“Fixing operational problems in clinical research, regardless of their cause, will usually end up requiring a discussion that we are trying to avoid.”

We humans are a mostly affiliative species. We like to be in groups, we like to do things together. In order to do things together, we have to get along. We all know how well people get along with each other – of such stories history is made! Corporations are the ultimate business expression of affiliation, and so they are the ultimate environments for people trying to get along. Most operational process problems in clinical research can be traced to people not getting along, for many reasons ­– philosophical, psychological, scientific, interpersonal, uninformed.

Fixing operational problems in clinical research, regardless of their cause, will usually end up requiring a discussion that we are trying to avoid, because of the inherent disagreement in the topic, or personal antipathy, or a power differential. These are difficult conversations to have. We avoid them for a short-term gain in affiliative peace. The long-term cost is a problem left to fester, a reinforcement of unproductive relationships, a solidification of the intimidation of power. In such an environment, true improvement is very hard to achieve.

Conversations Left Unsaid

There are dozens of conversations we should be having during the clinical research process that go unspoken:

§ How do we tell the sponsor physician that the key opinion leader is a lousy recruiter?

§ How do we call the bluff of the manager who says he “welcomes change” but never implements any?

§ How do we get the CEO not to promise a trial start date to Wall Street that is operationally impossible to meet?

§ How do we tell Quality Assurance that their interpretation of the regulations is incorrect?

§ How do we tell an employee, or a peer, that they aren’t good enough at their job?

§ How do we tell the functional director that sending a powerless representative to our meeting in her place is a waste of everyone’s time?

§ How do we tell the data management standards committee that the need for flexibility is more important than blind adherence?

§ How do we tell the boss that the process improvement project is a waste of time because we know she has no budget to implement the recommendations?

§ How do we contribute an honest “360-degree” performance review on our peers without having it backfire?

§ And most fundamentally: how do we call someone out for not doing what they were supposed to do?

I am sure you can supply your own examples. What is happening in each of these cases is that someone is failing to execute their work properly, affecting all of those around her, and yet we enable this failure by being afraid or embarrassed to try and do something about it. It is directly analogous to “enabling” an alcoholic.

Difficult Consequences

Unfortunately, there are very concrete consequences to our avoidance of difficult conversations. Let’s take a complex, real example which at first glance seems unimportant but which has considerable consequences: A study is in startup mode with a clear sequence of events, some of which include: finish the protocol, send it to the various site ethics committees (ECs), build the electronic case report forms (eCRFs) in the meantime, certify the sites’ knowledge of the trial software, distribute study drug, start enrolling. But what if one sponsor physician on the project insists that for marketing or professional reasons, Site “D” must be one of the first enrollers, and that Site is in a country that requires final eCRFs before EC approval. As we know, that interrupts the carefully planned, time-pressured startup sequence. To wait until final eCRFs are ready, and then wait for Site D’s ethics approval, will ruin the study timeline – but the study physician insists and he’s backed by his management.

A difficult conversation is required. The study manager (or worse from a political standpoint, the data manager) has to explain the timeline change and suggest letting Site D start enrolling later. In this example, no one on the operations side felt they could confront clinical affairs with this reality without being punished for not “helping the team” by “taking on the challenge.” Instead the corporate culture expected the operations team to perform their now customary heroics to cope with the site dilemma while preserving the timeline.

The consequences of avoiding the conversation are far-ranging. The study physician feels vindicated and immune from censure. The data manager feels de-motivated by the thought of the perpetual heroics required of her. The study manager quietly seethes and seeks a future opportunity to get revenge. All of these are the seeds of inefficient clinical research.

How We Can Talk

Some may despair at having difficult conversations, and some may shrug. Look at the power differential, look at the interdepartmental antipathy, look at the personalities involved. Shrugging or despair is not enough. The consequences on performance are too dire. While operational accountability has been loose for many years in our industry, this is ending rapidly. Inefficiency will mean your department ends up outsourced and your job (at sponsor or CRO) will be in jeopardy. In the long run it means foregoing clinical programs that may improve our medical care but we cannot afford to complete.

There are a few, challenging ways to have difficult conversations that may get us out of these situations:

§ Be courageous. Simply said and hard to do, but many situations described in clinical research operations can and are defused more easily that we expect – it just takes the courage to start the conversation.

