REVIEW

When Science is Not Enough: A Framework TowardsMore Customer-Focused Drug Development

Nektarios Oraiopoulos . William C. N. Dunlop

Received: April 20, 2017 / Published online: June 17, 2017� The Author(s) 2017. This article is an open access publication

ABSTRACT

Introduction: The purpose of this study was toidentify the key barriers to a customer-focuseddrug development process and develop a com-prehensive framework to overcome them.Methods: The paper draws on existing litera-ture, both academic and practitioner, across arange of disciplines (innovation management,marketing, organizational behavior, behavioraleconomics, health economics, industryreports). On the basis of this extensive review, aconceptual framework is developed that offersconcrete suggestions on how organizations canovercome the barriers and enable a more cus-tomer-focused development process.Results: The barriers to collaboration are orga-nized into three distinct categories (economic,behavioral, organizational), and within eachcategory, a one-to-one mapping between barri-ers and solutions is developed.

Conclusion: The framework is specificallydesigned with the objective of offering action-able and practical advice to executives who facethese challenges in their organizations. Thepaper provides a unique theoretical contribu-tion by synthesizing findings from several aca-demic disciplines with concrete examples fromthe pharmaceutical industry.Funding: Mundipharma International Limited.

Keywords: Business; Customer-focuseddevelopment; Drug development; Finance;Health economics; Market access; Patientaccess; Pharmaceuticals; R&D management;Value frameworks

INTRODUCTION

On 18 October 2007, Pfizer announced thewithdrawal of its inhaled insulin brand Exu-bera� from the market. Just a few months ear-lier, the company predicted the drug would be aUS$2 billion-a-year product. Yet, at the time thewithdrawal decision was taken, Exubera hadbarely reached sales of US$12 million, and ithad cost Pfizer over US$2.8 billion [1]. It is notuncommon for pharmaceutical companies toterminate drugs and write off substantialinvestments during clinical trials (and occa-sionally after market launch) owing to safetyconcerns. But what really made Exubera standout as one of the most unprecedented and

Enhanced content To view enhanced content for thisarticle go to http://www.medengine.com/Redeem/8598F06012A083E4.

N. OraiopoulosCambridge Judge Business School, Universityof Cambridge, Cambridge, UK

W. C. N. Dunlop (&)Mundipharma International, Cambridge, UKe-mail: will.dunlop@mundipharma.com

Adv Ther (2017) 34:1572–1583

DOI 10.1007/s12325-017-0567-y

stunning failures in the history of the pharma-ceutical industry [1] was the fact that the with-drawal decision was entirely based ondisappointing sales, rather than any safety orefficacy concerns [2].

One unique feature of the pharmaceuticalindustry is that the patient who receives thedrug is rarely the person who decides on thattreatment or even pays for it. Instead, pre-scribing decisions in developed healthcaremarkets are typically made by physicians, andthe payments are often made by insurancecompanies, government, or other organiza-tions, with little or no payment directly fromthe patient. This creates a unique environmentin which a successful commercialization strat-egy has to meet not only the regulatoryrequirements of the Food and Drug Adminis-tration (FDA) or European Medicines Agency(EMA) but also the requirements of the threeend customers: the patient, physician, andpayer. The term ‘‘payer’’ broadly refers to thehealthcare budget holder and can include, e.g.,Medicare in the USA. A high profile Europeanorganization, which reviews therapies from acost perspective and provides guidance, is theUK National Institute for Health and CareExcellence (NICE). Traditionally, pharmaceuti-cal companies have launched blockbusterdrugs by just focusing on patients and physi-cians. This is no longer the case: given the everrising reimbursement hurdle, a successfulcommercialization strategy relies on a soundunderstanding of the payer’s needs and priori-ties. This aspect has been overlooked in theexisting academic literature. In this article, wedevelop a framework to address this gap andenable organizations to better align their drugdevelopment process with the expectations ofthe end customer, and more specifically, thepayer.

Consider again the case of Exubera, a drugthat faced significant opposition from payers. Inthe USA, many insurance companies refused tocover treatment because Exubera was moreexpensive than existing treatments: US$5 com-pared with US$2–3 for injectable insulin [3]. Inthe UK, NICE argued that Exubera should onlybe approved for diabetics who had a proven fearof needles, as it offered no advantage over

existing treatments [3]. The Exubera story is byno means unique. In recent years, there havebeen several drugs that met regulatory safetyand efficacy hurdles, but failed to meet payerhurdles, e.g., Avastin� [4] and Zaltrap� [5].

