The Risky Business of Wound Research, Algorithms and Systems

DMCA.com Protection Status
Blog Category: 
Wound Research Data Review Including Outliers

by Margaret Heale RN, MSc, CWOCN

The research lecturer's name was Terry, and he had my respect and attention. Many of the students were dreading the research modules but were cheered by the prospect of Terry taking us through it.

Terry included this little gem in his introduction to research. A group of researchers came up with a hypothesis that spiders had ears in their legs. How could they prove this, they wondered. They procured some arachnids of exactly the same type and took great pains to match them for all important parameters. They checked carefully the hearing of each. Placing them in turn in a quiet, stable tray, each time they yelled "Run!" every spider would take off at speed at the sound. The next step was to remove the spiders' ears, so the legs of each spider were removed. Now when the researchers yelled "Run!" the spiders did not hear and did not move at all. So there you have it, clear undeniable proof, spiders have ears in their legs. It is very easy to see the flaw in this rather cruel piece of research, but sadly it is not always that easy.

Wound Study Design Types

When looking at randomized controlled trials one of the first things you read is a one liner, "subjects were matched," and there may be a list that includes stage of pressure injury, size of wound, age, sex, and a myriad of other things somebody decided to include. There may also be exclusion criteria such as uncontrolled blood sugar, obesity, and being over 60 years old. It makes sense to do this, and there is no doubt that once you have got homogeneous groups and compare the outcome of one with another, after whatever intervention you wish to discover the worth of, the result may look gratifyingly convincing. Please check that the legs of the spider are still there. For a very long time, wound care research compared the dressing being tested against the "standard of care," which happened to be "wet-to-dry dressing." Well, it is hardly surprising that the manufacturers managed to show their modern dressing on a clean wound worked better than a debridement technique!

If there are 1,000 patients and only 100 fit the criteria for researching a product, don't be surprised when the product doesn't work so well in the reality that is the other 90% of the population. This is not to say that research is not valuable and necessary, just that care needs to be taken when announcing a product or treatment regimen is research or evidence-based and hence a better option than a non-researched alternative.

As a wound clinician, I have determined case studies can be very useful tools at times, particularly when it comes to wound care products. Criteria for use and the rationale can be so clearly laid out, and comparisons with patients known to the reader stand out. Even a small group of case studies can "prove" a product or approach is a worthwhile venture.

Protocols and Algorithms in Wound Management

There are similar issues with protocols and algorithms. These are set with the majority in mind, so when a patient is different for some reason (an outlier), if the difference is not recognized the patient may be at risk. An example may help here. Protocol states patients likely to be on the OR table for over 4 hours must follow the special algorithm relating to pressure injury prevention. A patient who had a difficult to heal stage IV pressure injury of the sacrum that healed 6 months ago who is going in for surgery expected to last a few hours would be missed and be at high risk.

Outliers: Looking at the Patient as an Individual

The systems in place to prevent harm to patients work only if everybody involved appreciates the limits of the system. Example: a patient calls the hospital where he is to have surgery the next day and asks if it is okay to take his migraine medication as it puts his blood pressure up. The person taking the call says it should be fine but to mention it to the anesthesiologist. The next day the patient dutifully tells the nurse doing his pre-operation vitals that he took the migraine medication, and he also tells the nurse putting in the pre-operation IV. When he tells the anesthesiologist, the patient is not convinced the anesthesiologist is listening and points out his blood pressure is up. Sinus surgery finished, the patient wakes to hear "OMG... he's bleeding, get him back in the OR." Later, when the patient is awake in the recovery room, the surgeon comes and asks the patient why he didn't tell anyone he was hypertensive! Needless to say, the patient has to have IV antihypertensive meds and an overnight stay. Such systems failures are the reason why patients are harmed while under our care and why we all have to be on the alert for outliers that don't quite fit. They are at risk just because of their difference, whatever it might be.