§ Once engaged, depersonalize the situation, especially where the history of antipathy overshadows all. It’s not about what he or she said to you yesterday or last year, or what seems unfair, or what you think your career needs. It’s about the work, the study, the science.

§ Gather and focus on the facts. In all difficult conversations, the facts get lost in the power or the feelings or the legitimate stresses of time and money. Ignore these, and focus on facts that are indisputable and well-articulated.

§ Employ a work process that neutralizes the power differential, and actually makes difficult conversations routine. There are basic process designs that are intended to specifically overcome these difficulties.

§ And do not overlook the need for executive commitment to a culture that is both safe and efficient – safe for staff to challenge each other and those in power, and where staff are sincerely empowered to root out inefficiency.

Lots of things are difficult. Mostly we need to do them anyway. It is the responsibility we inherit to live in the remarkable society humans have created.

The first thing we do, let’s kill all the lawyers. – Wm. Shakespeare, Henry VI, Part 2, c. 1591

…And the procurement officers and outsourcing managers, while we’re at it. – Clinical Research Executive (Anon.), c. 2012

Professionalizing the contracting process in clinical development has gone astray. What was once meant to add a little bit of needed legal-oriented skill to project management has become a self-perpetuating web of complexity.

Here’s a true story: we were interviewing various managers of a sponsor’s clinical research department, exploring their skills at outsourcing. The departmental lawyer, the procurement officer, and the strategic outsourcing officer each said that s/he was the only one who was protecting the company from financial ruin, compliance risk and regulatory sanction. Not a single clinical person had the arrogance to assert such importance to the clinical research success. How did we get here? When and why did clinical trial expertise get so buried under layers of bureaucracy of our own making?

Every sponsor I visit has such a story. These days the stories are told as muted rumblings: monitors who report investigators furious at the site contracts sent by the sponsor’s legal department, trial schedules with built-in 3-6 month delays to allow for “CRO contract negotiations”, trial managers throwing up their hands at having to use the same CRO that failed them last time because “procurement likes them better.” Sometimes the storytellers don’t even sound like they are complaining – is this a sign of healthy acceptance, a sign of despair, or is it that no one remembers when life was easier?

When Bad Things Happen to Good Ideas

The overlapping sponsor roles of contracting/legal/procurement/strategic outsourcing all grew out of the rapid growth of CRO usage starting over ten years ago. This trend, combined with increased trial complexity, regulatory scrutiny and budget pressures, led first to a professionalization of Clinical Operations, which took responsibilities away from medical leadership, mostly appropriately. Clinical Operations diversified and specialized rapidly, and soon it was felt that study managers had too much learn, to much to be expert in, too much to handle – again, probably appropriately. One of the identified specialties was how to contract with the proliferating third parties.

But somewhere along the way, those who were supposed to advise study managers in contracting, and advise the enterprise in broad-based improvements such as standard contract terms and volume discounts, became in charge of these key elements of trial conduct. Indeed, at some companies they have become virtually in charge of the trials themselves, as a side-effect of negotiating with those who now actually do the trials (the CROs and sites, instead of in-house staff).

For instance, some companies report the procurement people now have more to say about which sites should be used than the clinical staff. While I am the first to criticize clinical managers for continuing to use under-performing sites because they want to use key researchers or long-time friends, it is no better to have procurement staff decide which sites to use on the basis of ease of negotiation or compatible liability clauses. One group should be advising the other, while the other actually decides. Now let’s see, should that be the outsourcing department or the medical director?

Professional procurement was intended to improve Clinical Operations management skills. But lawyers don’t manage trials and they don’t improve CRO relationships, although they can make them worse. The inappropriate empowerment of legal-oriented staff has introduced consistent and costly program delays; their issues can be an enormous time sink. Even the value of standardization is diluted by a level of inflexibility that is now getting worse and worse in sponsor contracting. And make no mistake: poor contracting experiences lead to poor vendor/site performance. This is strongly aggravated by the now common gamesmanship of sponsors delaying payments: poor payment performance starts site performance on a downward spiral. What was supposed to help is now aggravating and delaying the work. And the work, lest we forget, is drug development, not shiny and polished contracts. It is about good conduct practice, not good contract practice. It is, in fact, about people, not bid grids and invoice templates.