Executives in pharmaceutical companies arewell aware of the increasing role of the payer,and the fact that regulatory approval can nolonger be considered a guarantee for marketsuccess and profitability. According to Harrison[6], 24% of drug failures in phase II and phase IIIcan be attributed to commercial and strategyreasons. Given that the average developmentcost for a new drug is estimated at US$1.4 bil-lion (out-of-pocket) and US$2.55 billion (capi-talized) dollars [7], anticipating and avoidingsuch market failures earlier in the process couldpotentially save companies hundreds of mil-lions of dollars in research and development(R&D) costs. However, to achieve this, closecollaboration between the R&D and commer-cial teams is critical. Such a collaborationrequires that the project initiation and contin-uation decisions are made using the best possi-ble information, rather than subjective andbiased estimates. Numerous academic studieshave shown how better communicationbetween the commercial and R&D teams canlead to projects with a higher success rate, agreater percentage of sales coming from newproducts, and a higher likelihood of successfullydelivering the company’s strategic goals (see [8]and the references therein). As a result, theextant academic literature highlights theimportance of having a cross-functional teamthat brings together experts from variousdepartments and ensuring that the projectprogresses to meet both the technical andcommercial requirements [9, 10].

However, for most pharmaceutical compa-nies, and despite the growing realization aboutthe importance of the payer, the flow of infor-mation between the R&D and commercialteams is not always optimized. For example,development teams might not always includeappropriate comparators (e.g., existing stan-dards of care) in the clinical trials, but ratherrely on placebos. This might be acceptable forregulatory approval by the FDA or EMA, but it isoften not sufficient for the payer, who is likely

Adv Ther (2017) 34:1572–1583 1573

to require evidence of direct comparison withthe existing standards of care (when applicable).As a result, a number of projects that are unli-kely to meet the customer’s expectations con-tinue to consume valuable resources as a resultof biased information and organizational inef-ficiencies. By its very nature, developing newdrugs will remain a highly risky and complexprocess. Expensive failures are an inherent partof medical discoveries, and that is why it is socritical for organizations to identify these fail-ures as fast and efficiently as possible. This isprecisely why maintaining a focus on thepatients’ and broader societies’ true needsbecomes even more important. This challengeis equally critical for drugs that are developedin-house (by the internal R&D team of theorganization) or are in-licensed from externalorganizations (e.g., through various types ofpartnerships).

The goal of this study was to develop aframework that would enable a better commu-nication flow, and develop tangible suggestionson how to make the drug development processmore customer-focused. The first step was toidentify key barriers to cross-functional collabo-ration that have traditionally prevented anintegrated development process in the pharma-ceutical industry. To better understand theunderlying causes of these barriers, they werecategorized into those driven by economical,behavioral, or organizational reasons. For eachcategory, an extensive literature review wasconducted to identify the most effective man-agement practices to overcome the specificchallenges. This review covered a number ofacademicpapers that spannedawide spectrumofresearch (from organizational and behavioraleconomics to innovation management andorganizational theory). For each literature-basedsolution, specific examples on how the proposedsolution can be applied to the drug developmentprocess are provided. The central proposition ofthis work is that overcoming these barriers willallow pharmaceutical companies to prioritizeprojects that are more customer-focused and,therefore, generate more value (both for thecompanies and the patients). This article doesnot contain any new studies with human oranimal subjects performed by any of the authors.

CHALLENGES TO COLLABORATIONAND COMMUNICATION

Economic

The first and foremost characteristic of pharma-ceutical R&D is its highly uncertain nature,manifested directly in the high failure rates ofnew drugs. Developing new products is chal-lenging and entails a great deal of risk for mostindustries [11]. Even so, the attrition rates of newpharmaceutical products are at a different orderof magnitude compared even with the riskiestindustries: for those drugs that successfully reachclinical testing (testing in humans) fewer thanone in five (around 16%) will receive regulatoryapproval [12]. These high attrition rates havebeen relatively stable across the past few decades,unlike the cost of conducting clinical trials,which has consistently increased [13].

The high attrition rates mean that pharma-ceutical R&D teams are often susceptible toso-called progression-driven behavior wherethey focus their efforts on advancing a com-pound to the next stage without first ensuringthat this is the right decision. However, in manycases, potential failures could be detected ear-lier. A recent study by Pfizer examined thedecision-making processes behind 44 develop-ment programs [14]. A striking finding was thatall programs that had successfully completedclinical proof of concept (PoC) could provideclear evidence that proper testing procedureshad been followed in previous stages. On thecontrary, for almost half (43%) of those thatfailed at PoC, there was little evidence that allthe necessary testing procedures had been fol-lowed. Had proper testing taken place, thesefailures would have been detected much earlier,saving the company millions of dollars. Theneed for superior clinical outcomes against thestandard of care is even greater when a pricepremium is required to meet the cost of devel-opment. The existence of this progres-sion-driven behavior is also highlighted in anextensive longitudinal study of AstraZeneca’ssmall-molecule pipeline that examined the rootcauses behind project failures [15]. A possibleexplanation, suggested by the authors, is that

1574 Adv Ther (2017) 34:1572–1583

‘‘the use of volume-based metrics encouragedproject teams and leadership groups to progressprojects to the next phase in order to meetyearly goals’’.

