Welcome to my blog, which speaks to parents, professionals who work with children, and policy makers. I aim to show how contemporary developmental science points us on a path to effective prevention, intervention, and treatment, with the aim of promoting healthy development and wellbeing of all children and families.

Saturday, April 18, 2015

Tsarnaev Trial Puts Spotlight on Developmental Trauma and Mental Illness

A colleague of mine, an active advocate for identification and treatment of postpartum mental illness, recently posed an interesting concern. With Susan Smith- who in 1995 infamously drowned her children- in the news again because she and Boston Marathon Bomber Dzhokhar Tsarnaev have the same lawyer-my colleague wondered if there was insufficient attention to Smith's postpartum psychosis.

As I reviewed the media coverage, both of the original trial and Judy Clarke, Tsarnaev's and Smith's shared lawyer, I discovered that she was right- there was little to no mention of postpartum psychosis. However,  Clarke's tactic clearly achieved her goal of portraying her client's humanity and vulnerability.

Smith was spared the death penalty because Clarke uncovered a story of significant early childhood trauma and abuse. Smith's father committed suicide when Smith was 6, and her stepfather sexually abused her in a secret relationship that continued in to her adulthood.   A newspaper article from 1996 quotes Clarke: 
This is not a case about evil. . . . This is a case about despair and sadness...Her choices were irrational and her decisions were tragic. She made a horrible, horrible decision to be at that lake that night. She made that decision with a confused mind and a heart without hope. . . . [But] confusion is not evil, and hopelessness is not malice. 
What does it mean to conceptualize “mental illness” as separate and distinct from trauma? One could say that, with the context offered of childhood maltreatment, a "confused mind and heart without hope" is a more evocative description than "postpartum psychosis."

Her question got me thinking about compelling research by psychiatrist Martin Teicher about the neurobiological underpinnings of adult mental illness in the wake of childhood maltreatment. 

Conduction a vast literature review, Teicher and colleagues differentiated two groups with psychiatric disorders, diagnosed according to the DSM system, with depression, anxiety, substance abuse and PTSD. One group had experienced maltreatment and another had not.

Maltreatment is broadly defined as being “characterized by sustained or repeated exposure to events that usually involve a betrayal of trust.” 

It includes not only physical and sexual abuse, but also emotional abuse, including exposure to domestic violence, humiliation and shaming, as well as emotional and physical neglect. The incidence of childhood maltreatment ranges from about 14% in one-year prevalence to 42% in retrospective reviews covering the full 18 years of childhood.

The way maltreatment is defined has great significance in the way we think about the connection between childhood experiences and adult mental illness. The word “trauma” itself may convey a kind of “not me” response, but when the term is defined in this way, we see that these experiences are, in fact, ubiquitous.

Teicher and colleagues found two subtypes, with significant behavioral and neurobiological differences, despite the fact that individuals in both groups carry the same diagnosis. Those with maltreatment history have earlier age of onset, more severe symptoms, greater suicide risk and poorer response to treatment. There are distinct differences in brain structure and function, stress response, as well as epigenetic changes in gene expression.

These findings offer a window in to how childhood maltreatment gets in to the body and brain. They have great significance in terms of prevention, intervention and treatment. Focusing efforts on supporting young children and their families is a natural conclusion. There are implications for treatment of adults as well. Teicher writes:
Recent recommendations for adults with maltreatment- related posttraumatic stress are to adopt a sequential approach that begins with safety, education, stabilization, skill building, and development of the therapeutic alliance before endeavoring to revisit or rework the trauma, as this may be destabilizing.
He expresses concern over the way these two distinct groups have not been differentiated.
Overall, we suspect that unknowingly mixing maltreated and nonmaltreated subtypes in treatment trials may have left us with an incomplete understanding of risks and benefits. Stratifying study subjects by maltreatment history may provide more definitive insights and delineate a clearer course of action for each subtype.
In other words, rather than treating psychiatric diagnosis according to the current DSM system, that looks at symptoms and seeks to eliminate those symptoms, it is critical, in determining appropriate treatment, that we explore the developmental and historical context of the symptoms.

In his conclusion Teicher writes:
We propose using the term ecophenotype to delineate these psychiatric conditions. We specifically recommend, as a first step, adding the specifier “with maltreatment history” or “with early life stress” to the disorders discussed here so that these populations can be studied separately or stratified within samples. This will lead to a richer understanding of differences in clinical presentation, genetic underpinnings, biological correlates, treatment response, and outcomes.

If Susan Smith did have postpartum psychosis, then certainly the specifier “with maltreatment history” would have been indicated. Teicher’s work has relevance to this story in two ways. By supporting new families who are struggling in the face of parental mental illness, we offer the best opportunity to prevent mental illness in the next generation. In addition, as evidenced by Clarke’s success in avoiding the death penalty for her client, by telling the story of the individual, rather than simply naming an illness, we evoke the full complexity of experience that a current DSM diagnosis, without this specifier, does not.

