THE ADOPTION of a new clinical behaviour by a clinician and healthcare system

ADOPTING BEST EVIDENCE IN PRACTICESUPPLEMENTDiffusion of innovation theory for clinical changeRobert W Sanson-FisherT HE ADOPTION of a new clinical behaviour by a clinicianand healthcare system is a consequence of multiple factors,with research evidence being only one. Research on thediffusion or adoption of innovations suggests that a numberof themes come into play. For example, the willingness touse new drugs is influenced by the physician’s sex, specialty,medical school, Journal since graduation, 0025-729X locationThe Medical years of Australia ISSN: practice 15and practice volume,55-56 the relative proportion of elderlyMarch 2004 180 6 andpatients inMedical Journal ofpractice.1©The the physician’s Australia 2004 www.mja.com.auAdopting Best Evidence in PracticeABSTRACT■Maximising the adoption of evidence-based practice hasbeen argued to be a major factor in determining healthcareoutcomes. However, there are gaps between evidencebased recommendations and current care.■Bridging the evidence gap will not be achieved simply byinforming clinicians about the evidence.■One theoretical approach to understanding how change maybe achieved is Rogers’ diffusion model. He argues thatcertain characteristics of the innovation itself may facilitateits adoption. Other factors infuencing acceptance includepromotion by influential role models, the degree ofcomplexity of the change, compatibility with existing valuesand needs, and the ability to test and modify the newprocedure before adopting it.■The diffusion model may provide valuable insights into whysome practices change and others do not, as well as guidingthose who try to effect adoption of best-evidence practice.Diffusion theoryRogers2 has developed one of the better-known theoreticalapproaches to diffusion of innovation. This theoreticalframework is helpful when determining the adoption ofspecific clinical behaviours and when deciding which components will require additional effort if diffusion is to occur.It includes a consideration of aspects of the innovation (ornew technology), style of communication, steps in decisionmaking, and the social context.MJA 2004; 180: S55–S56The innovationAccording to Rogers2 there are five elements of a new orsubstitute clinical behaviour that will each partly determinewhether adoption or diffusion of a new activity will occur:relative advantage, compatibility, complexity, trialability andobservability.Relative advantage: Rogers defines “relative advantage” asthe degree to which an innovation is perceived as better thanthe idea it supersedes.Research provides information on the cost-effectivenessand potential benefit to patients of implementing a newclinical activity. However, the objective data may be lessimportant than the clinician’s perception of whether theinnovation will be advantageous. Decisions about implementing best-evidence practice are driven not only bypatient welfare but also by the interplay between the interests of the patient, the clinician and the healthcare system.For example, if a proposed change alters the balance ofpower between or within professional groups in a “negative”way, the innovation may not be implemented. Conversely, ifthe recommended behaviour increases the status of adoptingclinicians and brings in more revenue for individuals or theorganisation, the innovation may be readily adopted.Compatibility: “Compatibility” is a measure of the degree towhich an innovation is perceived as being compatible withFaculty of Health, University of Newcastle, Newcastle, NSW.Robert W Sanson-Fisher, PhD, Professor of Health Behaviour.Reprints will not be available from the author. Correspondence: ProfessorRobert W Sanson-Fisher, Faculty of Health, University of Newcastle, DavidMaddison Building, King Street, Newcastle, NSW 2300. Rob.sansonFisher@newcastle.edu.auMJAVol 18015 March 2004existing values, past experiences, and the needs of potentialadopters.To increase the probability of adoption, the innovationmust address an issue that clinicians or others perceive to be aproblem. For example, a new procedure that enables earlydetection of a life-threatening illness is likely to be adopted.Early screening tests are compatible with medical beliefs thatearly detection of disease is beneficial. Consequently, testsand procedures that appear to offer this capacity are morelikely to be adopted. Real-life examples include the rapidadoption of mammography screening3,4 and testing for prostate cancer,5 despite some debate about their effectiveness.Complexity: “Complexity” is a measure of the degree towhich an innovation is perceived as difficult to understandand use. A clinical procedure is more likely to be adopted ifit is simple and well defined. For example, altering apatient’s drug regimen is relatively simple, and thus changesin drug therapy can occur rapidly. In contrast, preventiveactivities such as detecting and treating patients with hazardous alcohol consumption6 and smoking7 have not beenadopted quickly, in spite of the potential health gain. Thismay, at least in part, be a result of the complexity of theseactivities. Attempts to intervene at the level of primaryprevention may be hampered by patients’ resistance andtheir lack of accuracy in self-reporting risk behaviours.Moreover, some clinicians may have insufficient expertise inthe consulting skills necessary to achieve change.Trialability: Rogers defines “trialability” as the degree towhich the innovation may be trialled and modified. TheS55ADOPTING BEST EVIDENCE IN PRACTICESUPPLEMENTability to test a potential medical intervention on a limitedbasis allows clinicians to explore the implementation of theprocedure, its acceptability to patients, and the potentialoutcomes. Rogers argues that the facility to undertake alimited cost–benefit trial of an intervention promotes faiththat the evidence is correct and that its implementation islogistically possible.