1. Olson, DaiWai M.

Article Content

Scientists use terms such as category, bin, group, class, and section to designate and differentiate variables. Humans like to categorize things. It makes it easier for us. However, I have begun to wonder of late whether this is really in the best interest of the nursing profession. This hit home for me as I was listening to a neurologist lecturing on stroke who began discussing results from a study where the modified Rankin scale (mRS) was used as a dichotomous outcome. This got me thinking, "[horizontal ellipsis]is it different if you are dead?"

Figure. No caption a... - Click to enlarge in new windowFigure. No caption available.

As a quick review, in 1957, Dr Rankin1 published an article in which he had a scale that ranged from 1 (no significant disability) to 5 (severe disability). In 1987,2 the scale was extended by adding the value "0" to designate "no symptoms." A 2005 reliability study3 found rather limited interrater reliabilty of the 0-to-5 scale. The current version, which emerged somewhere around 2008, now includes the value "6" to designate "dead." Interestingly, it is also in 20084 where we begin to see controversy over dichtomizing the mRS.


Dichotomize is a verb used to describe the process of dividing something into 2, not necessarily equal, parts. When a variable is dichotomized, all of the observations must be assigned to 1 of the 2 categories. Recent history has shown us that this approach does not always accurately represent all members of the group. Hence, dichotomous variables such as female and male, black and white, or young and old, although easy to understand, are falling out of fashion (albeit slowly). In response, researchers are developing new methods of describing sex, race, age, and many others. However, I'm not sure we have figured out what to do about neurological outcomes in nursing research.


The 3 most common dichotomization cutpoints for mRS are 1 or less, 2 or less, and 3 or less as good outcomes, versus greater than 1, greater than 2, and greater than 3 as bad outcomes. This approach means that being alive with moderate disability (mRS, 4) is scored exactly the same as being dead (mRS, 6). Setting aside the idiom that "there are things worse than death[horizontal ellipsis]," I really think that being dead is a different outcome than living at home and needing assistance with mobility and personal bodily needs.


Important Disclosure

I have authored articles with mRS dichotomized,5 and the Journal of Neuroscience Nursing has published-and will continue to publish-articles where mRS is dichotomized.6 Moreover, I am only using mRS as an illustrative example; it is not the only scale where death is lumped in with another outcome. This editorial is by no means a treatise on the mRS. Rather, it is a call for opinion, insight, and new direction.


Nursing is one of many healthcare professions. A distinction to our profession is that nursing is fundamentally patient focused and not disease focused. Although neuroscience nurses may study the same patients as their physican counterparts, our research has a unique scope. We have a different lens through which we examine different interventions and different outcomes. Moving our profesison forward requires the combined knowledge of our entire profession. What outcomes are sensitive to nursing care, and how should they be measured? In addition to letters to the editor, the Journal of Neuroscience Nursing welcomes Research, Reflections, Case Study, and Clinical Practice articles. I hope you are willing to share your knowledge.


Dr. Olson declares that he is the Editor of the Journal of Neuroscience Nursing.

Figure. No caption a... - Click to enlarge in new windowFigure. No caption available.



1. Rankin J. Cerebral vascular accidents in patients over the age of 60. I. General considerations. Scott Med J. 1957;2(4):127-136. [Context Link]


2. van Swieten JC, Koudstaal PJ, Visser MC, Schouten HJ, van Gijn J. Interobserver agreement for the assessment of handicap in stroke patients. Stroke. 1988;19(5):604-607. [Context Link]


3. Wilson JT, Hareendran A, Hendry A, Potter J, Bone I, Muir KW. Reliability of the modified Rankin scale across multiple raters: benefits of a structured interview. Stroke. 2005;36(4):777-781. [Context Link]


4. Weisscher N, Vermeulen M, Roos YB, de Haan RJ. What should be defined as good outcome in stroke trials; a modified Rankin score of 0-1 or 0-2? J Neurol. 2008;255(6):867-874. [Context Link]


5. Ortega-Perez S, Shoyombo I, Aiyagari V, et al. Pupillary light reflex variability as a predictor of clinical outcomes in subarachnoid hemorrhage. J Neurosci Nurs. 2019;51(4):171-175. [Context Link]


6. Alexandrov AW, Palazzo P, Biby S, et al. Back to basics: adherence with guidelines for glucose and temperature control in an American Comprehensive Stroke Center sample. J Neurosci Nurs. 2018;50(3):131-137. [Context Link]