Using Diffusion of Innovations and Academic Detailing to Spread Evidence-based Practices

Clark Carboneau

Healthcare in the United States is arguably the best in the world (Berwick, 2003). The science base for healthcare is broad and deep, with advancements progressing at a steady, if not hurried, pace. Yet an enormous amount of scientific knowledge related to healthcare, that is, evidence-based best practices, is underutilized across the industry. For example, when an evidence-based innovation is implemented successfully in one part of a hospital or clinic, it may spread slowly, or not at all, to other parts of the organization (Berwick). This article explains the theory and research associated with how change spreads across human systems, as described in E. Rogers’s Diffusion of Innovations and provides a communication strategy, academic detailing, for applying the theory to healthcare.

The Centers for Medicare and Medicaid Services (CMS) Change Concepts

CMS is working to increase the adoption rate of evidence-based best practices in specific clinical areas. If adoption of these innovations relied solely on the scientific research that supported its use, these innovations would be quickly adopted. However, some people naturally resist accepting even potentially constructive change. One example is the evidence-based practice of not shaving patients with razors prior to surgery, which decreases the risk of infection. The benefit of this process change is not new. Research on this topic dates back to 1971, with 561 surgery cases studied by Seropian (Rupp, 2003). More research was performed in 1973, when Cruse examined and calculated the infection rate of 23,649 surgery cases (Bratzler, 2004). The infection rate for patients shaved with razors was 2.3%; in comparison, the infection rate for patients with hair removed by clippers was 1.7% (Bratzler). Subsequent research by Alexander, Sellick, and Ko in the 1980s and 1990s confirmed these earlier studies (Rupp). The evidence has been validated for more than 30 years, yet some surgery suites in the United States still shave patients prior to surgery and thus increase the risk of infection. To do better, healthcare professionals must acknowledge and adopt evidence-based best-care practices sooner.

The History of Diffusion of Innovations Research

The branch of research concerned with how change is adopted is called the diffusion of innovations, and its history dates back 60 years. In 1943 researchers at Iowa State University wanted to understand the slow adoption rate by local commercial farmers of a new drought-resistant hybrid seed corn. The university promoted the use of this seed corn during the Great Depression in the late 1920s and throughout the Dust Bowl years of the 1930s. In addition to being drought resistant, the seed corn had a proven 20% increase in yield. Because of these benefits, it was expected that the farmers would be thrilled about the new seed corn and would quickly adopt its use.

Iowa State University hired Ryan and Gross to research the slow adoption rate. The researchers conducted one-on-one interviews with 259 farmers, and the importance of the farmers’ network was viewed as the primary reason for adoption. That is, the farmers placed more value on their friends’ and neighbors’ opinions about the seed corn than on the opinions of the university experts or the salespeople (Rogers, 2003). From this initial research, an S-shaped curve was developed that described the cumulative total number of adopters. The S curve demonstrated that adopting change usually proceeds slowly until 20% of the population has adopted the innovation. That percentage is called the tipping point, or the critical mass. After the tipping point is reached, the adoption rate accelerates quickly without any further intervention (Gladwell, 2002). The research also concluded that the frequency of adopters per year followed a normal bell-shaped distribution (see Figure 1).

Categorizing Adopters

Rogers combined the Iowa farmer survey data with the mean and standard deviations of the normal curve adoption rate (see Figure 1) to develop categories of adopters and their personality traits (Rogers, 2003). The social sciences, marketing, public health, agriculture, education, and other industries have successfully planned and implemented the spread of innovations through the use of this model (see Figure 2).

Description of the Adopter Categories

Five adopter types have been identified (Rogers, 2003).

• Innovators are 2.5% of the social population. They are venturesome, and they come up with ideas for improvement.
• Early adopters are 13.5% of the social population. They are respected risk takers and are the highest opinion leaders in the system. The early adopters are socially integrated (i.e., they communicate freely with the innovators and the early majority). The innovators create the idea, and the early adopters test it for possible adoption.
• Early majority are 34% of the social population. They are deliberate and the most interconnected people of the social system. The early majority adopts change after the early adopters demonstrate that the change can work.
• Late majority are 34% of the social population. They are skeptical about change but are subject to peer and economic pressure. They need a weight of evidence before they will adopt change. The late majority adopts the change after the early majority has clearly demonstrated that the change is a better way. This group will change if they can see that the change is good.
• Laggards are 16% of the social population. They are the resisters. They won’t change unless everyone around them has changed. Being risk-averse sometimes has negative consequences. In the hybrid seed corn study, some laggards lost their farms instead of adopting the new innovative seed corn.