There is an assumption that contracting staff are better, indeed necessary, to create good arrangements with CROs. The thinking is that research doctors are not good managers, that individual project managers will only serve their self interest and not the company’s, and that the art of contracting is so arcane as to be unteachable. I think this assumption is abetted by both parties: study managers are happy to have one less thing to do and procurement staff want to perpetuate an aura of indispensability, as anyone does. But regardless of whether these are accurate assumptions, what was the purpose of this assumption? What was the problem we were trying to solve, and did we solve it, or did we solve something and create another problem? The solution has grown out of proportion to the problem.

Going back to our earlier example, one of the assertions of procurement indispensability is risk aversion – that we have professionalized risk management in these staff and so that is where protection lies. Generally speaking, risk aversion in this particular area of research is overkill. Site and CRO performance is not dictated by contract terms, but rather only by the skill with which Clinical Operations managers (people, not paper) learn how to monitor, measure, communicate and hold accountable their service providers. Contracting per se does not lower risk, people do. I remember long ago, my first corporate lawyer wisely advised me that if I ever actually needed to use one of his contracts to confront a business problem, I had already lost half the battle.

When Good Things Happen to Bad Ideas

Here’s a way to change the suboptimal status quo for the better. We need to vastly reduce the procurement/contracting/outsourcing infrastructure created in clinical research and instead rely on a handful of expert advisors who work with those who should be in charge: the clinical research managers. In this advisory role, procurement and legal experts can advise on the topics of good contracting, standardized terms, possible volume discounts, metrics and penalties. This advisory role, instead of a controlling role, is where the legal profession was always meant to be. It will keep us focused on executing clinical trials instead of executing lawyers, as Shakespeare would have it, while improving the relationships among the essential parties in conducting clinical research.

“Whenever we say we’re doing the best we can, it means something has gone wrong”

How often do you hear, from someone who has just failed you, that “we’re doing the best we can”? It is the universal response of auto repair shops, phone companies, catalog merchants, hospital nurses – just about anyone who has not delivered the service you expected. Instead of an apology, or a promise to improve, we are meant to believe we are getting the best there is. It is a universal and automatic expression, and also almost certainly a lie. The worst of it is, those who say it probably believe it.

A Story

Here’s a true story. I was supposed to receive a time critical, very important package one morning from Fedex. Ten-thirty AM came and went, and no package. I found the tracking number and going online, I instantly found out that the package would be on time! Except that it wasn’t. So I picked up my phone and called Fedex, seeking a human rather than a database explanation. After the usual keypad entries and insipid hold music, and various transfers, I found myself speaking to a very kind and enthusiastic representative who told me there had been strong storms in the Midwest the night before, most packages were delayed, and delivery could not be estimated. When I pointed out that their website said it was “on time”, she had no explanation other than, “we’re doing the best we can”.

Really? Was Fedex that day actually doing the best they could? Employing one of the most sophisticated worldwide tracking and communication systems in commercial use, could they not have done something other than post out-of-date, incorrect information? Could they have added a message alerting users to expect delays? Could they have had a recording on their customer service line? Could they have promised delivery within one additional day, rather than not having any prediction at all? Could they have looked at a weather forecast the day before? Could they have offered a partial refund? I’m sure you can think of many more things Fedex could have done, and then ask yourself, did they do the best they could?

Best of What?

I worry that this phrase is not just an expression, a habit of speech, or even just a feeble attempt at placating customers. I think it is a reflection of two key aspects of our daily work:

1) we’re always told how good we’re doing and how hard we’re trying; and therefore

2) we settle for the status quo, believing that since we work long hours, we must be doing the best we can.

What this means for clinical research is very important: we ignore the opportunities, and avoid the cost and time, for improving how we work – and how our suppliers work – because it is daunting to think of changing it. It is so much easier to think we’re doing as well as we can.

Let’s review a few of the opportunities for improvement in front of us:

§ A large majority of our trials are behind their enrollment targets.

§ Clinical staff change study protocols repeatedly, even after study start.

§ Ethics committees’ reviews routinely cause months of startup delay.

§ A new study team rarely has the information to learn operational improvement from past study teams, nor do they seek this information.

§ Service providers routinely charge substantial amounts in change orders, in part because of misguided contractual terms insisted on by sponsors.