A second characteristic of the drug develop-ment process that further complicates collabo-ration and efficient decision-making is its longdevelopment cycles. Even if we exclude thediscovery phase of a new compound, it stilltakes 6–7 years from filing of a new molecularentity to getting regulatory approval (Pharma-ceutical Research and Manufacturers of America[16]). During these lengthy cycles, a number offactors can separate the actions of an individualor group from the end goal of the project:market launch. In addition, given the highlyspecialized nature of the tasks and activitiesinvolved, this process is highly fragmented:different groups in the company, or even dif-ferent companies (e.g., contract research orga-nizations), are involved at different stages. As aresult, scientists and managers are often rewar-ded on the basis of interim goals [17]. Whilesuch a reward scheme might seem understand-able, given the lengthy development times, aby-product of these interim goals is that man-agers are often held accountable for reachingonly a specific milestone, without bearing anyoverall responsibility for the subsequent clinicalor market access success of the drug. As such,the different organizational units tend to oper-ate in ‘‘silos’’ [17–19] that develop a ‘‘throw-o-ver-the-wall’’ mentality at the hand-over points.This can be hugely inefficient if seeminglypromising products that meet interim safetyand efficacy goals lead to commercial failures ifthe end customer’s criteria are not taken intoaccount early on and incorporated into theclinical development plan [20].

The high costs associated with the develop-ment process in the pharmaceutical industrycan pose a third critical obstacle to collabora-tion across the different organizational units.Given the high cost of conducting clinical trials,which often run into the hundreds of millions[7], the different business units (e.g., therapeu-tic areas) within the company often have tocompete for very scarce resources. To securetheir budget and, therefore, their future devel-opment, business units might convey overly

optimistic forecasts about the technical andcommercial success of a given drug [21].Because of the highly specialized nature ofthose forecasts, it is not a trivial exercise forsomeone outside the business unit to challengethem or the methods used to reach them.Clearly, this is extremely inefficient, as thoseforecasts are bound to fail in later stages, afterhaving consumed an enormous amount ofresources.

Taken together, the three key characteristicsof the pharmaceutical industry (high failurerates, long development times, and hugedevelopment costs) can create an environmentin which there is misalignment in incentivesbetween specific individuals or groups and theoverall organization. When that happens, col-laboration suffers and development teams focuson the short-term viability (or competence) oftheir narrowly defined ‘‘silo’’, rather than con-tributing to the end goal: developing cus-tomer-focused medicines.

Behavioral

Among the most pervasive managerial behav-iors is the practice of ‘‘throwing good moneyafter bad’’. As noted in Cooper [22]: ‘‘projects geta life of their own and become like expresstrains, slowing down at the stations, but neverwith the intention of stopping until they reachtheir ultimate destination, market launch’’.Royer [23] also analyses a number of productfailures, highlighting that development teamskept going even though there were clear andconsistent signs pointing to a near-certain fail-ure. For example, during the development of anovel lens, both the regulatory authorities andopticians expressed clear concerns about itsbenefits. However, the development team choseto ignore them [23].

The seminal work of Staw [24] demonstratedthat subjects were likely to continue makinginvestments in failing projects, despite evidenceof negative performance. Importantly, moremoney was invested in a project when thesubjects themselves, rather than an outsider,were also responsible for earlier funding deci-sions. This self-justification effect has been

Adv Ther (2017) 34:1572–1583 1575

replicated in a number of subsequent experi-ments [25, 26].

A second important bias that prevents effi-cient decision-making is the confirmatory bias.Here, it is not the past involvement of thedecision-maker that biases their future actions.Instead, their current beliefs affect how theyseek new information about the project.Specifically, cognitive studies have shown thatpeople tend to seek information that confirmstheir existing beliefs [27, 28], and at the sametime heavily discount information that contra-dicts those beliefs [29].

As discussed earlier, the R&D process in thepharmaceutical industry is a complex endeavorthat requires highly specialized expertise.Moreover, developing ground-breaking sciencefor unmet medical needs requires an extraordi-nary amount of perseverance and commitmentto succeed. However, those very qualities canoften lead to overconfidence [30]: scientistsbecome ‘‘true-believers’’ that their compoundwill be successful, despite all the evidenceagainst it. This creates an emotional attachmentthat further amplifies self-justification andconfirmatory bias. As a result, the scientists (orthe specific development team) develop a‘‘groupthink’’ mentality [31] where the teamisolates itself from the rest of the organizationand defends its product by being overly opti-mistic or by focusing on a narrow set of criteriathat might not include the end customer’sneeds.