Advancing Diffusion with One-on-One Education

Trying to spread an innovation from one care setting within a facility to other care settings requires a plan. One strategy is to script a messenger with evidence-based information and perform one-on-one education with the early adopter. Identifying the early adopter need not be difficult. All one needs to do is seek out the respected opinion leader of the work system. Early adopters will have a track record of innovativeness and of challenging “the way we’ve always done it” (Collins, 2000).

Targeting the early adopter for one-on-one education, a process known as academic detailing, has proven effective for changing behavior (Solomon et al., 2001). Based on the diffusion of innovations research, the goal should be to get 20% of the target audience to use the evidence-based method. From a psychological perspective, a 20% goal (e.g., 1 out of 5 practitioners) seems achievable, as opposed to a goal to change the behavior of 100% of the target audience. After the tipping point is reached, others within the system can be expected to adopt the method on their own. The early majority will observe the success of the early adopters and become motivated to change. In contrast, trying to convince the laggards to lead the change may not be the most efficient use of one’s time. This group traditionally avoids change and defends the status quo. Because a process worked in the past, the laggards will defend it. The laggards will eventually change but not until everyone around them has changed.

To gain the needed 20% of supporters, the early adopters must agree to test the change concept, and, if successful, they must promote the innovation to their peer network. Since 1991 at least seven randomized clinical trials have used an opinion leader and his or her network to change behavior. The control groups used patient chart audits with feedback to physicians on patient outcomes. In all these studies the use of an opinion leader and his or her network has proven successful. In contrast, the control groups did not make any significant change in behavior (Rogers, 2003).

Application to Healthcare—An Example of a CMS Change Concept

To implement the CMS change concept “prophylactic antibiotic received within 1 hour prior to surgical incision” (antibiotic given within 1 hour prior to surgery) requires physician buy-in and support. To gain this support requires a physician (an early adopter) to agree to test the preferred method and then to promote the new method to his or her peers if the results are agreeable. Advocating the need for change to other physicians was illustrated by the hybrid seed corn research. The farmers who adopted the hybrid seed corn were influenced more by their friends’ and neighbors’ endorsement than by that of the university experts.

This one-on-one communication or education strategy using the early adopter frequently is used successfully by pharmaceutical companies (Solomon et al., 2001). The premise is to script the message, so that all the important details can be conveyed within a short time frame (e.g., within 5 minutes). This method would use a conversation in an elevator or hallway rather than a formal presentation at a medical staff meeting.

This version of academic detailing education consists of five components (Pfizer Pharma-ceutical, 2003):

• State the promise.
• Explain the features.
• Describe the benefits.
• “Trial the close” and overcome objections.
• Close the dialogue and get a commitment for next steps.

Step 1. State the promise.

The promise must be an attention getter, and it has to outweigh the “pain of change.” Being able to prove your promise with evidence will relieve the discomfort that the thought of change creates. Be attentive to your audience members by studying their body language. If they do not seem to be listening, they may be visual learners. Try using a graph with comparative data to catch their attention.

This is the promise for our “antibiotic given within 1 hour prior to surgery” example:

• The surgical site infection rate will be reduced by making a few simple changes.

Step 2. Explain the features.

Features show how good something is. Features are the facts that can be proven with evidence. Features clearly describe why the evidence-based best practice should be used (e.g., better patient outcomes, more efficient processes, saved patient and physician time). As a rule of thumb, when presenting the features, try to be as objective as possible. If controversial issues surround the new method, it is better to briefly acknowledge both sides of the issues. If you are not perceived as being objective, you may lose credibility and risk losing a receptive audience.

Features for the “antibiotic given within 1 hour prior to surgery” example based on national 1999 numbers (Bratzler, 2004, Slide 3) are these:

• Average mortality rate from infections is 7.8%; noninfection mortality rate is 3.5%.
• Average length of stay from infections is 11 days; length of stay for patients with no infection is 6 days.
• Average readmission rate from infections is 41%; readmission rate for patients with no infection is 7%.
• An estimated 40%–60% of surgical site infections can be prevented.

Step 3. Describe the benefits.

Features show how good something is, but benefits sell its value. This is where the change concept is actually sold. If possible, the benefits should span more than one area; it would be useful, for example, to describe the benefit to the patients, the physicians, and the health system.

