What information consumes is rather obvious: it consumes
the attention of its recipients. Hence a wealth of information creates
a poverty of attention, and a need to allocate that attention efficiently
among the overabundance of information sources that might consume
it.
-Herbert Simon,
Nobel Laureate Economist, 1971
Knowledge is power.
-Sir Francis Bacon, 1597
Some books are to be tasted, others to be swallowed,
and some few to be chewed and digested.
-Sir Francis Bacon
Medical information that is relevant to orthopaedic surgeons
is continuously expanding and changing, while older information
is expiring24,25. The speed at
which new information is generated and disseminated challenges us
to adopt new strategies for acquiring useful knowledge and leaving
unusable information aside. Reading faster or spending more time
reading does not ensure that one is gaining more relevant orthopaedic
knowledge. In addition to the challenge of processing ever-increasing
amounts of orthopaedic information, there exists the timeless problem
of discriminating "the truth" from that which is "not-truth" (incorrect
or not clinically useful). The ability to sift through large amounts
of written material to uncover a few useful truths that will improve
one's practice is a skill that busy orthopaedists and current residents
should learn19,21.
We live in an electronic information age in which patients often
have more citations at their command than do practitioners15,16. Our present dilemma is that
a large quantity of information is quickly available, yet more and faster
information services have not translated into better care for patients15,20,21. The management or application
of information related to orthopaedic care, combined with the skill
of treating musculoskeletal disease, is what will set the orthopaedic
surgeon apart from the educated lay public and from physicians who
are not trained in managing orthopaedic disorders. As we enter the
third millennium, we must add constantly to our knowledge base and
skills repertoire while paring away unused or obsolete information
that clutters the mind and memory.
Orthopaedic practitioners gather information for four basic reasons26: (1) to keep up with new developments
in clinical medicine, (2) to answer a question related to a specific
patient, (3) to review and reinforce previously learned information,
and (4) for fun or to keep up with a subspecialty of interest. Depending
on why the information is needed and how it is to be used, different
information sources (such as journals, videos, lectures, and textbooks)
will provide the solution to the quest for practical knowledge1,17. The acquisition of knowledge
and the acquisition of the wisdom to use or apply this knowledge
are the ultimate goals of orthopaedic research and education (Fig. 1). This paper
introduces concepts that should help orthopaedic surgeons to manage
their information18,19 and time
while ultimately improving musculoskeletal patient care20.
In the past two decades, we have had to learn and apply new diagnostic
and treatment technologies, such as computerized tomography, ultrasound,
magnetic resonance imaging, arthroscopy, complex external fixation,
bone transport and callotasis, intra-medullary fixation, complex
spinal instrumentation, and replacement arthroplasty. There were
false starts and blind alleys. Remember polyester ligament substitutes,
hip and patellar resurfacing prostheses, laser meniscectomies, and
direct-current bone stimulators? An orthopaedist may embrace an
innovation or a new technique on the basis of little evidence and
much advertising. At other times a good idea, such as the use of
an endoprosthesis for the treatment of femoral head avascular necrosis,
does not prove itself over time. The fundamental knowledge of anatomy, surgical
approach, and physical assessment, learned in residency, will probably
serve for the lifetime of an orthopaedist without drastic revision
or change. Progress in basic science, on the other hand, will change
some of what orthopaedists need to know for recertification and
will gradually find application as new technology in patient care.
In the future, what will change markedly is the reporting of outcomes
based on clinical evidence. Rather than examining how many knee
implants survived for fifteen years, we will examine how many quality-adjusted life-years
a cemented implant provided to the average osteoarthritic or rheumatoid
patient compared with those provided by an uncemented implant. We
will also examine what additional productivity an arthroscopic partial meniscectomy
provided to the average person in his or her prime working years
compared with that provided by another reasonable treatment. In
other words, orthopaedic surgeons will be judged by the evidence
that their interventions are doing the most good for the most people
at a price that the country can afford.
For those who practice general orthopaedics, it remains relatively
easy to keep up with patient-evaluation methods and technologies.