Organizational

Developing and marketing new products is atruly interdisciplinary process that involves arange of individuals from different organizationalfunctions. Research has shown that employees’beliefs and values, and hence their behaviors, arelargely driven by the specific functions or groupsthey belong to. This is expressed in Dougherty[19]: ‘‘Departmental thought worlds partition theinformation and insights. Each has a distinctsystem of meaning which colors its interpretationof the same information, selectively filters tech-nology-market issues, and produces a

qualitatively different understanding of productinnovation’’.

The existence of an insular culture in whichpeople work closely with and learn only fromtheir own group, while excluding those outsidethe group, is a major barrier to collaboration,and ultimately to product performance. Theeffect is clearly demonstrated in Hansen et al.[32] who studied 121 product developmentteams at Hewlett-Packard: the study identifiedcertain teams that only sought solutions withintheir own team rather than reaching out toother divisions (even when the problem at handrequired an interdisciplinary solution). Inter-estingly, the data also revealed that when teamsdid reach out to other divisions, they did notnecessarily reach out to the ones that wouldhave the highest expertise for their particularproblem. People tend to approach those theyknow and have good relationships with [33] orthose they feel most comfortable with [34].

Such organizational barriers to cooperationbecome even more critical in highly complexdevelopment environments such as the phar-maceutical industry. First, the specialized nat-ure of the different processes and tasks creates adirect, and at first sight unavoidable, barrier tocollaboration. Specifically, development teamsin the pharmaceutical industry might consist ofexperts from target and drug discovery teams(from synthetic organic chemistry to molecularand cellular biology), clinical teams (frompharmacodynamics and toxicology to pharma-cokinetics), and regulatory experts. At the sametime, more commercially focused teams consistof experts spanning market access, marketing,medical affairs, finance, and corporate affairs, aswell as business development teams that iden-tify the most promising opportunities from theexternal environment. These individuals areexperts in their own area, and possess knowl-edge and experience that cannot be easily cod-ified or quantified. The organizational literaturerefers to such tacit knowledge as ‘‘sticky’’: andtransferring it across organizational functions isnot a trivial task [35, 36]. Hence, an integrateddecision-making process has to rely on the col-laboration and effective participation of allteams.

1576 Adv Ther (2017) 34:1572–1583

FRAMEWORK TO OVERCOMETHE CHALLENGES AND PROMOTEA CUSTOMER-DRIVENDEVELOPMENT PROCESS

Addressing Economic Barriers

Firstly, a customer-driven mentality requires afocus on the end goal (i.e., commercial success),and as such, a decision-making process that willeffectively filter products that will ultimatelynot meet the customer’s requirements. For thepharmaceutical industry, this implies thatincentives for the development teams need tobe designed in such a way that they promotetruth-seeking experimentation, rather thanexperiments that would accomplish short-termmilestones. To overcome such misalignedincentives, recent research has shown thatmotivating the right kind of experimentationrequires a culture with tolerance for failure[37–39]. In practice, and specifically for thepharmaceutical industry, a tolerance for failurewould require a stronger emphasis on rewardingthe decision-making process itself and not justthe interim outcomes. Such process-based met-rics encourage employees to focus on the tasksthat generate the highest value for the organi-zation, even when those tasks cannot bedirectly observed and measured [40].

A second key step towards a more cus-tomer-centered development process is tofacilitate transparency in the resource allocationprocess. Sharpe and Keelin [21] emphasize theimportance of an integrated framework thatenables transparency and consistency in theassumptions and assessment criteria that dif-ferent business units use to assess their R&Dprojects. In the example they describe, the keytask of the development teams was not to pre-pare reports that defended their projects, but todemonstrate and document their assumptionsand information sources in a succinct and reli-able way. On the other hand, the role of seniormanagement was to challenge these valuationsin a constructive way, and facilitate a discussionin which the strengths and weakness of eachproject were identified not in isolation, but withrespect to the R&D portfolio of the entire

organization. This iterative and complementaryrelationship between a top-down (where seniormanagement sets the priorities) and a bot-tom-up approach (where the developmentteams provide the critical information) isessential for effective project portfolio selection[41, 42].

Lastly, incentive schemes should strike abalance between bonuses based on individual orunit-specific performance and those based oncompany-wide performance. The latter are par-ticularly effective in enabling better collabora-tion and knowledge sharing, particularlyamong competing groups [43, 44]. Huckmanet al. [45] describe in detail how Wyeth phar-maceuticals restructured its incentive andreward systems, so that the bonuses of all eli-gible scientists in the development teams werebased on the degree to which the entire orga-nization achieved its objectives. The key objec-tive of this initiative was to motivate thedevelopment teams to look beyond theirdepartmental ‘‘silos’’ and strengthen synergiesacross the various therapeutic areas.

