Nursing Research: The Intellectual Tradition Nobody Teaches

The Braden Scale sits in the nursing flowsheet of nearly every hospital in the United States. Almost no physician knows where it came from.

Nancy Bergstrom and Barbara Braden built it and published the validation study in Nursing Research in 1987. Six subscales: sensory perception, moisture, activity, mobility, nutrition, and friction and shear. A nurse scores it on admission and again each shift. The score drives the specialty mattress, the turn schedule, and the wound consult that a physician then signs. Very few of those physicians could name the instrument's authors, the journal it appeared in, or the discipline that produced it.

That is the shape of the thing this essay is about. Nursing has a research tradition roughly seventy-five years old, with its own journals, its own institute at the National Institutes of Health, its own long methodological arguments, and its own formal account of what counts as knowledge in the discipline. Medical training does not mention that the tradition exists. Medicine consumes its output daily.

This is not a complaint about credit. It is a claim about what medicine cannot see. A discipline that does not know another discipline's methods exist cannot reach for them when its own methods run out. Medicine's methods run out regularly, at every question about what happens to a patient across months, in a house, with a family, under conditions the trial protocol excluded.

Where the Tradition Came From

Nursing Research began publishing in 1952. That date matters, because it means the discipline organized a research literature before it had doctoral programs built to supply one.

The supply problem was solved deliberately and strangely. Beginning in 1962, the Division of Nursing in the United States Public Health Service funded the Nurse Scientist Graduate Training Program. It paid for nurses to earn doctorates in other disciplines: sociology, anthropology, physiology, psychology, education. The understanding was that they would return and build nursing science with what they had learned elsewhere.

They returned with incompatible methods. That is the most important single fact about nursing research, and it explains most of what follows.

A discipline whose founding scholars trained as sociologists, anthropologists, and physiologists does not converge on one method. It inherits an argument. Nursing's research literature has been methodologically plural from the start, not as a stylistic preference but as a structural consequence of how its first generation of investigators was produced. Medicine's research tradition grew out of laboratory physiology and then out of the randomized trial, and it converged hard. Nursing's did not converge, and the failure to converge became one of its subjects.

Federal recognition took two more decades. Congress authorized a National Center for Nursing Research at the National Institutes of Health in the Health Research Extension Act of 1985, which became law when the veto was overridden. The center opened in 1986 under Ada Sue Hinshaw and was elevated to the National Institute of Nursing Research in 1993. Its budget has never approached that of the disease institutes. Its existence still changed what could be funded, and so what could be asked.

The Discipline That Wrote Down Its Epistemology

In 1978, Barbara Carper published "Fundamental Patterns of Knowing in Nursing" in the first issue of Advances in Nursing Science. The paper argued that nursing knowledge has four patterns that do not reduce to one another. Empirical knowing is the science. Esthetic knowing is the perception of the particular case as a whole rather than as a set of findings. Personal knowing is the clinician's knowledge of herself in relation to this patient. Ethical knowing is moral judgment under obligation.

Carper's claim was that a curriculum teaching only the first produces a clinician who is competent and incomplete.

The paper is assigned across nursing programs. Peggy Chinn and Maeona Kramer's Knowledge Development in Nursing, revised across many editions, is organized around it.

Medicine has never written the equivalent paper. Not because physicians lack esthetic, personal, and ethical knowing. Every physician deploys all three before lunch. Medicine has simply never made a formal disciplinary statement about what its knowledge consists of. Evidence-based medicine produced a hierarchy of evidence, which is a different object. A hierarchy ranks the reliability of empirical claims. It is silent on whether there are other kinds of clinical knowing and what their standards would be.

Nursing went further still. Jacqueline Fawcett proposed in 1984 that the discipline has a metaparadigm, four concepts that specify what nursing studies at all: person, environment, health, and nursing. The proposal was contested for years inside the discipline, which is the point. A field that argues in public about its own metaparadigm is a field that has one.

Ask a room of physicians to state medicine's metaparadigm. The question will not parse. Medicine has organizing concepts in abundance. It has never had to defend its right to exist as a field of inquiry, and so it never had to specify what it studies.

