A Study of Weight Loss as a Model for Clinical Research based on Bowen Theory
Posted by Laura Havstad
This is the summary my research and the goals that emerged, which I presented when I accepted the Polly Caskie Research Award at the last Spring Conference. It describes the basis of a paper I’m working on for publication in Family Systems, which I hope to submit very soon. Any questions or puzzles or competing ideas ? I’m interested in hearing.
The framework for research described is now being evaluated for it’s value in structuring a knowledge base of family systems data for the Bowen theory network and those who might wish to organize their own data, and access that of others, to research the process of differentiation of self and the family emotional system and how they impact various outcomes of interest. What ideas and questions would you put to the test of research if you could and if you had a reliable and valid sample of emotional process that was set up for systematic observation?
A Study of Weight Loss as a Model for Clinical Research based on Bowen Theory
I began the research on weight loss as a graduate student in 1978 a couple of years into the postgraduate program at the Family Center. Family evaluation interviews were used to gather family system data. As theory predicts, shifts in the family system preceded weight loss for the 28 overweight and obese women in the study. Their functioning shifted up when their chronic anxiety went down as a result of the shifts in their family system. Sometimes their functioning shifted up to meet a challenge when serious threats emerged in their relationships and anxiety was up. In total, the take away was shifts in the family system preceded weight loss for them all.
30 years later in 2008, I came back to these findings. I wondered about the potential for scaling up the research method for mainstream clinical research. Bowen theory predicts that changes in anxiety and functioning levels that emerge out of shifts in the family system will precede changes in most symptoms. It is a dream that this knowledge will one day be basic knowledge in clinical science. But mainstream clinical science needs proof based on the conventions of scientific method.
I set out to see if the method that produced such clear results in the weight loss study could be defined clearly and parsimoniously enough that mainstream clinical researchers could use it.
For mainstream research the first question that needed tackling was the reliability of observations. Could, to begin with, experienced observers of family systems use the method and independently agree about observations of shifts in the family?
The results of a small study said, yes. An intriguing question is could researchers new to Bowen theory be trained by the research method to see and document the family system and it’s impact on clinical outcomes?
An overarching framework for clinical research has emerged from these studies that rests on three broad variables within which all the other variables of the theory can fit: The first being the shifting relationship patterns in the emotional unit, associated with the second variable of shifts in the functioning of family members in level of anxiety and level of functioning, associated with the clinical outcome variable of shifts in symptoms. The method tracks how these three variables follow one after the other over time, and how shifts in the family system produce shifts in the functioning and anxiety levels of its members, which, in turn affect the course of their symptoms.
With this method it is possible to assess the impact of the family system on individual family members and to assess the impact of a shift in the functioning of a family member on the family system.
.The overall framework is simple which is important for the potential of theory to enter mainstream research. But the framework also has the flexibility to incorporate the dynamic interplay of all 8 or 9 variables of Bowen theory as a unified family system variable. The method is qualitative but applied to enough cases could produce an accumulating empirical basis for assigning quantitative values and algorithms to the system of theoretical variables. There is much work to do to realize this potential, but I believe I see a way
Laura,
It’s good to see the steps forward that you have taken with this project and the way your are conceptualizing the organization of the variables.
The interplay between family emotional patterns, functioning shifts, and symptom development would seem to be too complex a system to measure, but you are finding a way. The possibilities this holds for clinical practice are exciting to think about.
Stephanie
Yes, there is a way… Thanks for your encouragement, Stephanie.
Laura,
The results you state are intriguing, and I wonder how consistent these results are if applied to different symptoms in different families. You report that an individual is able to pull up in functioning during a time of significant shifts in family functioning. I think it makes sense theoretically, and that could be elaborated in each example. Is it also the case that an individual’s functioning can go down during periods of shifts? I also wonder if there is a way to take advantage personally of this information! I’m struck by the efforts we make to improve our functioning, and how our basic level of DoS seems to come out, somewhat regardless of our intentions to do better. I say “somewhat” because Bowen’s theory does offer a guide toward more responsible rather than more irresponsible actions–as well, of course, as a guide toward the long-term effort to raise DoS.
It sounds like a pretty solid longitudinal method. I might include some additional causal variables to flesh out ‘shift in family functioning’ as a cause. These days i’m looking for groups of causes working together, more than just single causes.
Jim Edd, Thanks for posting on this older post. Shift as I see it, encompasses all the multiple variables in Bowen theory as they change in unison and can be specified in more or less detail with all the sub variables in individual and family unit functioning that theory describes and more. But in total one can consider shift a measure of the family system as a single variable in relation to, for instance, physiological and environmental and societal variables as well. What I found fascinated in my original research is how profoundly the family system variable operated in the phenomenon of significant weight loss. I have a paper submitted to Family Systems on this and we’ll see where it goes.