Addressing Behavioral Biases

The first step in overcoming any behavioral biasis the realization that we are all susceptible tothem. This is far from trivial. As noted in Russoand Schoemaker ([30] and references therein),most humans, including managers, have arather poor understanding of the limits of theirknowledge. According to Lovallo and Kahne-man [46], such overconfidence often leadsmanagers to believe that they can overcome thechallenges in executing a project, even whenthere is clear evidence to the contrary. This isalso consistent with Hammond et al. [47] andSengupta et al. [48] who argue that even themost experienced and intelligent managers aresusceptible to behavioral decision traps. Assuch, a key recommendation in all of the abovestudies is that decision-makers should realizethe limitations of relying solely on their ownjudgment, and instead seek to validate itthrough objective feedback from other units orfrom outside the organization. It is importantthat people look at the data without being

Adv Ther (2017) 34:1572–1583 1577

influenced by their role in the project or theirpersonal connections with others working onthe project [23, 46, 47]. For the pharmaceuticalindustry, an advisory board could ensure thatchallenging questions about the payer’srequirements are asked early in the process,rather than a few months before market launch.There are also now more formal options toreceive this external customer feedback, withthe EMA facilitating a parallel regulatory andHealth Technology Assessment (HTA) scientificadvice procedure [49].

Secondly, Lovallo and Kahneman [46] stressthat companies have to balance optimism andenthusiasm (keeping employees motivated andfocused) with realism (ensuring objective fore-casts). Along the same lines, Russo and Schoe-maker [30] suggest that managers shoulddistinguish between the stages of deciding anddoing. The former requires realism and a fact-based approach, the latter relies more on moti-vation and encouragement. For pharmaceuticalproducts, a more objective assessment at thedecision stage can be achieved by setting theprogression criteria on the basis of lessons learntfrom comparable projects that have previouslybeen through the commercialization process. Asdiscussed in Lovallo and Kahneman [46] this‘‘outside view’’, which relies on data external(rather than internal) to the project, leads tomuch more objective assessment because ‘‘in-ternal view’’ assessments tend to come withnumerous cognitive biases that distort theirobjectivity. It is important to stress here thatexternal data are not limited to just quantifiablemetrics and analytics. Relying only on these islikely to result in companies missing what cus-tomers really value [50]. The essence of theexternal view is that it is inclusive of bothqualitative and quantitative insights.

Lastly, senior management should empha-size the need for objective and data-drivenestimates about the project’s profitability, andparticularly about potential factors that mighthinder its commercial success. Whereas man-agers are usually happy to produce optimisticforecasts and reports about their projects,research shows that they are much less effectivein seeking information that disconfirms theirbeliefs [29], especially when this information is

not readily available. For pharmaceutical prod-ucts, it is often difficult to provide precise esti-mates about a new drug’s cost-effectivenesswith respect to existing treatments. Modelsabout the economic value of a drug tend to beextremely sensitive to the particular assump-tions applied. A potential step forward would bethe development of open-source health eco-nomic models (for an example, see [51]) thatwould provide greater transparency and, inturn, more credible and consistent estimations,not only across the different companies but alsofrom the payer’s perspective.

Addressing Organizational Barriers

The first and foremost requirement for a collab-orative decision-making process is the existenceof unifying goals that require collective effortand commitment from all departments withinthe organization, including senior management[52]. For this reason, Morgan Stanley decided tointroduce peer-evaluation, even in their seniormanagement committees. According to TomDeLong, senior executive at the time: ‘‘OperatingCommittee members who normally did notshare the important knowledge of their divisionsrealized that at the end of the year they would beevaluating one another. All of a sudden theybegan to share more information, knowing theconsequences at year-end for their evaluations ifthey didn’t’’ [52]. For the drug commercializa-tion process, this implies that even though theoverarching governance body consists of indi-viduals from different functions, it shouldmaintain a clear and strong focus on the endcustomer’s needs [53].

The second important step towards over-coming an insular culture is the development ofso-called T-shaped management [54]: managerswho can simultaneously excel in their own area(the vertical part of the ‘‘T’’) and support collab-oration across the different areas (the horizontalpart). Such behavior is in sharp contrast withthat of ‘‘lone star’’ managers, who only focus ontheir own performance, or that of ‘‘corporatebutterflies’’, who seemingly collaborate witheveryone in the organization but without servinga specific purpose. Fostering a T-shaped

1578 Adv Ther (2017) 34:1572–1583

management approach means that collaborationneeds to be measured and evaluated. Hansenet al. [55] discuss how management consultancyBain and Company directly evaluates theirpartners on how much they have helped othercolleagues by collecting detailed data on theircollaborative activities. This evaluation com-prises a significant part of the partner’s annualcompensation (up to 25%). The concept of

encouraging T-shaped managers is also consis-tent with a number of other studies thatdemonstrate the importance of cross-functionalcollaboration in new product development set-tings [8, 56]. Such a cross-functional manage-ment approach is also imperative for managersin the pharmaceutical industry who need to beable to communicate effectively, both withfunctional experts and across functions.