Nursing had to. The obligation produced a literature.

The Methods Nursing Built

The methodological contributions are the part that would surprise physicians most, because several of them are already inside medicine, unlabeled.

Grounded theory was developed at the University of California, San Francisco School of Nursing. Issue 032 of this publication told that origin story at length: Barney Glaser and Anselm Strauss on the wards asking who knew a patient was dying and who was pretending not to, and Jeanne Quint Benoliel carrying the question into nursing. What matters here is the address rather than the fieldwork. Juliet Corbin, a nurse, later became Strauss's principal collaborator and co-author of Basics of Qualitative Research.

Grounded theory is now the default qualitative method across sociology, education, management, and health services research. It was built in a nursing school, on a nursing question, and the tradition that carried it forward was substantially nursing's.

Margarete Sandelowski, a nurse and methodologist, spent a career on the question of what qualitative health research is for. Issue 030 of this publication took up her defense of plain description; the work that bears on this argument is the later work. With Julie Barroso she built the procedure for synthesizing qualitative findings across studies, published as Handbook for Synthesizing Qualitative Research in 2007. Systematic review had solved that problem only for trials, where every finding arrives as an effect estimate and estimates can be pooled. Twenty studies of what patients understood a diagnosis to mean produce nothing poolable, and before Sandelowski and Barroso there was no defensible way to state what those twenty studies had collectively established. They sorted findings by how far each had been carried from the raw data, then specified separate procedures for aggregating findings and for interpreting across them. It is the part of her work medicine has the most use for, because it says what a body of qualitative evidence adds up to.

Nursing also polices its own methods in public. Baker, Wuest, and Stern published a paper in the Journal of Advanced Nursing in 1992 on what they called method slurring, the habit of running phenomenology and grounded theory together as though the philosophical commitments underneath them were interchangeable. Medicine's parallel literature on qualitative rigor is thin, because medicine has rarely treated qualitative work as something with standards available to violate.

Then there is instrument development, which is where the Braden Scale sits. Merle Mishel built the Uncertainty in Illness Scale and the middle-range theory it measures. Janice Morse built a fall-risk scale now running in hospitals that have no idea whose it is. These are validated, published, and embedded in the daily operation of American hospitals, and they came out of nursing science.

Behind the instruments sits a measurement problem nursing took on directly. If the intervention is not a drug and not a procedure, how do you count it? Harriet Werley and Norma Lang built the Nursing Minimum Data Set in the 1980s to make nursing care visible in the record at all. The Iowa group produced the Nursing Interventions Classification and the Nursing Outcomes Classification in the 1990s, taxonomies of what nurses do and what changes when they do it. The American Nurses Association's national database of nursing quality indicators followed in 1998. These systems are imperfect and have been criticized inside nursing for thirty years. The criticism is the tradition working. What matters is that a discipline faced the problem of measuring an intervention that has no dose and built something instead of declaring the question unanswerable.

The measurement work made the outcomes work possible. Linda Aiken is a nurse and a sociologist, and her group's 2002 paper in JAMA reported that each additional patient in a surgical nurse's assignment was associated with roughly a seven percent increase in the odds of a patient dying within thirty days of admission. Issue 002 of this publication carried the companion finding on bachelor's-prepared staffing and the European replication that followed it. Issue 004 carried failure to rescue, the measure Jeffrey Silber built in 1992 and Aiken's group turned into a staffing question, alongside the argument that most in-hospital arrests are visible in the nursing record hours before they happen. The staffing literature is the other half of that argument. It asks whether anyone is at the bedside often enough to look.

What neither of those issues said is where that work came from. It came from a discipline that had already spent thirty years arguing about how to count an intervention with no dose. Those findings are among the most cited in health services research. They are mortality findings, which is the currency medicine takes most seriously. They are nursing research, and they are not in the medical curriculum.

Nursing also arrived at the implementation problem before medicine had a name for it. The Conduct and Utilization of Research in Nursing project ran out of the Michigan Nurses Association from the mid-1970s under Jo Anne Horsley, funded by the Division of Nursing. Its question was not whether a finding was true. Its question was what an organization has to do to get a true finding into practice. Cheryl Stetler's model of research utilization dates to 1976. Gordon Guyatt's group at McMaster named evidence-based medicine around 1991, and the Evidence-Based Medicine Working Group's paper appeared in JAMA in 1992.