§ Joint steering committees between sponsor and service provider are neither truly “joint” nor “steering” anything.

§ EDC, ePRO, eTMFs and all the other “e’s” are surrounded with paper, negating their cost savings.

§ A focus on quality and oversight has turned into an endless loop of mistrustful triple-checking.

§ Outsourcing requires more in-house management resources than insourcing did.

And so on.

Clearly, in this 21st century world of clinical research, we are not doing the best we can. And yet, do we move through our days thinking about all the stupid things we do, or instead are we thinking about how we should stop and fix the stupid things we do? And let’s not shy away from an important new reality in clinical research: it is even harder to stop doing stupid things when we are paying outsiders to do them, at our direction!

Building the Airplane in the Sky, and Other Moronic Metaphors

The typical, and possibly most potent, argument for saying we are doing our best (insert shrug here), is that we can’t stop the pace of clinical research to fix it. Pick one:

§ We are building our airplane in the sky.

§ We are trying to change the tire on a speeding car.

§ We are trying to work on a speeding train.

True enough. But here we have once again a case of the pharmaceutical industry somehow thinking they invented business adversity (regulations, the unpredictability of research, recouping capital-intensive development, requiring a highly skilled workforce) and no one has ever faced such novel problems. Which is very silly. Think of the number of enormous (and ultimately successful) transformations the automobile industry has undergone in every aspect of their business, while they still built and sold cars. Think about journalism, entertainment, the computer industry. They’ve all had planes in the sky and trains on the track, but learned (eventually and sometimes painfully) how to change and execute at the same time.

Biopharma desperately needs to recognize when it is not doing the best it can, and figure out how to improve its operations other than mergers or outsourcing. Society needs our best; it is critically important that we find it.

It is a perverse use of the language that whenever we use this phrase, we’re doing the best we can, that it always means something has gone wrong. Our “best” is apparently never enough. When we’ve done something great, why don’t we ever stand up and say, truthfully: “we’re doing the best we can”? Shouldn’t that be the better — and only — time to say it, rather than mumbling it as an excuse for shortcomings? Let’s hope that someday, your organization, and even all of clinical research, can say this proudly.

“Consequences are good for us – they are the guides to behavior and decision-making.”

We are not hearing the truth. And we are not in danger of the consequences of the truth. This is not only a poor standard for human interaction. It is directly responsible for inefficiency and perpetuation of substandard performance in clinical development.

How many of us have dutifully filled out “RACI” charts, documenting who is responsible and accountable for various tasks or projects in the clinical development process? Having completed these charts, what happens next? Is our behavior, or the behavior of colleagues, materially changed? What level of concern do we carry when we have been assigned accountability? What are the consequences to us personally for not being successful in a task we are accountable for? Or for that matter, what are the consequences for succeeding in a task we are accountable for? I suggest that in most biopharmaceutical companies, where the projects are so large and interdepartmentally dependent, there are essentially no consequences of our actions, good or bad, in any manner that significantly guides behavior. In fact, what does guide our behavior, in a damaging way, is this fact that there are no consequences one way or another.

Developmental Immunity

What happens if we fail at our work? If we don’t lock a database on time, we miss a monitoring visit, we let our CRO go over budget, we obstruct process improvement, or – universally – we miss our enrollment dates? Do we lose our job? Our rank? Our stock options? Do we even lose our credibility? It is not at all clear that anything adverse occurs. In a very few urgent, high-stakes circumstances, the consequences may indeed be onerous. But mostly we live on, more or less unsanctioned, for as long as we care to hold on to our jobs.

What happens if we do a great job? If we bring a project in on time or under budget, keep the CRO from changing project managers, enthusiastically contribute to process improvement, write reports with no corrections needed, or – amazingly – meet the enrollment targets? Do we get promoted, get a raise, get more stock options? We hope our credibility improves. In the rare entrepreneurial biopharma environment, there may be a financial benefit. But in an environment where pay raise percentages are severely limited by policy, where cost-cutting is shrinking supervisor discretion, and where special recognition is culturally disapproved, the good news will go unrecognized. Mostly we live on, more or less unrewarded, for as long as we care to hold on to our jobs.

We do a better job raising our children than managing our companies. Good parents teach their children the consequences of their behavior, either overtly or subtly, every day. Morality, socialization, even just getting through the day, would not be possible without the concept of actions and consequences. Indeed, we probably learn more about life from the application of consequences than we do from the classroom. In the absence of consequences, we get widespread mediocrity.