Economic

Setup process-based metrics that tolerate

milestone failure Focus on end-goal

metrics rather than short-term ones, even if projects are terminated

earlier.

Develop robust and transparent criteria to

allocate resources between projects

Focus on the quality of informa�on and the

assump�ons behind the calcula�ons.

Link incen�ves to the end goal

Consider linking employee incen�ves

directly to overall company performance.

Behavioral

Promote transparency in assump�ons and

discussion

Encourage sharing of selected informa�on, such as open-source pharmacoeconomic

models.

Encourage individual awareness of decision

traps and tes�ng of ideas (internal/external)

Encourage payer customer feedback via

advisory boards and scien�fic advise etc.

Dis�nguish between deciding (over-

confidence obscures judgment) and doing

(over-confidence promotes execu�on)

Use previous case studies and analogues to set the

right benchmarks.

Organiza�onal

Facilitate unified goals

Create interdependencies and

informa�on sharing between departments.

Foster T-shapedmanagement (measure,

evaluate, reward)

Leaders are ac�vely encouraged to work

beyond their immediate departments.

Encourage organiza�onal

understanding of customer needs

Use comprehensive frameworks (such as the

BEACON) to communicate internally

the payer's needs.

Economic

Setup process-based metrics that tolerate

milestone failure Focus on end-goal

metrics rather than short-term ones, even if projects are terminated

earlier.

Develop robust and transparent criteria to

allocate resources between projects

Focus on the quality of informa�on and the

assump�ons behind the calcula�ons.

Link incen�ves to the end goal

Consider linking employee incen�ves

directly to overall company performance.

Behavioral

Promote transparency in assump�ons and

discussion

Encourage sharing of selected informa�on, such as open-source pharmacoeconomic

models.

Encourage individual awareness of decision

traps and tes�ng of ideas (internal/external)

Encourage payer customer feedback via

advisory boards and scien�fic advice etc.

Dis�nguish between deciding (over-

confidence obscures judgment) and doing

(over-confidence promotes execu�on)

Use previous case studies and analogues to set the

right benchmarks.

Organiza�onal

Facilitate unified goals

Create interdependencies and

informa�on sharing between departments.

Foster T-shapedmanagement (measure,

evaluate, reward)

Leaders are ac�vely encouraged to work

beyond their immediate departments.

Use comprehensive frameworks (such as the

BEACON) to communicate internally

the payer's needs.

Fig. 1 Actions to address barriers to customer-focused drug development

Adv Ther (2017) 34:1572–1583 1579

Lastly, companies should aim to createbridges between business units that tradition-ally were perceived as ‘‘distant’’ and cultivatethe exchange of knowledge that would allowthe creation of a common perspective: howeach unit contributes to the customer-drivenstrategy of the company. For example, Dunlopet al. [53] developed a conceptual framework toaddress the key needs of the payer and, there-fore, overcome potential reimbursement hur-dles. Importantly, the framework accounts forboth technical and commercial aspects. Themnemonic BEACON (Burden/target popula-tion, Environment, Affordability/value, Com-parator, Outcomes, Number of studies/qualityof evidence) was developed to promote theunderstanding of the payer’s needs throughoutthe pharmaceutical organization. The diversityin the criteria reflects the different skillsrequired to appropriately assess this and, inturn, the reason why the existence of across-functional team is an integral part of asuccessful commercialization process. More-over, a key enabler for such integration is that itis embraced and perceived as a fair process [57]by all the teams involved in it. As discussed inLoch and Tapper [40], a fair process is necessaryas it helps employees overcome suspicion andaccept changes by engaging them (asking forinput), explaining the choices (using consistentcriteria), and setting clear expectations (whatneeds to be achieved in the future).

CONCLUSIONS: SUMMARIZINGTHE FRAMEWORK

The proposed framework is presented in Fig. 1.It shows the three key categories of barriers tocollaboration (economic, behavioral, and orga-nizational) together with specific actions thatpharmaceutical companies (and others) cantake to address these and enable a more cus-tomer-focused development process.

