Outcomes at SunCloud Health

We here at SunCloud Health are thrilled to announce that after two years of hard and persistent work, we are very close to being in a position to release the first results of our patient reported outcome measurements! Built in to our DNA is appreciating the immense value this data provides to us, our patients, their families and our payers- and then figuring out what to do with it. Seeking to be a pioneer in value based delivery of behavioral healthcare we are inspired by internationally respected healthcare thought leaders such as Michael E Porter, Thomas H Lee, MD and Leemore Dafny. We aspire to be like hospital systems such as The Cleveland Clinic and Germany’s Schoen Clinic who have already proven that a focus on value over volume is not only good for patients but it is good for business. Currently we using the data we are collecting to influence our ongoing clinical care, providing our clinical team with patient reported information on outcomes that matter to our patients. Our outcomes are incorporated in real time into the process of care, allowing us to track progress and make clinically appropriate changes as we interact with our patients.  We can see where and when patients report they are improving (or not) and compare that to clinical observations.  We are able to use this data when justifying level of care recommendations both to our patients and their payers.  We use the data to see where our patients feel we can improve, and where they feel we may be excelling.  We can drill down in to an individual patient’s results and we can see average trend lines among all of our patients.  We collect data on admit, weekly during treatment, at discharge and then post discharge. Admittedly post discharge collection remains our biggest challenge, and this will likely always remain the case.  Not yet are we able to compare ourselves to other providers in our field, but we are one of the few who hope this is coming- and for our patients sake, the sooner the better. We look forward to sharing our process and our results in an honest, transparent and meaningful way in the very near future. We hope this will be an inspiration to all providers as we all seek a common goal of improving quality and reducing cost. Although the process of collecting, using and sharing data like this can be time consuming and might appear at first to be Daunting, in the end doing so is ultimately in the best interest of our patients- and that is really what matters most. Outcomes need to be standardized and risk adjusted in order to mean much. Patients beware! We should all be thrilled to see that a few providers are beginning to publish health outcomes for mental health and addiction treatment.  We see them on websites showcasing patients progress and occasionally we see them presented at industry conferences by providers advertising their successes.  We use outcomes in healthcare as one way of measuring the quality of care we are providing for our patients, their families, and their payers. Only once we have outcomes can we then take the next crucial step of measuring value for our patients and their families where value is a function of outcomes that matter to patients per dollar spent. Once we are finally able to measure value can we then make the badly needed transition to a system that is centered around quality instead of solely quantity (aka fee for service).  This transition has proven in many other areas of medicine at major health institutions to align the interests of patient, provider and payer and results in improved quality at a lower cost. In mental health and addiction treatment the outcomes that matter to our patients are often difficult to measure.  If our job is to mend a broken bone, we x ray the bone a few months post-surgery and it has either healed or it has not healed.  In our field, it’s not so easy…  We want to know if our patient’s overall mood has improved, if they are able to engage in healthy relationships, if they feel satisfied with their employment situation, if they are feeling less alienated and more connected in their community, if their eating habits and their relationship with food and their bodies are healthy and if they have been sober…  In order to assess much of this type of information we use validated patient reported outcome measurement tools such as the PHQ-8 or 9, the EDEQ (Eating Disorder Questionnaire), the ASI-LITE (Addiction Severity Index), the OQ-45.2 and many of us use a patient satisfaction survey.  These measurements tools are not perfect, but they have proven to be statistically and clinically reliable.  Many of us who are collecting this data wisely use it to measure ourselves.  Where we see internal deficiencies, we try and improve.  If nothing else the data collected serves as one additional piece of information, we can use to support our patients. Externally, those of us confident in the quality of care that we are providing want to transparently provide outcomes to prospective patients, their families and their payers as they assess where to go for help (and why). To be useful, however, these outcomes must be standardized, and they must be risk adjusted.  Standardization means we are all using (and publishing) the same tools for the same patients in the same way at the same time.  If we are using the EDE-Q and our friends across the street are using the EAT-26 (another tool used to measure eating disorder progress), for instance, the two cannot be compared against one another and therefore any comparative analysis is useless.  Agreeing on which tools to use is a process that we hope will happen soon.  Once we all agree, however, this is fairly straight forward. Risk adjustment is a bit more complicated and yet as if not more important.  Risk adjustment takes into consideration the underlying health status of the patient being measured.  The more complicated the patient’s condition, the “riskier” they are and therefore the more resources will be needed to treat them effectively.  Further, the more complicated the patient, generally the less improvement we might expect to see or at the very least the improvement in a complex patient will look different.  Without both standardization and risk adjustment to outcome results, we simply have no way to determine how well patients at a facility are improving as compared to similar patients at another facility.  