Thursday, April 2, 2015

What Are We Saying When We Diagnose Autism in Infancy?

Recent research suggests that while intervention is needed, we ought to be carefully considering this question.  A fascinating and important study by Jonathan Green in the January 2015 Lancet beautifully described in an article titled, The Social Network: How Everyday Interactions Shape Autism, shows that autism research is coming out from the shadows of the “refrigerator mother” theory. This theory, first identified by Leo Kanner in 1949 and popularized in subsequent decades by psychoanalyst Bruno Bettelheim, claimed that autism was due to lack of maternal warmth.
While this theory has been widely discredited, it led to a kind of backlash, where autism is understood and researched as a biological disorder that resides exclusively in the child. Many contemporary autism researchers pose the question, "How early can one determine if a child does or does not have autism?" analogous to the way one does or does not have diabetes or food allergies.
However, contemporary research at the interface of developmental psychology, neuroscience and genetics, showing how the brain changes in relationships, flies in the face of this formulation.
Given what we know about the plasticity of the brain, rather than framing the question as “Does he or does he not have autism?” a more appropriate question might be, “How to we, in the face of biological vulnerabilities, hold parents through uncertainty to give a child the best opportunity to grow in to what D.W.Winnicott termed his “true self.” (A question echoed by Stanley Greenspan’s DIR Floortime model)
As Green’s research beautifully demonstrates, holding uncertainty does not translate to “do nothing.” As the article about his study states, “An added benefit is that the treatment is easy for parents to do and doesn’t require a diagnosis.”
While this research is specifically about autism, it has relevance for any parent-infant pair that is struggling to connect. The essence of the intervention is a clinician who has a relationship with a parent, who offers space and time to listen to parent and child together. The following case from my behavioral pediatrics practice offers an example of an intervention similar to what Green offers in his research study.
Mary was convinced that her 3-month-old son, Liam, was autistic. She felt she couldn’t connect with him. Her oldest child, Jack, now 7, carried diagnoses of autism that had not been made until he was 4. Her middle child, Jane, had recently been diagnosed with anxiety. Mary was overwhelmed with fear that Liam would follow a similar path.
Mary told me that Liam was quiet from birth. He hardly even cried in the delivery room. Despite the doctor’s reassurances, Mary wondered from those first moments if there was something “wrong with him.” Then as the weeks went on not only was he quiet, but he seemed to her not to be connected. She would put her face close to his and try to engage him to look at her face and follow. But she was rarely successful. As the weeks went on her efforts intensified while her anxiety escalated.
With a full hour together, we sat on the floor and observed Liam together.
I noticed it right away. My initial attempts to engage him by talking to him and looking in to his face were met by a rather remote expression.  He appeared to be looking past me, perhaps at the lights on the ceiling, but it wasn’t clear. I saw Mary’s rising alarm. Resisting a similar reaction in myself, I said, “Let’s give it time.”
Liam lay on a blanket on the floor, at first continuing his seemingly random scanning of the room. I spoke quietly to him, noticing how he was sticking out his tongue. I imitated his movements and gradually he began to engage. Mary noticed that he seemed to be responding to my mirroring of his expression. Then we observed a remarkable transformation. In the quiet calm of this space, so dramatically different from the normal chaos of his everyday life, he seemed to come out of his shell. It started with a smile, at first seemingly random, but then clearly in response to my smile.
Mary continued to speak with him in a soft voice, but rather than putting her face up close to him, she spoke in a more natural way as part of our conversation. Liam became increasingly animated. Mary and I noticed, with rising joy and relief, that not only was he fixing and following on his mother’s face, but he was cooing in a responsive conversation with her. He kicked his legs and moved his arms in an expression of increasing delight.
Mary is not a “bad mother.” Liam's challenges are not her "fault." She is parent overwhelmed by the stress of caring for three young children and her understandable anxiety about the future of her infant. The space and time to listen gave us opportunity to notice that the intensity of her attempts to engage him were having the opposite effect. 
Relief flooded Mary, but alongside what threatened to be a paralyzing sense of guilt and fear. Had she caused him harm by missing his cues? But I pointed out how easy it had been for us to engage Liam. Clearly Mary had been doing something right. Research(link is external)has shown that even when parents miss these cues in 70% of interactions, as long as these “misses” are recognized and repaired, development moves forward in a healthy way.
When I saw them together a month later, Mary spoke joyfully of the fun the family was having with Liam, who had developed in to an engaged and happy baby. Now, taking a few minutes every day to have some quiet time with Liam, she fell deeper in love with him every day. She marveled at his complexity as a person even at the tender age of three months. This “disruption” led to new levels of love and intimacy between Mary and her son.
If an intervention similar to the one described in Green's study was available to all parent-baby pairs who are struggling, we might find that biological vulnerabilities, rather than leading to a diagnosis of autism, or some other disorder, can be transformed in to adaptive assets.