There are few new diagnoses or new pathological conditions; there
are some improvements in confirming diagnoses. The greater challenge,
and frustration, is keeping up with current aspects of patient management:
new medications (for example, cyclooxygenase-2 inhibitors and quinolone
antimicrobial agents), new applications of technology (for example,
computer-assisted surgery and cloning of human tissue), and refinements
in techniques (for example, arthroscopic procedures compared with
open procedures). Even with common problems that have relatively
simple treatments, the increasing constraints of a busy practice
minimize the time available to critically evaluate new treatments. Lack
of time limits the ability to do extensive literature searches and
to use scientific measures of outcomes in one's practice.
Guilt accumulates in direct proportion to the growing pile of
unread journals and the mounting stack of invitations to desirable
courses that promise hands-on learning experience22.
This problem is compounded when a patient arrives with an article
from the Internet that describes the outstanding results from the
latest technique, which combines laser arthroscopy with genetically
engineered clones of autologous chondrocytes. In addition, the generalist must
cope with a dizzying array of new classes of medications that improve
pain control, bone density, or fracture-healing. One has a hard
time keeping abreast of all areas of orthopaedic interest.
As this river of information flows by, it is difficult to identify
which information, knowledge, and wisdom we really need and then
incorporate it into practice14,18,27.
Should you be the first, or the second, or the last orthopaedist
in your community to perform a thermal reduction of the shoulder
capsule with use of an arthroscope? The old adage "Never be the
first or last orthopaedic surgeon to learn a new technique" should
be tempered by the credo "First of all, do no harm (primum
non nocere)." Both sayings are relevant if we are to achieve balance
in this setting.
A number of studies in the medical literature have shown that
there is an unacceptable lag between the time of publication of
credible scientific information that should change medical practice
and the time of the adoption of this change by practitioners3,22,26. Fineberg11 identified
the impact that twenty-eight so-called landmark papers had on the
practice of medicine and discovered that only two had an effect
within two years of publication. Clearly, the concept of a landmark
article must be reconsidered11,17.
Superficially, the solution to the information problem is easy.
Find a way to get the information to orthopaedists, and they will
change their way of doing things. However, the evidence suggests
that, even with appropriate information, clinicians are reluctant
to change their management and treatment behavior4.
Evans et al.10 found that the
strongest predictor of a physician's knowledge and management of
hypertension was the clinician's year of graduation from medical
school.
Two additional problems remain. The first is that the knowledge
needed to make certain management decisions does not exist. A new
surgical procedure has a waiting period before a meaningful comparison
can be made with established procedures. We do not know if patients will
be better ten years from now if they have their shoulder instability
treated with an arthroscopic laser or diathermy capsular reduction
rather than with more conventional open surgical plication of capsular
tissue. The choice of a new surgical procedure should be based on its
benefits compared with those of the existing treatment and on the
hope that it will achieve at least equal success. Once an innovation
demonstrates comparable or improved outcomes, it becomes popular
because of reduced cost, reduced morbidity, or ease of performance
compared with those of the previous technique. The long period between
the development of an innovation and the acquisition of results
underlies the conservatism of most orthopaedic surgeons and increases
the difficulty of evidence-based persuasion to improve surgical
practice.
The second problem with information management is that once we
are aware of evidential knowledge we are hesitant to put this knowledge into
practice3,14. Researchers who
have studied the diffusion of information in medicine have reported
that innovations in surgical and medical practice are widely adopted
only after they are first adopted by a so-called opinion leader2,12,14, an influential member who
is trusted by others in the community. These researchers also indicated
that physicians obtain information from many sources and place greater
credibility on some than on others. No matter where or from whom
the information is obtained, for the most part its roots can be
traced to one source: medical journals2,5,22,25,28.
Journals are the major sources of new information for both researchers
and practitioners6,23,26. However,
these two groups approach this information quite differently. When
researchers evaluate articles in their area of expertise, they usually
are familiar with all of the research that has been published previously23,28. Clinicians may not be so well
read in the dialogue generated by previous publications and, in a
sense, are listening to the middle of a conversation. Reading a
journal article is similar to listening to only part of a conversation
or hearing only one person's half of a conversation24,25.