Yet this is critical if any value is to be placed in reviewing any one provider’s results. For example, take the example of an addiction treatment facility which publishes abstinence rates showing that 80% of its discharged patients remain abstinent at 365 days post treatment; Or Look at a treatment center that treats depression which publishes a decrease from 14 to 9 from admission to discharge in their PHQ-8;  Or, consider the results published by a treatment center that treats eating disorders where they show a decrease from 4 to 2 in their EDE-Q from admit to discharge while the non-clinical person measures a score of 1.4.  On the surface these results appear to show that treatment in these respective programs works.  This is all very good news and probably true to some degree. However, if we want to know whether these results are good, bad or average as compared to other programs treating similar patients, on their own in the vacuum in which they are presented these scores fail miserably in this effort. First, we must ensure that what we are measuring is precisely what others are measuring and then doing so the same way at the same time (ie. weekly in treatment, 30 days post treatment, etc).  Using the abstinence measure example above, for example, if the definition of abstinence is not clearly defined the same way across all providers measuring it, then no fair comparison can be done.  Does it mean no use since discharge, no use in the past 30 days, past 2 days???  If one provider is asking the question via email and the other is asking it in person, it’s probably not all that useful to assume the results can accurately be compared.  Standardization is critical. Such is the case in all data analyses. As for risk adjustment, imagine if the addiction facility that publishes the 80% abstinence rate at 360 days post discharge generally treats patients who have never been in treatment before, are relatively young, have not struggled with the disease for a long time and who do not have any co-occurring depression, anxiety, or trauma…the 80% score might look fantastic but what if another provider uses the same tool, measures the same way and yet shows a 60% abstinence rate? However, when looking more deeply at this provider using some form of risk adjustment model), we see their patients are significantly more complex in that they are generally older, have in many cases been in and out of treatment for much of their lives and have co-occurring trauma, depression, and anxiety along with a host of medical co morbidities.  Would we say this provider is producing lower quality work because their abstinence rate is 60% as compared to 80%?  Probably we would not. In fact, we might say the opposite; namely, that the first provider could possibly be doing better. The same would be said for the other two examples above. I have seen one published report that shows patients improve from a 14 to an 8 using the PHQ-8 scale and another report whose patients improve from an 11 to a 2!  If the latter is telling the truth, undoubtedly their patients are far less complex and acute than the former. Either that or they have some proprietary miracle intervention they are using in which case I suspect we would all know about it by now…  Thus, the two results are likely not comparable and yet if not scrutinized a potential patient might assume the latter provider is a better fit for them! If a facility that treats depression for example sees mostly relatively mild cases of depression and their “success rate” is say 80%, does that mean they are doing far better work than the facility that specializes in treatment resistant depression where patients have generally had at least one suicide attempt and whose “success rate” is 20%?  The answer is no.  Without any sort of risk adjustment on the acuity and complexity of patients being measured, we really have no idea what we are looking at when trying to compare results between multiple providers. One clear example is in looking at the survival rate (an outcome) of stage 1 lung cancer versus the survival rate of stage 4 lung cancer.  With stage one the survival rate is anywhere between 70-90% whereas with stage 4, it’s around 10%.  If we were considering Mayo and Cleveland Clinic for our cancer care, we would likely ask them what their survival rates are not just for any cancer but specifically for the stage and type of cancer that we might have.   The same analysis cannot be done without risk adjustment and standardization in addiction and mental health treatment and yet without it these outcome measurements don’t mean much. As mentioned above, standardization is the easy part.  We simply need to agree on a universally accepted set of measurement tools for the conditions we treat and then collect them the same way.  Easier said than done but still relatively simple once we agree to do it.  The risk adjustment part is more difficult.  These diseases are often difficult to diagnose and “stage”.  It takes time to get to know our patients and as time passes, we often see that there is far more than what met the eye initially at the time of the evaluation.  Still, we can start with something and some of the ways we risk adjust at SunCloud are by using the following.

• Level of care patient is coming from previously? Coming from inpatient, higher risk. Coming from outpatient, lower risk.
• How many diagnoses does the patient have? The more they have, the higher risk they are.
• Has the patient been hospitalized for their addiction/mental health issue in the past 5 years?
• Does the patient have a supportive family?
• Are their medical comorbidities at this time as a result of the patient’s condition?
• Any history of overdose or suicide attempt?

Once we finally have standardization and eventually some form of risk adjustment model, the outcomes some of us collect will be far more valuable to patients, their families, and their patients. Meantime some data is undoubtedly better than no data and those of us collecting it and publishing it should be commended. We just advise everyone to take all of it with a grain of salt. At least for now. Read more from David Newton, SunCloud Health’s Director of Operations, on today’s issues in behavioral health,  about the need for accountability, transparency and outcomes in behavioral health.