Last week I was asked by the Australian and New Zealand Academy of eating Disorders and by the National Eating Disorders Collaboration lead a discussion on what are the core competencies for someone working in the field of eating disoders. As part of that I gave a presentation entitiled "The Model or The Clinician - Who comes first?" I have attached a link to a PDF of the slides here.
I thought I might just provide some context for some of the slides so it makes some sense.
We start with a brief review of the evidence base in the eating disorders (along with one of my favourite quotes from Christopher Hitchens "What can be asserted without evidence can also be dismissed without evidence".
Slide 3 reminds us we do have Clinical Practice Guidelines to direct us, as well as a more recent Cochrane review and also (slide 4) great review by Bulik (up to 2005) and then Hay (2005-20013). Both published in International Journal of Eating Disorders.
Slide 5 suggests there is the strongest evidence for FBT in the treatment of adolescents with anorexia nervosa, but Slide 6 reminds us that that data (for full recovery - which is the standard to which I hold myself) indicates full remission for 50% of patients. (i.e. it's good but 50% ain't great).
Slide 7 & 8 makes similar point for CBT for BN. Good data suggesting good outcome for most, but again about 50% of patients remain symptomatic at end of treatment. There are also other treatments that are helpful for BN and BED, such as Interpersonal Therapy or DBT. The weight of evidence suggests however a first line treatment for BN should be CBT.
Slide 9 covers adult AN, where our results are poor. Most treatments get about 20% of patients to full remission, and get a substantially higher number to partial remission. 10,11, 12 summarise the latest study from Germany - the ANTOP study. This study again found that the type of treatment (a manualised psychotherapy or CBTE) didn't really matter in terms of outcome. Our treatment manuals aren't guiding us to what we should specifically do with adult patients in AN. I would point out however that the majority of the manuals focus on improving nutrition.
So we have some data to guide us on what may be helpful in the treatment of eating disorders. However, the question is how many practitioners are using the evidence base which suggests some empirically supported treatments exist.
Slide 15 was chosen not because the Tobin study is the best one indicating the Research Practice Gap, but because it has the coolest name (I know what you did last summer ... and it wasn't CBT) and it allowed me to put up a poster of SMG. There is a concerning "What Manual" with 21% of clinicians asked do they use a manual in the treatment of BN. I am more interested in the "flexible" user group, as this is where manual and clinician come together (maybe).
16 is on who uses manuals. Well when I here CBT, research focused and young I tend to think of my Doctor of Clinical Psychology and Master of Clinical Psychology students (some of who are pictured in the slide). When the first graduate the don't have the clinical experience, so should be sticking close to the manualised treatment where ever possible.
17 is some of the reasons people don't use manuals. I am most worried about those who don't know the manual. Slide 18 is an interesting study that indicated that even of those for whom CBT was their treatment of choice, few had formally being trained in the manual. ANZAED and the various state based training organisations must roll out training in evidence based treatment. I know Victoria have done this with FBT, as has NZ, but NSW is far behind.
RCT's dont exclude most patients - its an excuse we use when our results aren't as good as RCT. I have never met an eating disordered clinician who doesn't think they have the toughest patients that have all been excluded from RCT. The evidence generally suggests that treatments do disseminate if you use them.
Maybe our treatment models are "ignorant". I am thinking here of adult anorexia nervosa. Maybe drive for thinness or drive for control just doesn't cut it. I look forward to looking at the data coming out of neuropsychiatry labs like Walter Kaye's and the emerging research on Cognitive Remediation Training that may increase the outcomes for CBT.
The most common reason proffered for not using the Empirically Supported Treatment is that therapy is an art that should not be restrained by a cookbook. Surprisingly I agree with this to some extent. Mainly in the restraints that an RCT puts on the length of treatment. Maybe 40 sessions is not enough for adult anorexia (see the ANTOP studies weight graph that shows people keep getting better after treatment). Sometimes patients may need 30 sessions of FBT rather than stopping at 20 as the manual and RCT protocols suggest.
However, there are ideas in the manuals that help our patients. Our therapies need to reflect this.
As clinicians we need to remember that Evidence Based Practice is a three legged stool. Yes, one leg is Empirically Supported Treatments (our manuals). The other legs are clinical experience and also patients values and preferences.
Each of our camps tend to grab onto on leg of the stool and pull. The stool then looks a bit like a "hitting stick" (Slide 20) to beat people over the head with. Or as Phillipa Hay once wrote in response to our article "tyrannise" people with THE EVIDENCE.
However, as clinicians we need to however be aware of our own humanity and vulnerability to thinking errors - most importantly the confirmation bias (slide 21). This is when we do what we think works and collect evidence that it works and discount evidence it doesn't. This bias can lead us to do things that don't work continually, because we assume they do. It is this that, in my opinion, using manuals grounded in science helps us with.
