When I first arrived at the Philadelphia VA Medical Center to practice psychology in 2001, my patients – mainly male Vietnam veterans – were leery. I had a PhD and could be viewed as an authority figure, a likely trigger for distress for those who felt mistreated by the military or the Department of Veterans Affairs.

When I suggested an intervention or put forward an interpretation of their behavior, they would bristle. Regardless, I used to say to them, “I’m planting seeds. Please don’t rake them out. I’ll water them over time.”

Week after week, year after year, I’d use psychotherapy techniques that had been shown to be effective in randomized trials to reduce post-traumatic stress disorder (PTSD) symptoms, increase effective coping and help people live more meaningful and healthy lives. My patients got better, for sure. And it might have been because of those interventions – some aspect of one or their combination. But that’s not what my patients remembered as being most effective, and that’s not what I remember either.

What they remembered more – what they gave the most credit to – was the time we spent together, the bond forged over years of therapy. This is called the therapeutic relationship. And while evidence suggests that it is a critical part of psychotherapy, the impact of the relationship often isn’t studied in clinical trials for trauma survivors with PTSD.

The relationship can make a big difference for patients

It’s telling that the relationship between patients and therapists is sometimes described as the “therapeutic alliance.” The alliance, often thought of as the quality of the partnership and the mutual collaboration between therapist and patient, is essentially our positive emotional bond.

And research suggests that the therapeutic alliance and empathy (the therapist’s ability and willingness to see from and communicate their patient’s point of view) are particularly effective in contributing to positive change.

For instance, a 2013 meta-analysis on the efficacy of therapy for adult depression found that specific psychological techniques accounted for 17.1% of overall patient improvement, while relationship variables contributed 49.6%. That’s huge.

A series of meta-analyses commissioned by the second American Psychological Association’s Task Force on Evidence-Based Therapy Relationships indicates that the effects of the therapy relationship on psychotherapy outcome are substantial – can make therapy more or less effective.

Some prominent trauma clinicians have argued that the therapeutic relationship is as important (or even more important) to helping trauma survivors improve as the therapy technique used. But most clinical researchers developing research-based interventions for trauma survivors with PTSD do not measure it as a variable in their psychotherapy trials.

Why doesn’t the relationship factor into clinical research?

When psychologists conduct research to find out how effective one type of psychotherapy is over another, we bring people into our academic laboratories and essentially flip a coin.

One individual goes into one type of psychotherapy and another goes into another type of psychotherapy and we compare the two to see if one type improved symptoms better than another. We sometimes include a measure of quality of life, such as increased time spent with loved ones. If the people in a particular therapy condition get better, we label that therapy as “efficacious.”

As a behavioral scientist, I understand the power of randomization and believe in the gold standard of the controlled trial. But is it accurate? Is it sufficient? Does it capture the art of psychotherapy?

There are a few reasons that the the therapeutic relationship isn’t evaluated in clinical trials for trauma survivors with PTSD. It may be because measuring the process of therapy and the changes that take place in the therapeutic relationship is harder than measuring the techniques. Or it could be that government funding priorities do not emphasize these “soft” variables and thus there are few dollars to support this type of research.

In addition, it seems that graduate training in psychology nowadays places much more emphasis on learning standardized protocols than on how to develop and maintain healthy relationships with our patients. And in the treatment of survivors of trauma, particularly trauma arising from betrayals from other human beings, the relational dynamics can be complex and perhaps difficult to teach.

The power of the therapeutic relationship

I spent four years working at the Philadelphia VA Medical Center, and when I left, my patients presented me with gifts. One was a plaque with all their names engraved on it and a message that said, “In appreciation for being our fearless leader and guiding us through life and leaving us a path to follow. We owe you our lives.” Before I could say a word, one veteran handed me a dozen red roses and said, “These are the seeds that you planted. Look how they’ve grown.”

I wept. I thanked them for their effort, honesty, vulnerability, courage and, of course, their service. And then I asked them what had helped them the most in our work together.

One by one they recounted stories – playful or touching moments. They didn’t attribute significant change to any of my clinical box of effective techniques, the experiential exercises, the homework assignments. Not one. Not once.

Some may say that my veterans didn’t accurately recall or attribute effectiveness to the techniques. Or that the techniques were what really mattered and that my patients' relationship with me and our relationship with one another was only the vehicle to that delivery. But there is plenty of research that shows just how important that bond between patient and therapist is.

While psychotherapy is not solely about the relationship, it is those moments when we try to connect with our patients that make us bloom.

Joan Cook receives funding from the National Institute of Mental Health.