt long ago. But why should I purchase and use an evidence-based model to make us more effective? No one cares. We have a culture of clinical minimalism. Our therapists do the minimum because that’s all they have to do.”

Is There a Future for Public Sector Psychotherapy?

In the face of such obstacles, how can we transport effective models of therapeutic practice to frontline clinicians? How can we convince administrators and key government agencies to support and fund these efforts? What can we do to make sure that public agencies don’t continue to devolve into institutions of dubious therapeutic quality?

In the evidence-based world of today’s psychotherapy, the first step is demonstrating that what you do actually works. About 10 years ago, I began to develop a model from the ground up for treating teenagers diagnosed with either oppositional defiant or conduct disorder.

In developing the model, I used process research to identify the essential techniques and strategies in structural and strategic family therapy that led to positive changes in out-of-control adolescents aged 10 to 18. This type of research is so labor intensive that it isn’t often conducted in our field. I had to watch hundreds of hours of videotapes to locate and analyze precise moments of change.

I spent an entire summer in the basement of Charles Fishman, one of the founders of structural family therapy, watching tape after tape of his and Salvador Minuchin’s work with angry, aggressive adolescents and children. The next year, I approached Jay Haley, the founder of strategic family therapy, and his colleague Neil Schiff. They let me spend hundreds more hours with their tapes. (To better understand how the model was developed, please go to and click on the link “research effectiveness.”)

Next, I spent two years field-testing the techniques and strategies I’d identified with 83 conduct-disordered adolescents aged 12 to 17 and their families. My goal was to provide a practical approach, offering crystal-clear instructions for therapists, especially about what to do if a specific intervention failed. Having been a troubled teen myself, I was driven to help future teenagers avoid the pitfalls my family and I had stumbled into.

The result of this work was an evidence-based model called Parenting with Love and Limits (PLL), which integrates the most effective structural and strategic therapy methods for treating troubled teenagers in a hybrid format that combines group and family therapy. Its goal is to restore nurturance between parent and child and reestablish a mutually respectful family hierarchy. The approach is “manualized” in a step-by-step group therapy leader’s guide, videos, and a 7-step family therapy “coaching” survival kit. (You can visit the website for more information.)

Developing an empirically supported model and breaking it down into easily understandable techniques may be essential first steps in getting a method into public sector therapy. But finding a way to convince public agencies to adopt and implement such an approach is a challenge that’s befuddled many developers of promising evidence-based treatment models.

Structural Family Therapy as the Framework for Change

I’d like to say I had immediate success transporting PLL to the public sector, but that wouldn’t even be close. In fact, I had so little luck during the next six years that I nearly gave up the struggle. In the early days, agencies would use parts of the PLL model for a while and then go back to their old way of doing things. I kept asking myself what I was doing wrong. I remember calling my wife in tears from a plane that was taxiing down the runway after leaving yet another public agency that wasn’t using my model. To make matters worse, I’d left my carry-on suitcase in the airport restaurant. At that moment, I almost gave up. But luckily I remembered Thomas Edison’s words on the subject of failure: “I have not failed 5,000 times in trying to invent the light bulb. I have successfully discovered 5,000 ways that do not work and I do not need to try again.”

I remember the exact day and time when my own personal light bulb went on. I’d contracted with a state and its local agencies to train their therapists, but as I walked into the training session, I saw 20 angry clinicians staring me down. I was clueless as to why and
didn’t know what to do about it, so I just began the training. After half an hour, one of the therapists raised his hand and asked, “Why are we here? Our bosses told us that we had to come to this training, but we still don’t know why.” As far as these clinicians were concerned, they’d been ordered to listen to some stranger exhort them to add one more difficult job—a new therapy method—to their already packed days, without discussing it with them beforehand. Needless to say, they weren’t ready to “buy into” the new program.

I later realized that despite all my attention to how to heal tough families, I’d failed to apply the principles of family therapy to tough organizations. As Minuchin himself might have said, I’d tried to restructure the family (the local agency) without first joining and being invited into their world. An aha moment occurred for me right then and there: the effective application of the methods of good structural family therapy were vital to achieving transportability.

The Need to Join

In Family Therapy Techniques, Minuchin and Fishman write that joining occurs when the family and the therapist enter therapy with the same goals, but the family must first invite the therapist to enter their system if they’re going to work together to make systemic changes.

In the same way, I realized, the public agency needed to invite me into their system if my model was to be successfully transported. To achieve this, I’d need to persuade the stakeholders at various levels of the system—therapists who’d be implementing the model, clinical directors who’d be supervising the therapists, and administrators who’d be funding the program—that my goals were beneficial to them. I soon got my chance to put these ideas to work within a complex mental health community in a large Southern state.

Responding to an invitation from the state Department of Juvenile Justice, I started off identifying the four levels of “clienthood” involved in the consultation: the state funding agency that was behind the invitation for using PLL, the local public mental health agency service provider, the community of other agencies that would be supplying the referrals, and the frontline therapists who were to be the primary focus of the training.