You know I'm using a boring quote when I use a picture of a cute baby to keep the audience interested. I prefer the quote (from the book "Mistakes were made .... but not by me - which btw is going on my tombstone) about science being 'arrogance control'. Clinicians, and unfortunately I can't exclude myself here, can be some of the most arrogant people I now. We think we know that what we do is right and, if unchecked, our confirmation bias keeps it going.
Im not just bashing clinicians who wont use the evidence base. We need to be aware that as science evolves the evidence base will keep changing and we need to be ready to integrate new evidence into our formulations. For example the evidence for the cognitive behavioural model of BN is not as strong as I thought. I had selectively neglected a body of literature that indicates that dietary restraint may not lead to binge eating - although my clinical experience tells me it does. We all need to keep changing and adapting.
So, Im not bashing flexible manual users (which surprised some in the audience). Indeed I think I am one. The manual keeps me "anchored" to principles that are helpful. I try to integrate the concepts in the manual using my clinical experience - sometimes this is not the order the manual tells me. As clinicians we need to be aware of our tendency to drift into the latest clinical 'fad' or basing out therapy on the latest book we have read or workshop we have attended unless they have data to support them (ACT show me your data please).
Slides 24 & 25 are about the idea that maybe what we do is better than the manualised treatment. I feel all practitioners need to be accountable and to measure (and publish) their outcomes - this is the cornerstone of the generation of Practice Based Evidence. Be able to show what you do works. If it does - teach me what you do (make it disseminate-able).
We finish with the photo's of me cooking a BBQ, Jamie Oliver and Heston Blumenthal. I suggest manuals are a bit like cookbooks (which some take as a criticism). I can't cook. For me to cook chicken pizza - you need to give me a McCain's frozen Pizza that has all the ingredients included with strict instructions on what to do (including remove the plastic, turn on the oven to 180 degrees, place on a tray in the oven, leave in for 25 mins etc. Detailed step by step instructions. Heston may prepare a chicken pizza for you. It wont have chicken, pizza base and may be served as a gas - it is art, but it is not chicken pizza. Jamie however, has some instructions for chicken pizza, He includes a chicken and a pizza base. He can add to that base whatever he likes that might make it better. He doesn't need to be told when to remove the pizza from the oven - he just knows (he could tell you when to though). He uses the key ingredients in an understandable manner. He adds to the recipe in a way as guided by his experience, but his chicken pizza can be replicated by those who want to. When it comes to the treatment of our patients, I think Jamie serves as a model that integrates clinician and model.
I thought I might just provide some context for some of the slides so it makes some sense.
We start with a brief review of the evidence base in the eating disorders (along with one of my favourite quotes from Christopher Hitchens "What can be asserted without evidence can also be dismissed without evidence".
Slide 3 reminds us we do have Clinical Practice Guidelines to direct us, as well as a more recent Cochrane review and also (slide 4) great review by Bulik (up to 2005) and then Hay (2005-20013). Both published in International Journal of Eating Disorders.
Slide 5 suggests there is the strongest evidence for FBT in the treatment of adolescents with anorexia nervosa, but Slide 6 reminds us that that data (for full recovery - which is the standard to which I hold myself) indicates full remission for 50% of patients. (i.e. it's good but 50% ain't great).
Slide 7 & 8 makes similar point for CBT for BN. Good data suggesting good outcome for most, but again about 50% of patients remain symptomatic at end of treatment. There are also other treatments that are helpful for BN and BED, such as Interpersonal Therapy or DBT. The weight of evidence suggests however a first line treatment for BN should be CBT.
Slide 9 covers adult AN, where our results are poor. Most treatments get about 20% of patients to full remission, and get a substantially higher number to partial remission. 10,11, 12 summarise the latest study from Germany - the ANTOP study. This study again found that the type of treatment (a manualised psychotherapy or CBTE) didn't really matter in terms of outcome. Our treatment manuals aren't guiding us to what we should specifically do with adult patients in AN. I would point out however that the majority of the manuals focus on improving nutrition.
So we have some data to guide us on what may be helpful in the treatment of eating disorders. However, the question is how many practitioners are using the evidence base which suggests some empirically supported treatments exist.
Slide 15 was chosen not because the Tobin study is the best one indicating the Research Practice Gap, but because it has the coolest name (I know what you did last summer ... and it wasn't CBT) and it allowed me to put up a poster of SMG. There is a concerning "What Manual" with 21% of clinicians asked do they use a manual in the treatment of BN. I am more interested in the "flexible" user group, as this is where manual and clinician come together (maybe).
16 is on who uses manuals. Well when I here CBT, research focused and young I tend to think of my Doctor of Clinical Psychology and Master of Clinical Psychology students (some of who are pictured in the slide). When the first graduate the don't have the clinical experience, so should be sticking close to the manualised treatment where ever possible.