In all of these ways, consequences are good for us – they are the guides to behavior and decision-making. Economic markets are ruthless appliers of consequence. If nobody needs or wants our new drug, no one will pay for it. If we can’t hire people because our salaries are too low, we can’t get out work done. If we consistently mistreat our CRO with unreasonable demands or poor communication, they will give us inattentive service in return. These are the consequences that teach us how to succeed as people and businesses. So why don’t they apply internally?

Risk, Rewards, Fear, Teams, Truth

There are at least five consequences of the lack of consequences:

· The absence of truth

· Irrational fear of truth-telling, and consequences

· Dilution of the value of rewards

· Hiding individual accountability behind teams

· An unhealthy absence of risk.

What does truth have to do with this? Telling the truth should have consequences – good ones. Too often we are afraid of bad consequences, and so routinely withhold the truth, thus depriving us of the good consequences of the truth as well. For instance, if we are afraid to say that a manager is not performing well, then all sorts of bad things happen: that manager does not learn she needs to improve, her staff continues to suffer, the work continues to suffer, improvement is blocked, staff cynicism grows. But most supervisors are more afraid of telling someone they need to improve or else. It’s the “or else” – the consequences – that must be at the heart of the performance feedback. Otherwise we’re back to stagnating in mediocrity.

There is good reason to think that this situation will get worse over time. It has been widely observed that we are raising a generation (at least in the US) without consequences – where everyone gets a trophy in every competition, and it is culturally incorrect to admonish poor performance. Presumably there will be a rude awakening – in college, or on the job. But what if corporate environments imitate the youth culture as this generation becomes the management layer?

Perhaps the best cover given to poor performance in most clinical development groups is the reliance on teams. I’ve written this before and it can’t be said enough: teams cannot be held accountable for anything. Only individuals can. While I suppose it is possible to demote or fire a team, it is highly unlikely. Team accountability makes the application of consequences almost impossible. You can give everyone a t-shirt if the team succeeds; can you take the shirt off their back when they miss the deadline?

A powerful fairytale for our fearful, politically-correct corporate cultures is the Emperor’s New Clothes. Clinical development groups are filled with courtiers afraid to point out that the manager next door, or their team leader, is naked. Trying to articulate and enforce consequences is risky. What if staff get offended and leave? What if they resent one person being rewarded but not them? Will you keep your job if you reveal the emperor has no clothes? Will you open your company to lawsuits if you tell an employee the truth about their poor performance?

We have to create clinical research environments where it is safe to be a truth-teller about process performance, or we will never improve our productivity. We have to know there will be consequences, fairly and quickly applied, for all of our actions. Truth, and consequences. In the fairytale, it is the little boy who has the courage to tell the truth. We need him now, at a company near you.

“What are we in clinical development going to do to contribute to therapeutic innovation?”

At a recent holiday party, one of my neighbors said to me, “YOU are in clinical trials – why do they take so long and cost so much? What’s up with that?” After I explained that I wasn’t personally responsible for the state of drug development, I gave my standard answer (concern for safety, regulatory requirements, minimum drug exposure periods, hypotheses complexity, etc.), but I thought about the question later. With all the advances in science in the past twenty years, what have been the advances in how we prove drugs are safe and effective?

A simplistic overview over the past decades would conclude that there really have only been two big ideas in clinical development in this period: enabling technology, and outsourcing. Basically, EDC (electronic data capture) got rid of the paper and CROs got rid of the people. Of course, more than that has happened, as we will discuss below, but regardless, what have the results been? Fewer people, fewer companies, fewer drugs. While we are ultimately dependent on breakthrough science, what are we in clinical development doing to contribute to therapeutic innovation?