Following this analogy further, much of what is written in journals
can be considered "chatter" among researchers with the practitioner
listening in23,25. Just as bits
and pieces of chatter overheard in a public arena can be dangerous
if taken out of context, medical research chatter can be misapplied
to clinical situations and can become as harmful and inappropriate
as gossip. Most people, orthopaedists included, enjoy listening
to chatter and gossip and argue that it is useful to keep abreast
of the lay public and to recognize future trends. But consider some
of the false starts in clinical practice when medical gossip was
translated into action: carbon-fiber polyethylene, fluoride treatment
of osteoporosis, and the injection of chymopapain into the lumbar
intervertebral disc. In these examples, clinical practice was changed
on the basis of preliminary or intermediate evidence that was not
subsequently supported by ongoing scientific clinical evaluation.
On the other hand, orthopaedists chastened by their experiences
of changing their treatment (going along with the fad) might justifiably
develop a cynical mistrust of journals. Our journals remain the
primary medium for communicating new information but, unfortunately,
not necessarily for providing knowledge. In the above examples of
intermediate outcomes leading to treatments that did not prove successful,
the facts were laid out on the table for all to see but they were
left up to the individual reader to interpret. The array of medical
facts (disease-oriented evidence) is often an intermediary for knowledge,
and many times the facts alone do not yield a successful treatment.
The acquisition of knowledge from information is a daunting task
for which few clinicians are adequately prepared. Most general orthopaedists
refer to many sources for quick, reliable knowledge: textbooks,
discussions with colleagues, newsletters, review articles, consultations
with subspecialists, equipment-company representatives or brochures,
and instructional meetings. The benefit of these knowledge sources
is apparent: they are quick and easy to use, and the knowledge is
often immediately applicable. However, each source of orthopaedic knowledge
has limitations. Colleagues who are no better informed than you
may feel pressured to give an answer or may be biased by their experience
or self-interest17,18,21. A textbook
may be outdated by the time that it is published. Industry representatives
are obviously biased by profit interests. Newsletters and review
articles may give only one side of an argument or may fail to put
new information in context with the old7.
How, then, do we find a compromise between relying on other sources
and converting information into knowledge for ourselves? There is
a need for a reliable approach to the practical and time-efficient
management of information.
When a busy orthopaedist picks up a journal, calls a colleague,
or attends a conference, the goal is to spend the least amount of
time and effort to find the best information. Some but not most
of what is acquired in this process is knowledge. Information that
is useful has three attributes7,8,13,23:
(1) it must be relevant to the practitioner's needs, (2) it must
be valid, and (3) it must require little work to obtain. These three
factors can be expressed in the following usefulness equation23: usefulness of orthopaedic information
= (relevance ¥ validity)/work.
The relevance component of the equation starts with the notion
of applicability to one's practice and extends into the domain of
patient outcomes. We need patient-oriented evidence with which we
can evaluate the efficacy of interventions in terms of results that
patients care about and that we as clinicians want for our patients. This
has been designated Patient-Oriented Evidence that Matters (POEM)23 as opposed to Patient-Oriented Evidence
(POE) that does not have a relevance measure or Disease-Oriented
Evidence (DOE). POEMs are rare and are scattered among the huge
number of DOEs.
DOEs consist of information aimed at increasing our understanding
of a disease: its etiology, pathophysiology, prevalence, prognosis,
and so on. These studies are important in all of medicine and in
orthopaedics in particular. We must understand how a disease works
before we can routinely diagnose, treat, or prevent it with any certainty.
A study that reports the prevalence of redislocation, wound infection,
and adhesions in a consecutive series of patients treated with arthroscopic
thermal plication is a DOE. In contrast, a randomized trial that
compares arthroscopic thermal plication with arthrotomy by using
a valid instrument to measure the quality of life of throwing athletes
with a dislocating shoulder is a POEM. For the most part, POEMs consist
of ready-to-use knowledge and DOEs consist of available information
of uncertain clinical utility. More importantly, a POEM tells us
what our patients want to know and would like us to fulfill as a
diagnostic or treatment goal. A physician will change his or her
practice to incorporate the evidence (the POEM) in order to improve
patient outcomes.
