Sunday, November 10, 2013

The Quality of Palliative Care: It Can and Must Be Measured


Is it important to provide high quality palliative and supportive care in patients with serious illness?  Does it matter if we treat pain in patients with metastatic cancer?  Does it matter if we listen to patients and provide care consistent with their goals?

I'm sure most health system leaders would answer an emphatic yes! to these questions.  But do they really mean it?  Will their actions match their words?  Unfortunately, the answer to that question will usually be no.  If you are serious about improving a care process, you need to measure that process, and use those measurements to improve.   Performance measurement has led to many process improvements in health care such as increasing the use of Beta Blockers after MI and the use of screening eye exams to prevent vision loss from diabetic retinopathy.  Shouldn't we also be using performance measurement to improve the quality of palliative and supportive care in the seriously ill?

This is what makes a study recently published in JAMA Internal Medicine a crucial leap forward in efforts to improve palliative and supportive care in seriously ill patients with advanced cancer.  A team led by first author Anne Walling (@DrAnneMWalling) and senior author Karl Lorenz of UCLA and the Greater LA VA Health System developed measures to assess the quality of palliative care in the Veterans Affairs Health System for patients with advanced cancer (metastatic colon, lung, or pancreatic cancer).

The quality indicators, which focus on pain, non-pain symptoms, and communication, are common sense measures of good and necessary care for seriously ill patients.  Each quality indicator takes the form of an IF, THEN statement outlining a clinical scenario and appropriate action.  For example:
  • If a patient with cancer pain is started on long-term opiod treatment THEN the patient should be offered a prescription or non-prescription bowel regimen within 24 hours or there should be a documented contraindication (52% compliance rate)
  • If a patient with cancer undergoes chemotherapy THEN prior to chemotherapy he or she should be informed about the risks and benefits of treatment, including likely benefits and adverse effects  and whether treatment intent is curative or palliative--16% compliance rate---:( 
In a wonderful accompanying commentary, Melissa Aldridge and Diane Meier (@DianeEMeier) from the Mount Sinai Department of Geriatrics and Palliative Medicine laud this study for proving that the quality of palliative care can be measured.  They applaud the VA Health System for supporting this large scale proof of concept effort to measure and improve palliative and supportive care.  

The study has good news and bad news for the VA health system.  As Aldridge notes, 86% of these patients received palliative care at some point in their illness reflecting the deep penetration of palliative care in the VA.  On the other hand, Veterans received only 49.5% of indicated care processes--So only a coin flip chance of getting the right thing done.  Clearly the VA has a long long way to go to improve the quality of palliative care for persons with advanced illness.   Since the VA has invested far far more in palliative care than the other US health systems, these numbers would certainly be far worse for the typical US patient with advanced cancer.

If we care about improving palliative and supportive care, we have to start by measuring it.  Hopefully this study will galvanize national efforts to improve the quality of life and well being for persons dealing with serious advanced illness.

by: Ken Covinsky @Geri_Doc

3 comments:

@kesleeman said...

Thanks for this post. I agree it is absolutely crucial that we measure the quality of palliative care we provide. I’m not sure that this JAMA Int Med paper is such a leap forward though. The paper describes a method of measuring process – shouldn’t we instead focus our attention on measuring patient and carer reported outcomes? If we have learned anything in the UK over the past few months from the LCP, it’s that process measures (how many people had morphine prescribed, how many people had written information given) are not an adequate proxy for outcomes (how many patients were pain free, how many patients fully understood the goals of care).

ken covinsky said...

Dr. Sleeman--Great question. You have a great point that we do need to keep in mind that our real goal is making things better for our patients, and our ultimate goal really needs to be focused on outcomes. It may not always be the case that improving processes improves outcomes, so some healthy skepticism is warranted.

Still, I think this study is an important step towards getting us to that goal. I am not sure how we get to the ultimate study of showing that we are improving outcomes without a reasonable way of measuring the processes that we at least think will get us there. Unlike a lot of process measures, the approach used here is sensible, in that each process is targeted towards specific clinical scenarios. The targeting, and the highly specific nature of the processes may make it more likely that they will improve outcomes. Also, a lot of these processes fall into the "just the right thing to do category." ie--it is hard to argue that patients with metastatic cancer should not be informed whether chemotherapy is being delivered with palliative or curative intent.

In terms of some of the issues you raise, the Aldridge editorial is a good read--while the editorial is very positive about this study, they note further refinement of these measures will likely be needed (ie, the current version requires a lot of time and resources to measure).

Like any challenging area, there are many pieces to the puzzle--and it will take many studies to get us to the right place. While I think this study is an important part of the puzzle, the point that we will not have the puzzle solved until we show we are improving outcomes is valuable.

Natural Death Centre charity said...

We absolutely support evidence based practise for caring for the dying and their families/carers as a way forward. As well as encouraging research we wish to promote how it is disseminated and in particular encouraging Open Access (Open Science) so mere mortals like me can read them easily – JAMA is closed access.

However we would also wish to advocate that if the outcome of research is to reflect on and improve practise and so improved outcomes then this can be done in tandem with other mediums. I’m talking of engaging health and social care professionals in reflecting on practise using the arts – yep, you heard it right. Empiricism did come from an arts basis ! I understand many palliative care teams have Journal Clubs to discuss a research paper and part of that is critiquing the methodology and then exploring their current practise and deciding what actions are needed. Well, how about occasionally having a poetry reading, watch a film and then discussion. (Tweeters see: #DeathArt ) Now this may improve palliative and supportive care - though it’s virtually impossible to measure it and I wouldn’t want to try as measurement isn’t the only true validity of knowledge.

At @ndccharity we collaborate with academic departments on research and encourage cross boundary research to promote innovative thinking out of the box. (Tweeters see #DeathAcademics )