Information Technology Leads the Way

Information technology has generated considerable innovation, time saving, and enhanced analysis over the last fifteen years. EDC has been transformative, the very definition of “enabling” technology. EDC has made adaptive trial design possible, even though the concept was around for decades. EDC has altered the relationship between sponsor, sites and monitors fundamentally. It has enabled a series of related technologies to blossom (electronic patient-reported outcomes, interactive response technology for patient randomization and drug supply management) and made other technologies infinitely more practical and useful (data warehouses, clinical trial management systems, etc.).

ePRO has given us data direct from the patient which can be relied upon for accuracy and timeliness – a dramatic change. Improved adverse event systems that have enabled rapid and comprehensive regulatory filings have eliminated much burdensome hand work and improved analysis for signal detection. The electronic submission of a panoply of regulatory documents – most importantly NDAs/BLAs – has progressed a great deal from the CANDAs of twenty years ago. Sponsors now understand how to apply technology to the fundamentals of trial documents, in an electronic trial master file. Data warehousing is replacing clinical data management systems and helping sponsors grapple with the flood of new data types (images, samples, genomics data, lab data) which go well beyond simply replacing the paper case report form.

Strategy Innovation Lags Behind

But what can we say about innovation in the strategy for and process of clinical development in recent years? In particular, what can we say the industry has achieved other than exploiting technology and outsourcing the heavy lifting?

There has been no lack of concern. For years, industry meetings (and private executive discussions) have focused on critical issues such as reducing the “white space” on the timelines between one trial and another, one program and another. We have focused on the continuing, nearly unchanged dilemma of slow patient enrollment times. We have struggled with resourcing trial tasks through the peaks and valleys of clinical development pipelines. We have despaired at the continuing dilemma of study protocols taking too long to develop and changing too often once they are finally finished. But most of all we have been under constant pressure to work faster and cheaper without compromising scientific or ethical integrity.

So what have we done in response? There is not much to highlight. Investigative sites have improved their trial competency and the efficiency of their operations. We have traveled to many new corners of the world looking for trial subjects, hoping this will accelerate enrollment completion. Study designs have evolved usefully, not just toward adaptive designs, but with an eye to combining Phases, using branching protocols, and exploring secondary study endpoints as a means to do better, faster science. The “virtualization” of study teams (again enabled by today’s technology) has tried to overcome geographic barriers to hiring and assembling the best talent to manage and execute trials.

We have played endlessly with organization charts, chasing the ever-elusive answer to the question of which comes first – functional discipline, operational process, or therapeutic specialty? Having tried every possible combination with equivocal results, we should not expect an answer to that one soon.

Most noticeably, we have tried probably every possible staff resourcing strategy under the sun: outsourcing, in-sourcing, off-shoring, in-shoring, functional service providers, full study outsourcing, complete outsourcing, contract employees, and more. In the end, what we as an industry have developed as a process innovation is the concept that any and every piece of the clinical development process can be done by someone else. We haven’t innovated how the outsider does it, just the fact that an outsider is the one who does it. There is no question that outsourcing has been a huge business success – the CRO market is estimated to exceed $20 billion in annual revenues. But is it more than a shell game, moving resources using the same mediocre processes from the sponsor to the outside?

Where Will the New Ideas Come From?

Let me be clear that clinical development runs much better today than twenty years ago. We not only have technology to thank for this, but the creativity of innumerable specialists within and outside sponsor companies and CROs who have tackled specific niche problems of clinical research and improved them. We can fully expect that these incremental improvements will continue. But what will emerge as the “next big idea?”

Undoubtedly there will continue to be technology-based improvements. Perhaps putting the power of unlimited computing literally in the hands of everyone, everywhere, can be harnessed for a leap forward in the research process. Globalization (of sites, patients and scientists) will become more and more prevalent and could be more powerful than software. There are so many questions, we can hope that some of them will lead to exciting answers: How will the industry adapt to governmental and public backlash on profits? Will personalized medicine offer more than new justifications for very high prices? How will investigative sites, and more specifically, patients/trial subjects, be changed by the breadth of instantly accessible information (regardless of accuracy or relevance)? Will global economics permanently change the dynamics of the industry, governments, insurers and patients?

Meanwhile…

As we start a new year, with the challenges in clinical development only getting more difficult, what answer do we have besides getting someone else to do the work? Do we await the next technology revolution and hope those in another industry show us the way once again? Almost all change comes either as a reaction to adverse circumstances, or through the disciplined cultivation of “eureka” moments. The adverse circumstances are here now; eureka we will have to wait for. In the meantime, are we working at our tasks as efficiently as we possibly can? Certainly not. Maybe the idea, for now, is an old one: work harder at working smarter.

©Waife & Associates, Inc., 2012