■the clinician seeks reinforcement about the innovationdecision (eg, discussion and comparison with peers).Individuals and organisations will move through the decision process at different rates, depending upon whether theyare innovators, or early or late adopters. Innovators arecharacterised by their tolerance of high levels of uncertainty.Observability: “Observability” is the degree to which theThe social contextresults of the innovation are visible to others. “Visibility” ofan innovation stimulates peer discussion, as colleagues of aclinician adopting a new procedure often request information about it. If respected and influential clinicians argue forand demonstrate the application of a new procedure ortreatment approach, it is likely to have a positive impactupon adoption rates.8 The more charismatic the personproviding the role model, the greater the chance that agreater number of other professionals will adopt the advocated change in clinical behaviour. In surgery, new techniques are often adopted very quickly, as there is a commonperception that there are disadvantages in being “leftbehind” by not adopting new technology.9The systems most likely to respond easily and quickly toinnovation are ones that have a culture of creativity andinnovation, a relatively flat hierarchical system, and strongleadership that is committed to effecting change. In contrast, the healthcare system has a hierarchical model, withseparate organisational structures for each professionalgroup. The system is often bureaucratic, with social normsthat hinder rapid change. However, within this system, it ispossible for clinicians to change some aspects of theirclinical activities relatively rapidly, as there are few restraintson determining the choice of care. Changing to evidencebased clinical behaviour may mean modifying the system sothat it adequately monitors the frequency of the activity, theoutcomes, feedback to the clinician and contingencies fordesired clinical behaviour.Communication styleChannels of communication used to convey informationabout clinical practice include research publications, databases (eg, the Cochrane database), the mass media, attendance at lectures an
d workshops, visits from interest groups,and videos or audiotapes.Current research suggests that the most effective communication strategy is face-to-face exchange.8 It provides anopportunity to tailor information to recipients and allowsthe advocate of the change to explore and, if necessary,modify the reasons why a shift in clinical behaviour shouldoccur. Interpersonal communication is usually more effective when there is a high degree of professional resemblancebetween the individual attempting to introduce the innovation and the recipient. This may partly explain why clinicalaudits undertaken by medical practitioners are more likelyto lead to adoption of a new practice than those performedby allied health staff.8The decision processThe diffusion model2 postulates five steps in the decisionmaking process:■ researchers acquire knowledge about the proposed clinical change;■ the individual clinician is persuaded about the advantagesof the innovation;■ the clinician engages in activities that will lead to a choiceabout adopting or rejecting the innovation (eg, reading,attending workshops, communicating with individualswho have experience in the field);■ the innovation is incorporated into the daily activity ofthe clinician; andS56SummaryDiffusion theory offers a plausible explanation for why someclinical activities are adopted rapidly and others only withdifficulty, despite strong evidence of their potential benefits.Some clinical behaviours may be adopted relatively easilybecause of the nature of the behaviour itself, while others mayinvolve a complex interplay between social systems, communication style and the decision-making process. There is aneed to prospectively test the assumptions of the model in thehealthcare environment using rigorous experimental design.Competing interestsThe author is a Senior Advisor to the National Institute of Clinical Studies.References1. Tamblyn R, McLeod P, Hanley JA, et al. Physician and practice characteristicsassociated with the early utilization of new prescription drugs. Med Care 2003;41: 895-908.2. Rogers E. Diffusion of innovations. New York: Free Press, 1983.3. Goodman SN. The mammography dilemma: a crisis for evidence-based medicine? Ann Intern Med 2002; 13: 363-365.4. Stanley DE. The mammography dilemma. Ann Intern Med 2003; 138: 771.5. McDougall GJ Jr, Weber BA, Dziuk TW, Heneghan R. The controversy of prostatescreening. Geriatr Nurs 2000; 21: 245-248.6. Schorling JB, Klas PT, Willems JP, Everett AS. Addressing alcohol use amongprimary care patients: differences between family medicine and internal medicine residents. J Gen Intern Med 1994; 9: 248-254.7. Franzgrote M, Ellen JM, Millstein SG, Irwin CE Jr. Screening for adolescentsmoking among primary care physicians in California. Am J Public Health 1997;87: 1341-1345.8. Bero LA, Grilli R, Grimshaw JM, et al. Closing the gap between research andpractice: an overview of systematic reviews of interventions to promote theimplementation of research findings. BMJ 1998; 317: 465-468.9. Denis J-L, Hebert Y, Langley A, et al. Explaining diffusion patterns for complexhealth care innovations. Health Care Manage Rev 2002; 27: 60-73.❏MJAVol 18015 March 2004