ACKNOWLEDGEMENTS

This study was funded by MundipharmaInternational Limited. Editorial services wereprovided by Dr. Joanna Todd (Stellar MedicalCommunications Limited) and were funded byMundipharma International Limited. Wewould like to thank Antony Mattessich(Managing Director Mundipharma Interna-tional) for actively supporting and encourag-ing research towards customer-focused drugdevelopment. The article processing chargesand Open Access fee associated with thispublication were funded by MundipharmaInternational Limited. All named authorsmeet the International Committee of MedicalJournal Editors (ICMJE) criteria for authorshipfor this manuscript, take responsibility for theintegrity of the work as a whole, and havegiven final approval for the version to bepublished.

Disclosures. William Dunlop is an employeeof Mundipharma International Limited. Nek-tarios Oraiopoulos is an employee of theUniversity of Cambridge. Judge Business SchoolExecutive Education Ltd received funding forthe literature search and analysis.

Compliance with Ethics Guidelines. Thisarticle does not contain any new studies withhuman or animal subjects performed by any ofthe authors.

Open Access. This article is distributedunder the terms of the Creative CommonsAttribution-NonCommercial 4.0 InternationalLicense (http://creativecommons.org/licenses/by-nc/4.0/), which permits any noncommer-cial use, distribution, and reproduction in anymedium, provided you give appropriate creditto the original author(s) and the source, providea link to the Creative Commons license, andindicate if changes were made.

1580 Adv Ther (2017) 34:1572–1583

REFERENCES

1. JohnsonA. Insulin flop costs Pfizer $2.8 billion. 2007.http://www.wsj.com/articles/SB119269071993163273. Accessed 17 August 2016.

2. Gutierrez C. Pfizer washes its hands of Exubera. 2007.http://www.forbes.com/2007/10/18/pfizer-earnings-closer-markets-equities-cx_cg_1018markets47.html.Accessed 17 August 2016.

3. Rao A, Purkayastha D. ‘‘Exubera fiasco: what wentwrong?’’, teaching note 508-115-8. Hyderabad:ICMR Center for Management Research; 2008.

4. National Institute for Health and Care Excellence.Bevacizumab in combination with paclitaxel andcarboplatin for first-line treatment of advancedovarian cancer. 2013. https://www.nice.org.uk/guidance/ta284?unlid=5013752442016211165322.Accessed 17 August 2016.

5. National Institute for Health and Care Excellence.NICE issues final guidance aflibercept for colorectalcancer. 2014. https://www.nice.org.uk/news/press-and-media/nice-issues-final-guidance-aflibercept-for-colorectal-cancer. Accessed 17 August 2016.

6. Harrison RK. Phase II and phase III failures:2013–2015. Nat Rev Drug Discov. 2016;15(12):817–8.

7. DiMasi JA, Grabowski HG, Hansen RW. Innovationin the pharmaceutical industry: new estimates ofR&D costs. J Health Econ. 2016;47:20–33.

8. Griffin A, Hauser JR. Patterns of communicationamong marketing, engineering and manufactur-ing—a comparison between two new productteams. Manag Sci. 1992;38(3):360–73.

9. Olson EM, Walker OC Jr, Ruekert RW, Bonner JM.Patterns of cooperation during new product devel-opment among marketing, operations and R&D:implications for project performance. J Prod InnovManag. 2001;18(4):258–71.

10. Slater SF, Mohr JJ, Sengupta S. Radical productinnovation capability: literature review, synthesis,and illustrative research propositions. J Prod InnovManag. 2014;31(3):552–66.

11. Calantone RJ, Schmidt JB, Di Benedetto CA. Newproduct activities and performance: the moderatingrole of environmental hostility. J Prod InnovManag. 1997;14(3):179–89.

12. DiMasi JA, Feldman L, Seckler A, Wilson A. Trendsin risks associated with new drug development:success rates for investigational drugs. Clin Phar-macol Ther. 2010;87(3):272–7.

13. Munos B. Lessons from 60 years of pharmaceuticalinnovation. Nat Rev Drug Discov. 2009;8(12):959–68.

14. Morgan P, Van Der Graaf PH, Arrowsmith J, et al.Can the flow of medicines be improved? Funda-mental pharmacokinetic and pharmacologicalprinciples toward improving phase II survival. DrugDiscov Today. 2012;17(9/10):419–24.

15. Cook D, Brown D, Alexander R, et al. Lessonslearned from the fate of AstraZeneca’s drug pipe-line: a five-dimensional framework. Nat Rev DrugDiscov. 2014;13(6):419–31.

16. Pharmaceutical Research andManufacturers of Amer-ica (PhRMA). Biopharmaceutical research & develop-ment: theprocess behindnewmedicines. 2015.http://www.phrma.org/report/biopharmaceutical-research-and-development-the-process-behind-new-medicines.Accessed 16 Sept 2016.