In the past, whenever I’d secured funding from the local or state agency before the training, I’d met the therapists and community stakeholders only on the first day of training. This time, even though the state juvenile justice system had given a green light to the project, I now realized that the community and agency had to invite me into their own system, and that they had to take responsibility for the program with me if the implementation of PLL was to “take.” So I was careful to tell the mental health agency and the therapists upfront when I went to meet them before the scheduled training that they could still choose to reject the model.

I knew that to get a multilevel buy-in from all the stakeholders, I needed to create a shared vision from the beginning of the relationship. I decided that the best way to do this was to set up a live demonstration of PLL with an actual family. I wanted to get the administrators and therapists in the audience excited about doing treatment differently—to show them what it meant to go from case management and individual therapy to a model that combined group and family therapy with parents and their children.

The selected family included an out-of-control, violence-prone 12-year-old named Jake. On the phone, the mother said, “I have to be the peacemaker. My husband and Jake can’t stand each other, and they fight all the time. I feel that I’m on the verge of a nervous breakdown.” I discovered that Jake listened to his grandmother more than to anybody else, but that she was extremely old and frail. Still, when I asked her to, she agreed to come to the demonstration. You could have heard a pin drop as the family arrived and the audience watched the grandmother struggle down the aisle on her walker.

I knew from speaking to the father on the phone that he was blind to the central role his sternness played in making his son Jake misbehave. So, after I said that there were unspoken and unhealed wounds in the family, I chose an emotionally compelling movie clip to capture the family’s attention—or as Fishman and Minuchin put it, I raised the “intensity to go above the family’s threshold of deafness.”

The scene was from The Horse Whisperer, in which Robert Redford ties a traumatized and brutalized horse’s legs together to stop it from thrashing wildly about. When the horse goes down, Redford asks the horse’s owner to tenderly massage the animal. The horse moans softly as it responds to this gentle care, and its emotional wounds begin to heal.

Tears flowed from the eyes of both father and son as they instantly saw the parallels between the traumatized horse and the trauma in their relationship. The horse represented Jake, who needed structure but also nurturance. The horse’s owner represented the dad, who was unsure of what to do or how to be nurturing. However, the real breakthrough in this session occurred when the grandmother gently touched the father and told him that Jake needed his dad, and that he could be as good a father as he’d been a son.

During the course of this demonstration, I illustrated other treatment strategies of the PLL model, such as the use of enactments to help the son and father speak to each other and to give the mother the opportunity to tell the family how tired she was of being put in the middle. However, it was clearly the movie clip that made the most powerful emotional impact. After the session, the father turned to the audience and said that, with all the therapies he’d tried over the years, this was the first time he’d felt hope for his family’s future.

Once the family had left the room, the atmosphere became electric. The therapists in attendance were excited about what they’d just witnessed, feeling the same hope that the father had expressed. Even the skeptical probation officers began to talk about families on their caseload whom they’d like to refer to the program immediately.

The next day, I sat down with the local director of juvenile justice and the executive director of the service provider to develop a strategic action plan. We figured out how PLL could be customized to fit their current gaps in service.

Before the live demonstration, I’d secured the buy-in from the state for the PLL model on the basis of increased delivery of treatment and cost savings. The state employed other evidence-based programs for adolescent conduct disorders, but they were delivered using a one-on-one individual family therapy format. This meant that one therapist could see only one family at a time. The state representative bought in when he realized that a PLL therapist working in a group format could treat more than twice the number of families. In addition, the cost per youth using PLL was 50 to 80 percent less than using the other models.

The five-day training that followed was a breeze. The therapists were excited and pumped. They’d even read the books and group therapy leader’s guide before I stepped off the plane!

Restructuring, Connecting, and Supervising

Despite this auspicious beginning, it was still crucial to have the PLL model actually take root in the agency. In the past, after the initial, five-day, onsite training wrapped up, another PLL supervisor or I would maintain an arms-length relationship with an agency, consisting mostly of two-hour phone supervision about stuck cases once or twice a month. This time, I decided to experiment with regular onsite visits within the first three to four months after the training. Like the live presentation, these visits proved crucial in the joining and restructuring process. For example, when the agency was having trouble jump-starting referrals, I rolled up my sleeves and set up a meeting between the agency and local referral sources.

At the meeting, I suggested that a probation officer and a PLL therapist meet with client families together for the first intake appointment, thus diminishing the possibility that the family would play them off against each other. Through role-plays, I taught them how to do motivational interviewing. Both groups loved the idea of working together and started implementing the strategy immediately. Soon the splitting engendered by the families stopped, the relationship between the probation system and the agency was strengthened, and the referrals from the probation system nearly doubled.

These onsite visits have continued, changing the relationship from a formal consultation to a dynamic partnership with a high degree of mutual trust and respect. During the onsite visits with the PLL staff, we’ve covered everything from intensive case discussions to producing a mission statement, but what’s taken the partnership to a higher level is our new approach to ongoing supervision.