Until recently, DOEs were the only sources of information that
we had about many illnesses and treatments. We knew, for example,
that sodium fluoride increased the apparent density of bone. Yet
when it was used to prevent vertebral fractures due to postmenopausal
osteoporosis, the rate of vertebral fracture actually increased. This
paradoxical finding is explained by the brittle quality of the new
bone that forms with sodium fluoride treatment. This is an example
of how a surrogate outcome (increased bone density) did not predict
the clinical outcome (prevention of vertebral compression fractures).
The application of DOEs to orthopaedic practice requires assumptions
that translate information about a general phenomenon to your patient. (For
example, joint laxity produces pain; therefore, if lax joints are
surgically tightened, the pain will resolve.) The application of
POEMs requires only that the condition of your patient matches that
of the patients in the study. (For example, my patient has a ruptured
anterior cruciate ligament, and if it is properly repaired she will
most likely be able to return to athletic activities.)
POEMs exist for diagnosis and prevention as well as treatment.
Diagnostic POEMs are particularly helpful in applications of new
technology, such as polymerase chain-reaction evaluation of joint
fluid for suspected staphylococcal infection.
POEMs involving highly prevalent conditions have the greatest
impact on our patients and therefore the greatest relevance8,23,24. In contrast, DOEs involving
rare or unusual disorders have the least relevance. Evidence that a
medication can prevent progression of stage-I avascular necrosis
of the femoral head is a POEM for a common condition, while a chemical
assay for p53 messenger RNA expression in the diagnosis of clear-cell
sarcoma is a DOE for a very rare condition (Fig. 2).
The validity of information defines to what extent the knowledge
gained as a result represents "the truth." Well designed clinical
trials based on statistical and experimental plans minimize bias
and are most likely to provide valid conclusions20.
Such clinical trials are lengthy, expensive, and fraught with potential
weaknesses. The assessment of validity is also difficult and time-consuming
for clinicians with inadequate training in statistics or epidemiology.
Validity assessment can be done individually, can be done with other orthopaedists,
or can be delegated, with great care, to a nonorthopaedist who has
the appropriate training (such as a biomedical statistician) and
the available time. It is not enough to ask a respected subspecialist
or academic orthopaedist whether an article is good.
There is no evidence that subspecialty surgeons are better at
validating new information than are general orthopaedists. The best
strategy is to consult a colleague with expertise in statistics, attend
orthopaedic journal clubs that spread the burden of finding and
evaluating new information, or search for published rigorous evaluations4,7,21,25 such as those presented in
the new Evidence-Based Orthopaedics section in this issue of The
Journal (see page 873). Review articles, even in highly respected
orthopaedic journals, do not provide rapid critical evaluation of
evidence-oriented publications.
The effort that it takes to obtain and validate useful information
is the work that is the denominator in the usefulness equation.
Work consists of factors such as the time and expense necessary to
obtain information and the mental effort required to ensure its
relevance and validity. Too much time, energy, or expense will increase
the work factor and decrease the usefulness of the information.
In some instances, the information is essential and the investment
in time and other resources is mandated by the need for an answer. For
the busy orthopaedist, a high usefulness score should fulfill most
information needs. The best sources of information are those that
provide highly relevant and valid information with minimal work.
Studies evaluating the impact of medical research have shown
that well designed clinical trials that should influence medical
care fail to do so because the results are diminished by the surplus
of clinically unimportant information8,22.
Clinicians, frequently overwhelmed by the volume of DOE literature,
fail to identify the high-quality POEM information that is available8,12. Incorporating POEM information
into clinical practice is in the best interest of patients because POEMs
provide the relevant answers to patient problems.
Understanding the relationship between relevance, validity, and
work helps us to master the management of medical information. Distinguishing
between POEMs and DOEs will minimize the potential for misapplying
harmful medical gossip22,23. In
addition, focusing on the identification of POEMs will reduce the
time necessary to remain up to date within all areas of orthopaedics.
Allowing oneself the luxury to skim over DOE information does wonders
for improving self-esteem and increasing free time without creating a
guilty conscience. One major caution: this approach to medical information
management creates a double-edged sword23,24,27.
One edge empowers clinicians to skim most of the published literature.