As for labeling a system or a group of treatments as a bundle, this is just asking for a problem. For me, the term bundle means just lopping everything together in a "not so neat" package. A pressure injury bundle might include putting a couple of heel offloading boots on every patient with a Braden score under 16. This will miss the usually fit, well-nourished 78 year old with rheumatoid disease, scoliosis, on prednisone, who is moving well but recovering from a total hip replacement. So the lesson is keep your eyes on differences, outliers, the tails of the normal bell-shaped curve, high risk, low volume. For safety, keep your eyes wide open.

Suggested Reading
Nichol AD, Bailey M, Cooper DJ, POLAR; EPO Investigators. Challenging issues in randomised controlled trials, Injury. 2010;Suppl 1:S20–3.

About the Author
Margaret Heale has a clinical consulting service, Heale Wound Care in Southeastern Vermont and draws on her extensive experience as a wound, ostomy and continence nurse in acute and long-term care settings to provide education and holistic care in her practice.

The views and opinions expressed in this blog are solely those of the author, and do not represent the views of WoundSource, Kestrel Health Information, Inc., its affiliates, or subsidiary companies.

Comments

I appreciate the succinct summary of this very important topic. Many clinicians look for a case study patient that is similar to their own patient when they are looking for a solution to a difficult wound problem. Although case studies have the limitation that they are cherry-picked data, they do demonstrate clearly the capabilities of the choice being featured. That is why it is important to look for a product that is supported by the full range of evidence types: well designed RCTs, quasi-experimental designs like historic control studies, descriptive studies like facility evaluations, case studies, and expert opinions.

It is easy to assume that an RCT is always better evidence than a case study, but a seriously flawed study design can result in very misleading findings. The first thing to look at is the comparator. If the researchers did not even define standard care, suspect the worst. The most egregious study I have seen recently "proved" that a dressing decreases heel breakdown by comparing it to a wad of gauze taped to the patient's heel. The "standard care" group had a whooping 50% rate of heel pressure damage because the researchers actively created a pressure point. Peer-reviewers overlooked this ethics violation and a leading journal published this study!

Most industry-sponsored wound studies compare the product the authors are being paid to test against plain gauze, which has not been the standard of care in most countries for decades. Others simply compare two of their own products to each other, and then tout the one that came out on top as a statistically superior product. In contrast, studies by independent authors, while usually smaller and less rigorous, are more likely to use as comparators what you and I are using today. They are also less likely to exclude the patients that we typically see. Findings from real life settings conducted by independent researchers are more likely to translate to our practices.

I recently summarized all of the evidence documents that include polymeric membrane dressings as a unique wound dressing subtype. Looking only at studies of pressure damage, I found 20 level 1 and 2 reviews and studies, 5 observational studies, 46 case studies or case series documents, and 6 expert reviews. Importantly, these documents were written by 72 completely independent authors, 16 authors who received compensation (often an unrestricted grant) and only 5 company employees. This highly relevant, mostly independent, body of evidence is very compelling.

Again, Ms. Heale, thanks for calling attention to such an important topic. Before we applaud a research study, we need to make sure the spiders have legs.

Sincerely,
Linda Benskin, PhD, RN, SRN (Ghana), CWCN, CWS, DAPWCA, FRSM
Clinical Research & Education Liaison, and Charity Liaison
Independent Nurse Researcher and Ferris Mfg. Corp.

Benskin LL. Evaluating All Available Evidence Pertaining to Using PMDs for Pressure Injuries. presented at NPUAP 2018 and SAWC 2018.

Excellent point about using wet to dry as 'standard of care'. It is key, when evaluating a wound dressing study, to look at the comparator to not only evaluate if it is still a care standard but to also ensure that proposed dressing actions are congruent.

Add new comment

Important Notice: The contents of the website such as text, graphics, images, and other materials contained on the website ("Content") are for informational purposes only. The Content is not intended to be a substitute for professional medical advice, diagnosis, or treatment. The content is not intended to substitute manufacturer instructions. Always seek the advice of your physician or other qualified health provider with any questions you may have regarding a medical condition or product usage. Refer to the Legal Notice for express terms of use.