In the old days, the PLL staff and I would do some form of typical case-management supervision: a therapist would present his or her best recollection of a stuck PLL session from case notes and we’d recommend a strategic family therapy intervention to get things moving. Neither we nor the therapist had any way of knowing whether the therapist was making tactical mistakes in sessions or using the model correctly. The result was low adherence to the PLL model and frustration among the therapists because their tougher clients weren’t improving.

From my one-way-mirror supervision days, I knew that this supervision format was the answer, but didn’t know how it could be adapted to an agency without one-way mirrors. To address this challenge, I investigated a relatively new method of supervision, known as interpersonal process recall (IPR), whereby the therapist videotapes a session of a stuck case on a DVD camcorder and the clinician’s performance is rated using a five-point scale. I adapted the IPR method to focus on seven specific skills essential to the implementation of both the group and family therapy aspects of PLL. (Please go to to view the IPR method.)

After the therapist videotapes a session, he or she keeps the original tape and sends a copy to the PLL home office. During supervision, we watch the same DVD session on our respective laptops and review the same IPR form together. It’s the closest thing to one-way-mirror supervision we’ve been able to devise.

Before the introduction of PLL, the therapists had been conducting all their sessions behind closed doors. Therefore, they were understandably anxious about this new form of supervision. However, their anxiety quickly dissipated when the onsite clinical director, Joan, took the lead and we used videotaped IPR one-way-mirror supervision on 16-year-old Max, who was experiencing bouts of emotional and physical abuse from a father who refused to come into therapy. To give the mother the support she needed, we had her best friend, who couldn’t attend the session, on speakerphone. Bringing in the client’s “village” to move a stuck system was a brand new concept for the therapists, who’d worked primarily with individuals.

The therapists were riveted as they watched the videotape showing how Joan and I, acting as a consultant during one of my onsite visits, worked together to get the mother unstuck. As they watched the video, they used the IPR form to rank us on seven separate skill sets. They loved the idea of critiquing us and being active participates in the supervision process.

A critical moment in the session came when we had to use the structural technique known as “unbalancing” to take sides with the best friend against the mother, to try to get her unstuck. Agitation and tension were clearly visible in her as we explained that her unwillingness to bring her husband into therapy would cause a relapse in Max’s symptoms of disrespect and acts of aggression toward her. We even brought up the issue of race, suggesting that the mother might not trust us enough to bring the father in because we were both white and the family was African American.

The unbalancing maneuver seemed to work, as the mother’s friend became our cotherapist, gently but firmly convincing the mother how important it was for the father to be present. The friend brought out the fact that race was an issue, because the family was from an African village that deeply distrusted outsiders. To address this issue, the friend agreed to ask her husband, who was a friend of the father, to speak with him about joining the therapy.

The Power of Relationship

We still have a long way to go in reviving public sector psychotherapy, but the lessons learned from this community represent a hopeful beginning. In the end, successfully transporting an evidence-based model wasn’t as much about our statistical outcomes as about building relationships. Doug Sprenkle, a researcher at Purdue University, explained best why the interface between research and real-life therapy can be so difficult: “Researchers sometimes disdain clinicians, fail to listen to their wisdom, and don’t typically work very hard at making their work clinically and easily accessible.”

This disconnection trap is easy for us researchers and therapists to fall into when we become enamored with a promising new approach, particularly when we’ve helped develop it and know from solid research how well it can work. There’s a natural tendency to think that once we show other clinicians this wonderful new method, there’ll be general acclaim and everyone will fall in step behind us. Can’t they see how much better this approach is than doing things the old way?!

However, the validity and effectiveness of the approach turned out not to be enough. As we learned, it was necessary to prepare the way for its implementation by taking the time to join with the community on multiple levels, and then continue to join by going the extra mile with onsite visits and videotaped supervision to maintain high-quality delivery of treatment. Once the critical work of joining had been done, the implementation process seemed to take on a life of its own. As the quality of work improved, so did morale. As morale improved, so did the agency’s effectiveness. As effectiveness improved, so did the economics.

In the bible story, Jesus observes that one shouldn’t put new wine into old bottles, which may break and let the wine run out. But that’s what I’d tried in my initial work with PLL. I’d forgotten that agencies, like families, are systems, and that they interlock with multiple other community systems. So, if we didn’t engage all the players in all the systems when we presented the new approach, even the most brilliantly conceived project in the world would collapse like a house of cards.

It’s an old lesson that bears repeating: we have to pay attention to relationships more than numbers, and to direct supervision more than case notes—which is what our family therapy founders never stopped telling us.


Scott Sells, Ph.D., CEO of Parenting with Love and Limits and a former professor at Savannah State University, serves as a ­consultant to the Department of Juvenile Justice. He’s the author of Treating the Tough Adolescent and Parenting Your Out-of-Control Teenager. Cynthia Franklin, Ph.D., is a Stiernberg Professor at the University of Texas at Austin, School of Social Work. She’s the author of more than 100 publications.