What Pharma Can Learn From Other Industries
By Clara Heering, VP, clinical risk management, ICON plc
The auto industry is a different animal today than in 1910 or even 1975. Just as the manufacturing theories of Henry Ford and Taiichi Ohno transformed automakers, inspirational methodologies and philosophies have evolved the operations of many organizations in the past few decades.
Airlines, for example, have cut the number of crashes in half through a risk reduction methodology called human factors analysis (HFA). The British Royal Navy and many countries’ armed forces have improved battlefield decision-making and success through a “mission command” philosophy that enlightens troops to the “why” behind their orders, which allows troops to adjust effectively as field conditions change.
Despite inspired changes in many industries over the last century, clinical trials today remain structurally unchanged since the 1950s. As we look to overcome rising costs and steep failure rates, we may profit from re-examining some of the ideas behind some industrial evolutions. Inspirational examples abound, but a few — namely in risk mitigation and leadership communication strategies — are particularly worth revisiting because some drug developers are already drawing upon them to evolve their clinical development process.
Mitigating And Reducing RiskLessons From The Airline Industry
Airlines have large workforces with varying backgrounds and performance goals depending on job function. This combination can create gaps in communication that are often the source of risks to the operation, such as mid-air collisions and flight delays. The airline industry has dramatically improved safety records and on-time departures by adopting HFA, a framework for empirically tracing errors’ root causes and mitigating those underlying sources of risk, which may include training deficits, technology issues, or culture and leadership factors. As not all errors emerge from immediately apparent reasons, the systematic analysis involved in HFA provides insights into complex operations that can be otherwise difficult to discern.
Drug developers can use HFA to minimize and mitigate trial risks that affect patient experiences and trial outcomes. Let’s consider a site that has multiple issues with informed consent. A traditional monitoring approach would log a problem, potentially classify the site as high risk, and deploy general retraining on informed consent. HFA instead looks for trends in the data, which in this case reveals that consent errors became more frequent after that site had enrolled 100 or more patients. This indicates that the root of the problem was not deficits in training, but constrained resources. To mitigate risk, a CRA may limit the site to no more than 75 patients and provide some relevant additional investigator training.
Lessons From The Military
For centuries, militaries operated with a command and control leadership style, in which subordinates blindly accepted their commander’s orders. Many militaries, including the Royal Navy and U.S. Army, recently switched to a new leadership style called mission command as evidence increasingly demonstrates that command and control is neither effective in preventing error, nor in truly engaging troops.
In command and control, a captain may bark, “Guard this door.” In mission command, according to Dr. David Slavin, a Royal Navy veteran and former head of risk technologies at Pfizer, troops receive the order and insight into why it is important to the operation as a whole. The captain may instead instruct, “Our goal is to prevent the enemy from infiltrating this building. Be aware that they are known to use tear gas and may try to enter through windows once they discover the door is guarded.” This background information increases troops’ ability to adapt, successfully execute their purpose in the plan, and protect the operation. It also serves to remind them that they are part of a team and that their work matters.
Whether due to risk aversion, indifference, or other factors, clinical trial operations tend to employ a command and control style of leadership. For example, CRAs may be instructed to manually verify source data and address any errors with retraining, regardless of cause.
Under mission command, CRAs would be instructed that a subset of data points and related errors are critically important to the quality of the trial and patient care. For many CRAs, this instruction foments increased alertness to risks, often those not explicitly captured in training, and better performance at identifying the real reasons why the errors came to be. In combination with HFA, the mission command approach arms staff with the flexibility to anticipate and resolve the inherent complications of conducting clinical research in a large, diverse network of sites, particularly at sites with multiple problems.
Leadership Communications
Lessons From NASA
During a visit to NASA headquarters in 1962, then-U.S. President John F. Kennedy asked a janitor what he was doing. The janitor replied, “Well Mr. President, I’m helping put a man on the moon.” The janitor knew the work he did, no matter how far removed from the actual act of a moon landing, contributed to the success of that team.
A similar philosophy towards patient engagement can enhance clinical trial quality. It can be as simple as a physician taking an extra minute to explain how each visit leads up to helping that patient and others regain health. When a patient feels like their actions are contributing to curing a disease, they are going to take that job very seriously, whether that means taking the correct number of pills at the right times of the day or not missing their monthly appointments. Patients are inspired to contribute when they are valued.
Lessons From Behavioral Decision Science
Great communication is not the act of talking or writing, but listening to both the recipient and his or her environment. In fact, environmental influences play a pivotal role in a recipient’s interpretation of a message and his or her behavior thereafter. If information is not personally relevant to the recipient, there is a risk that the he or she will not act on the intended message.
Baruch Fischhoff, a behavioral decision professor at Carnegie Mellon University, applied this concept to change teen pregnancy prevention in a Washington D.C. neighborhood. A number of safe sex and condom distribution campaigns there had proven to be poorly efficacious. Fischhoff’s team first built a mental model of the neighborhood’s teenagers, noticing that they perceived boys had the final word and girls had no power. For the girls, this was their reality and the insurmountable impediment for previous safe sex campaigns. Thus, the educational materials and video that Fischhoff’s team produced provided real-life dialogs in which girls found many ways to say, “No.” When trial participants watched the video, they were prompted to mentally rehearse responses to hypothetical scenarios and could choose which and how many video segments to watch.
Three months after the intervention, participants who watched the video were significantly less likely to report being diagnosed with a sexually transmitted disease (STD) and had a higher rate of abstinence than those who watched commercially available STD risk and prevention materials. By understanding the mental model for decision-making in this population, Fischhoff was able to alter the actions of these individuals.
Fischhoff is now exploring, in partnership with ICON, the mental model of patients consenting to clinical trials for various indications. His goal is to reduce the risk that errant assumptions investigators may hold about patients’ likely concerns, environmental pressures, and upbringing could lead to patients misperceiving a trial’s requirements, procedures, benefits, and risks.
The patient decision-making models he has constructed have prompted an evolution of the informed consent form (ICF). Fischhoff observed that the typical ICF, a document tens of pages long that a patient reads in a doctor’s office, inadequately prevents the risk that information that patients perceive as material to the trial (e.g., an inconvenient side effect or a schedule of visits) will be inadequately addressed, leading to recruitment and retention issues.
To evolve the ICF in context of patients’ mental model during consent, Fischhoff has developed videos and multimedia that more quickly and clearly deliver trial information that patients deem material. The educational content is delivered on a platform available to patients in a doctor’s office and while at home, helping patients to digest information in a comfortable setting and to have more informed conversations with their clinicians. Initial results indicate that an improved consent process can affect the rate of patient retention, compliance, and recruitment.
Putting It All Together
Lessons From Henry Ford
In 1913, Henry Ford's innovative assembly line reduced the time to build a car from 12 to 3 hours. While all the parts necessary for building a car existed before the assembly line, Ford was the first person to integrate the elements with a focused goal in mind.
Just as Ford combined methods, clinical trials today can combine methodologies and approaches already proven to work on their own. Risk-based monitoring, adaptive design, and electronic informed consent all have a growing body of evidence for significant improvements to clinical trial performance and clear regulatory support.
While no single person, industry, or methodology has all of the answers, the collective work of innovations is already in progress today — for example those described in the recently published book Re-engineering Clinical Trials — have transformative potential when implemented in parallel.
Perhaps that’s the greatest lesson of all: stepping back from individual innovations and contemplating how our industry works in the larger context of human nature and historical progress may lead us to breakthroughs otherwise too big to notice.