17 is some of the reasons people don't use manuals. I am most worried about those who don't know the manual. Slide 18 is an interesting study that indicated that even of those for whom CBT was their treatment of choice, few had formally being trained in the manual. ANZAED and the various state based training organisations must roll out training in evidence based treatment. I know Victoria have done this with FBT, as has NZ, but NSW is far behind.
RCT's dont exclude most patients - its an excuse we use when our results aren't as good as RCT. I have never met an eating disordered clinician who doesn't think they have the toughest patients that have all been excluded from RCT. The evidence generally suggests that treatments do disseminate if you use them.
Maybe our treatment models are "ignorant". I am thinking here of adult anorexia nervosa. Maybe drive for thinness or drive for control just doesn't cut it. I look forward to looking at the data coming out of neuropsychiatry labs like Walter Kaye's and the emerging research on Cognitive Remediation Training that may increase the outcomes for CBT.
The most common reason proffered for not using the Empirically Supported Treatment is that therapy is an art that should not be restrained by a cookbook. Surprisingly I agree with this to some extent. Mainly in the restraints that an RCT puts on the length of treatment. Maybe 40 sessions is not enough for adult anorexia (see the ANTOP studies weight graph that shows people keep getting better after treatment). Sometimes patients may need 30 sessions of FBT rather than stopping at 20 as the manual and RCT protocols suggest.
However, there are ideas in the manuals that help our patients. Our therapies need to reflect this.
As clinicians we need to remember that Evidence Based Practice is a three legged stool. Yes, one leg is Empirically Supported Treatments (our manuals). The other legs are clinical experience and also patients values and preferences.
Each of our camps tend to grab onto on leg of the stool and pull. The stool then looks a bit like a "hitting stick" (Slide 20) to beat people over the head with. Or as Phillipa Hay once wrote in response to our article "tyrannise" people with THE EVIDENCE.
However, as clinicians we need to however be aware of our own humanity and vulnerability to thinking errors - most importantly the confirmation bias (slide 21). This is when we do what we think works and collect evidence that it works and discount evidence it doesn't. This bias can lead us to do things that don't work continually, because we assume they do. It is this that, in my opinion, using manuals grounded in science helps us with.
You know I'm using a boring quote when I use a picture of a cute baby to keep the audience interested. I prefer the quote (from the book "Mistakes were made .... but not by me - which btw is going on my tombstone) about science being 'arrogance control'. Clinicians, and unfortunately I can't exclude myself here, can be some of the most arrogant people I now. We think we know that what we do is right and, if unchecked, our confirmation bias keeps it going.
Im not just bashing clinicians who wont use the evidence base. We need to be aware that as science evolves the evidence base will keep changing and we need to be ready to integrate new evidence into our formulations. For example the evidence for the cognitive behavioural model of BN is not as strong as I thought. I had selectively neglected a body of literature that indicates that dietary restraint may not lead to binge eating - although my clinical experience tells me it does. We all need to keep changing and adapting.
So, Im not bashing flexible manual users (which surprised some in the audience). Indeed I think I am one. The manual keeps me "anchored" to principles that are helpful. I try to integrate the concepts in the manual using my clinical experience - sometimes this is not the order the manual tells me. As clinicians we need to be aware of our tendency to drift into the latest clinical 'fad' or basing out therapy on the latest book we have read or workshop we have attended unless they have data to support them (ACT show me your data please).
Slides 24 & 25 are about the idea that maybe what we do is better than the manualised treatment. I feel all practitioners need to be accountable and to measure (and publish) their outcomes - this is the cornerstone of the generation of Practice Based Evidence. Be able to show what you do works. If it does - teach me what you do (make it disseminate-able).
We finish with the photo's of me cooking a BBQ, Jamie Oliver and Heston Blumenthal. I suggest manuals are a bit like cookbooks (which some take as a criticism). I can't cook. For me to cook chicken pizza - you need to give me a McCain's frozen Pizza that has all the ingredients included with strict instructions on what to do (including remove the plastic, turn on the oven to 180 degrees, place on a tray in the oven, leave in for 25 mins etc. Detailed step by step instructions. Heston may prepare a chicken pizza for you. It wont have chicken, pizza base and may be served as a gas - it is art, but it is not chicken pizza. Jamie however, has some instructions for chicken pizza, He includes a chicken and a pizza base. He can add to that base whatever he likes that might make it better. He doesn't need to be told when to remove the pizza from the oven - he just knows (he could tell you when to though). He uses the key ingredients in an understandable manner. He adds to the recipe in a way as guided by his experience, but his chicken pizza can be replicated by those who want to. When it comes to the treatment of our patients, I think Jamie serves as a model that integrates clinician and model.
RSS Feed