The benefits for the “antibiotic given within 1 hour prior to surgery” example are these:

• Patients will be healthier.
• Physicians’ infection rates will decrease.

Step 4. “Trial the close” and overcome objections.

“Trialing the close” refers to holding an engaging dialogue with the physician in order to gauge the success of convincing the physician to try the evidence-based best practice. Ask the physician, “What questions do you have about this information?” or “Do you agree with the evidence?” The response shows how receptive the physician is to the change concept. If the physician is skeptical, try saying, “There must be a reason you feel this way” or “What would change your position on this?” The purpose of holding this conversation and overcoming any objections is to start building some common ground with the physician. Clarifying an understanding of the physician’s concern is necessary. After paraphrasing the concern, ask, “Do I really understand your concern?”

Trialing the close and overcoming objections in the “antibiotic given within 1 hour prior to surgery” example would involve sharing the surgical infection prevention literature with the physician and then asking these two questions:

• “Does this evidence address your concerns?”
• “Are there any other issues you have concerning this evidence?”

Step 5. Close the dialogue and get a commitment for next steps.

Close the dialogue by repeating the key benefits and restating the areas of agreement. The statement must be specific, and a desired action must be requested if you are to gain the person’s commitment. For example, ask, “May I leave this surgery infection prevention material with you?” or “May we discuss this evidence again in a few days?”

The commitment in the “antibiotic given within 1 hour prior to surgery” example would be a statement like the following:

• The physician has agreed to review the evidence and to meet next Tuesday in the café to discuss it further.

Conclusion

Using diffusion of innovations theory with academic detailing can accelerate the spread of evidence-based best practices in healthcare. Testing the change concept will enable the early adopters to see for themselves that the evidence-based method works. Observing the early adopters using the new practice and hearing of its effectiveness motivates the early majority to change. The tipping point is reached when the early adopters and the early majority change their practice to match the change concept. The late majority will be the next to adopt the change, followed by the laggards, and the evidence-based practice will eventually become the new status quo.

References
Berwick, D. M. (2003) Disseminating innovations in health care. JAMA: The Journal of the American Medical Association, 289(15), 1969–1975.

Bratzler, D. (2004) A national initiative to improve care for Medicare patients. Kearney, NE: Surgical Infection Prevention Collaborative.

Collins, B., Hawks, J., & Davis, R. (2000). From theory to practice: Identifying authentic opinion leaders to improve care. Managed Care, 9(7), 56–62.

Gladwell, M. (2002). The tipping point. Boston: Little Brown.

Pfizer Pharmaceutical. (2003). Using academic detailing to influence behavior. Omaha, NE: Author.

Rogers, E. (2003). Diffusion of innovations. New York: Free Press.

Rupp, M. (2003). Prevention of surgical site infections. Lincoln, NE: Surgical Infection Prevention Collabor-a-tive.

Solomon, D., Van Houten, L., Glynn, R., Baden, L., Curtis, K., Schrager, H., et al. (2001). Academic detailing to improve use of broad spectrum antibiotics at an academic medical center. Archives of Internal Medicine, 161(15), 1897–1902.

Disclaimer: The analyses upon which this publication is based were performed under Contract Number 500-02-NE03, entitled “Utilization and Quality Control Peer Review Organization for the State (Commonwealth) of the State of Nebraska,” sponsored by the Centers for Medicare & Medicaid Services, U.S. Department of Health and Human Services (DHHS). The content of this publication does not necessarily reflect the views or policies of the DHHS, nor does mention of trade names, commercial products, or organizations imply endorsement by the U.S. Government. The author assumes full responsibility for the accuracy and completeness of the ideas presented. This article is a direct result of the Health Care Quality Improve-ment Program initiated by the Centers for Medicare & Medicaid Services, which has encouraged identification of quality improvement projects derived from analysis of patterns of care, and therefore required no special funding on the part of this contractor. Feedback to the author concerning the issues presented is welcomed. (750W-NE-QI-07-01/2005)

Author’s Biography
Clark Carboneau, MBA CQM, industrial engineer, earned his Deming Scholars MBA degree in management systems from Fordham University. He has 18 years’ experience in applying quality management tools and methods with over 10 years’ experience in healthcare as a director of quality improvement.

For more information on this article, contact Clark Carboneau by phone at 800/458-4262, ext. 517, or by e-mail at ccarboneau@neqio.sdps.org.