The frameworks that followed are the ones medicine now uses without attribution. The Iowa Model of Evidence-Based Practice, from Marita Titler and colleagues, is a nursing product. So is PARIHS, Promoting Action on Research Implementation in Health Services, published by Alison Kitson, Gill Harvey, and Brendan McCormack in Quality in Health Care in 1998 and revised as i-PARIHS in 2016. Kitson is a nurse. PARIHS is among the most widely used frameworks in implementation science across every health profession, which is a field substantially built on nursing's earlier work on the same problem.

What the Asymmetry Costs

Issue 032 of this publication argued that the trial apparatus is not the problem, and I am not reopening that here.

The narrower argument is this. There is a class of clinical question that the trial is not built to answer, and nursing has spent seventy years building methods for exactly that class. When a physician hits one of those questions, the tools are on a shelf in a room she was never told about.

The class is not exotic. It is most of what happens in a clinic. What does this patient believe is going on inside her body, and what will she therefore do with the prescription. What changed in the household when the diagnosis arrived. Why does a protocol that holds on one unit fail on the unit upstairs with the same policy and the same equipment. What has to be true of an organization before a proven finding survives contact with it. Each of those is answerable. None is answerable by randomizing anyone, because none is a choice between two available options.

The methods that answer them have names and published standards. Issue 026 of this publication observed that nursing doctoral training has treated them as first-class designs for fifty years, and the consequence shows on either side of a dissertation defense. A nursing PhD student defends a qualitative study on its own terms. A resident who proposes one is asked what her sample size is, and the question is rarely a request for information. It is a ruling that the work does not count. That is the asymmetry, and it is not about access. The papers are not hidden. What medicine lacks is a position from which a qualitative finding could be authoritative, because the only formal account it ever wrote of its own knowledge was a ranking of empirical claims, and a finding that is not an effect estimate has no rung.

Issue 003 of this publication argued that nursing named the theory-practice gap and medicine did not. Marlene Kramer's Reality Shock, published in 1974, is the empirical study behind that naming, and it is a study medicine could have replicated on its own interns at any point in the last fifty years. Issue 012 argued that Risjord's standpoint account was an introduction to a philosophical tradition that continues well past where he stopped. Each of those is one thread. The tradition is the cloth, and it is not taught anywhere a physician will encounter it.

A medical student reads no nursing research in four years. A resident reads none in three. An attending meets nursing findings only after they have been stripped of provenance, as a score in a flowsheet or a ratio in a policy memo.

The result is a physician workforce using the products of a research tradition it cannot name, cannot evaluate, and cannot contribute to. When a Braden score looks wrong, the physician has no way to reason about the instrument, because she does not know it is an instrument. When the staffing question reaches a hospital committee, she has no access to the literature that settles most of it.

Nursing research is not an alternative to medical research. It is a parallel apparatus, differently built, aimed at questions medicine has mostly declined to ask, holding answers medicine would use if it knew they were there.

About the author

Peter Schindler, MD, PhD is an Assistant Professor of Medicine and Associate Program Director of the Community Health Center Family Medicine Residency Program at the University of Nebraska Medical Center. He practices at Winnebago Comprehensive Health System, OneWorld Community Health Center, and Nebraska Medicine. He completed a Primary Care Research Fellowship at McGill University and holds a BSN from the University of Wisconsin-Oshkosh, an MS and PhD in nursing from Emory University's Laney Graduate School, a Diploma in Tropical Medicine from the Liverpool School of Tropical Medicine, and an MD from the Medical College of Wisconsin-Green Bay. He is board certified by the American Board of Family Medicine. The Interprofessional publishes new essays every week at the intersection of medicine, nursing, and the clinical knowledge that lives between them.

Disclaimer. The views in The Interprofessional are Peter Schindler's own and do not represent the official positions of the University of Nebraska Medical Center, Winnebago Comprehensive Health System, OneWorld Community Health Center, Nebraska Medicine, or any other affiliated institution.