The other edge gives clinicians the responsibility to seek, evaluate, and,
most importantly, implement the new knowledge that improves patient
care (POEMs).
Now, what do you tell a tournament tennis player who asks for
cloned cartilage resurfacing of a defect in the humeral head of
her dominant shoulder? There is no published patient-oriented evidence
that supports this approach in the shoulders of athletes, and this
treatment is considered experimental. Preliminary experience by
some pioneering surgeons may exist, but there is no published data.
A less radical treatment would be to temporize until better POEM information
becomes available. Alternatively, if the patient is insistent after
both of you weigh the available (DOE) evidence, she can become your
first patient to receive a cartilage implant in the humeral head
and she can be part of a collaborative effort to generate POEM data
for this treatment. Whatever the decision, you should remain open
to the necessity of changing directions if POEM information becomes
available.
Deciding to implement a new procedure has a significant impact
not only on the quality of medical care but also on its cost. General
orthopaedists on the front lines and in the trenches of the battle
against musculoskeletal disease recognize the power derived from
the appropriate management of information. As summarized by Eddy9: "In a field filled with uncertainty
and doubt, the difference between `when in doubt, do it' and `when
in doubt, stop' could easily swing $100 billion a year."
The goal of orthopaedic information mastery is to complement
an evidence-based practice of orthopaedics: musculoskeletal evaluation
and management, operative and nonoperative treatment, and hospital
and community-based treatment. We support a paradigm shift toward
a more scientific, modern, and (hopefully) cost-effective approach
to improving patient outcomes based on useful knowledge from reputable
sources in orthopaedic journals. Funding for lengthy and expensive
clinical trials that generate POEMs is more likely if the orthopaedic community
demonstrates the need for POEMs to those funding sources. The "quality"
orthopaedic journals will demonstrate the need for POEMs by publishing
more evidence-based material. The "quality time" that we spend "digesting"
POEMs, as suggested by Sir Francis Bacon, will improve orthopaedic
outcomes20 and save us valuable
time.
Shepard R. Hurwitz, M.D.
David Slawson, M.D.
Department of Orthopaedics (S. R. H.), Box 159Department of Family
Medicine (D. S.)University of Virginia School of MedicineCharlottesville,
Virginia 22908E-mail address for S. R. Hurwitz: srh5u@virginia.edu
Allen Shaughnessy, Pharm.D.
Harrisburg Family Practice Residency ProgramHarrisburg Hospital205
Front Street, P.O. Box 8700Harrisburg, Pennsylvania 17105-8700
Arndt, K. A.: Information excess in medicine. Overview, relevance to
dermatology, and strategies for coping. Arch. Dermatol.,128: 1249-1956, 1992.1281249
1992
[PubMed]
Becker, M. H.: Sociometric location and innovativeness: reformulation
and extension of the diffusion model. Am. Sociol. Rev.,35: 267-283, 1970.35267
1970
Becker, M. H.: Factors affecting diffusion of innovations among health
professionals. Am. J. Pub. Health,60: 294-304, 1970.60294
1970
[PubMed]
Bucknall, C. A.; Morris, G. K.; and Mitchell, J. R.: Physicians' attitudes to four common problems: hypertension,
atrial fibrillation, transient ischaemic attacks, and angina pectoris. British Med. J.,293: 739-742, 1986.293739
1986
Coleman, J. S.; Katz, E.; and Menzel,
H.: Medical Innovation: A Diffusion Study. Indianapolis, Bobbs-Merrill,
1966.
Connelly, D. P.; Rich, E. C.; Curley, S. P.; and Kelly, J. T.: Knowledge resource preferences of family physicians. J. Fam. Pract.,30: 353-359, 1990.30353
1990
[PubMed]
Curley, S. P.; Connelly, D. P.; and Rich, E. C.: Physicians' use of medical knowledge resources: preliminary
theoretical framework and findings. Med. Decis. Making,10: 231-241, 1990.10231
1990
[PubMed]
Ebell, M. H.; Barry, H. C.; Slawson, D. C.; and Shaughnessy, A. F.: Finding POEMs in the medical literature. J. Fam. Pract.,48: 350-355, 1999.48350
1999
[PubMed]
Eddy, D. M.: Clinical decision making: from theory to practice. Three
battles to watch in the 1990s. J. Am. Med. Assn.,270: 520-526, 1993.270520
1993
Evans, C. E.; Haynes, R. B.; Gilbert, J. R.; Taylor, D. W.; Sackett, D. L.; and Johnston, M.: Educational package on hypertension for primary care physicians. Canadian Med. Assn. J.,130: 719-722, 1984.130719
1984
Fineberg, H. V.: Clinical evaluation: how does it influence medical practice?. Bull. Cancer,74: 333-346, 1987.74333
1987
[PubMed]
Haynes, R. B.; Sackett, D. L.; Guyatt, G. H.; Cook, D. J.; and Gray, J. A.: Transferring evidence from research into practice: 4.
Overcoming barriers to application [editorial]. ACP J. Club,126: 14-A15, 1997.12614
1997
Kitchens, J. M., and Pfeifer, M. P.: Teaching residents to read the medical literature: a controlled
trial of a curriculum in critical appraisal/clinical epidemiology. J. Gen. Intern. Med.,4: 384-387, 1989.4384
1989
[PubMed]
Knott, J., and Wildavsky, A.: If dissemination is the solution, what is the problem?. Knowledge: Creation, Diffusion, Utilization,1: 537-578, 1980.1537
1980
LaPorte, R. E.: Global public health and the information superhighway. British Med. J.,308: 1651-1652, 1994.3081651
1994
Linzer, M.; Brown, J. T.; Frazier, L. M.; DeLong, E. R; and Siegel, W. C.: Impact of a medical journal club on house-staff reading
habits, knowledge, and critical appraisal skills. A randomized control
trial. . Am. Med. Assn.,260: 2537-2541, 1988.2602537
1988
Mann, K. V., and Chaytor, K. M.: Help: is anyone listening? An assessment of learning needs
of practicing physicians. Acad. Med.,67 (Supplement 10): 4-S6, 1992.67 (Supplement 10)4
1992
Menzel, H.: Sociological perspectives
on the information-gathering practices of the scientific investigator
and the medical practitioner. In Bibliotheca Medica: Physician for
Tomorrow, pp. 127-128. Edited by D. McCord. Boston, Harvard Medical
School, 1966.
Oxman, A. D.; Sackett, D. L.; and Guyatt, G. H.: User's guides to the medical literature. I. How to get
started. The Evidence-Based Medicine Working Group. J. Am. Med. Assn.,270: 2093-2095, 1993.2702093
1993
Sackett, D. L.: Evidence-based medicine [editorial]. Spine,23: 1085-1086, 1998.231085
1998
[PubMed]
Shaughnessy, A. F.; Slawson, D. C.; and Bennett, J. H.: Becoming an information master: a guidebook to the medical
information jungle. J. Fam. Pract.,39: 489-499, 1994.39489
1994
[PubMed]
Sheldon, T. A.; Guyatt, G. H.; and Haines, A.: Getting research findings into practice. When to act on
the evidence. British Med. J.,317: 139-142, 1998.317139
1998
Slawson, D. C.; Shaughnessy, A. F.; and Bennett, J. H.: Becoming a medical information master: feeling good about
not knowing everything. J. Fam. Pract.,38: 505-513, 1991.38505
1991
Slawson, D. C.,, and Shaughnessy, A. F.: Assessing medical information. Hosp. Pract. (Office Ed.),31: 103-104, 1996.31103
1996
Slawson, D. C., and Shaughnessy, A. F.:: Teaching information mastery: creating informed consumers
of medical information. J. Am. Board Fam. Pract.,12: 444-449, 1999.12444
1999
[PubMed]
Smith, R.: What clinical information do doctors need?. British Med. J.,313: 1062-1068, 1996.3131062
1996
Stross, J. K.,, and Harlan, W. R.: The dissemination of new medical information. J. Am. Med. Assn.,241: 2622-2624, 1979.2412622
1979
Wyatt, J.: Use and sources of medical knowledge. Lancet,338: 1368-1373, 1991.3381368
1991
[PubMed]