17. Hauser JR, Simester DI, Wernerfelt B. Internal cus-tomers and internal suppliers. J Mark Res.1996;33(3):268–80.

18. Souder WE. Achieving organizational consensuswith respect to R&D project selection criteria.Manag Sci. 1975;21(6):669–81.

19. Dougherty D. Interpretive barriers to successfulproduct innovation in large firms. Organ Sci.1992;3(2):179–202.

20. Cooper RG. Third-generation new product pro-cesses. J Prod Innov Manag. 1994;11(1):3–14.

21. Sharpe P, Keelin T. How SmithKline Beechammakes better resource-allocation decisions. HarvBus Rev. 1997;76(2):45–6.

22. Cooper RG. Winning at new products: acceleratingthe process from idea to launch. Cambridge:Addison-Wesley; 2001.

23. Royer I. Why bad projects are so hard to kill. HarvBus Rev. 2003;81(2):48–56.

24. Staw BM. Knee-deep in the big muddy: a study ofescalating commitment to a chosen course ofaction. Organ Behav Hum Perform. 1976;16:27–44.

25. Bazerman MH, Beekun RI, Schoorman FD. Perfor-mance evaluation in a dynamic context: a labora-tory study of the impact of a prior commitment tothe rate. J Appl Psychol. 1982;67(6):873–6.

26. Bazerman MH, Giuliano T, Appelman A. Escalationof commitment in individual and group decisionmaking. Organ Behav Hum Perform. 1984;33(2):141–52.

Adv Ther (2017) 34:1572–1583 1581

27. Jonas E, Schulz-Hardt S, Frey D, Thelen N. Confir-mation bias in sequential information search afterpreliminary decisions: an expansion of dissonancetheoretical research on selective exposure to infor-mation. J Pers Soc Psychol. 2001;80(4):557–71.

28. Bolton LE. Stickier priors: the effects of nonanalyticversus analytic thinking in new product forecast-ing. J Mark Res. 2003;40(1):65–79.

29. Biyalogorsky E, Boulding W, Staelin R. Stuck in thepast: why managers persist with new product fail-ures. J Mark. 2006;70:108–21.

30. Russo JE, Schoemaker PJH. Managing overconfi-dence. Sloan Manag Rev. 1992;33(2):7–17.

31. Janis IL. Groupthink: psychological studies of pol-icy decisions and fiascoes. Boston: Houghton Mif-flin; 1982.

32. Hansen MT, Mors ML, Løvas B. Knowledge sharingin organizations: multiple networks, multiple pha-ses. Acad Manag J. 2005;48(5):776–93.

33. Hansen MT, Løvas B. How do multinational com-panies leverage technological competencies? Mov-ing from single to interdependent explanations.Strateg Manag J. 2004;25(8/9):801–22.

34. Casciaro T, Lobo MS. Competent jerks, lovablefools, and the formation of social networks. HarvBus Rev. 2005;83(6):92–9.

35. Zander U, Kogut B. Knowledge and the speed of thetransfer and imitation of organizational capabili-ties: an empirical test. Organ Sci. 1995;6(1):76–92.

36. Szulanski G. Exploring internal stickiness: impedi-ments to the transfer of best practice within thefirm. Strateg Manag J. 1996;17(S2):27–43.

37. Azoulay P, Zivin JSG, Manso G. Incentives andcreativity: evidence from the academic life sciences.RAND J Econ. 2011;42(3):527–54.

38. Ederer F,MansoG. Ispay forperformancedetrimentalto innovation? Manag Sci. 2013;59(7):1496–513.

39. TianX,Wang TY. Tolerance for failure and corporateinnovation. Rev Financ Stud. 2014;27(1):211–55.

40. Loch CH, Tapper UAS. Implementing a strat-egy-driven performance measurement system foran applied research group. J Prod Innov Manag.2002;19(3):185–98.

41. Terwiesch C, Ulrich K. Managing the opportunityportfolio. Res Technol Manag. 2008;51(5):27–38.

42. Loch C, Kavadias S. Implementing strategy throughprojects. In: Morris PWG, Pinto J, Soderlund J,

editors. The Oxford handbook of project manage-ment. Oxford: Oxford University Press; 2011.p. 224–51.

43. Birkinshaw J. Strategies for managing internalcompetition. Calif Manag Rev. 2001;44(1):21–38.

44. Schlapp J, Oraiopoulos N, Mak V. Resource alloca-tion decisions under imperfect evaluation andorganizational dynamics. Manag Sci.2015;61(9):2139–59.

45. Huckman RS, Pisano GP, Rennella M. Wyeth phar-maceuticals: spurring scientific creativity withmetrics. Harvard Business School Case 607-008.2010.

46. Lovallo D, Kahneman D. Delusions of success: howoptimism undermines executives’ decisions. HarvBus Rev. 2003;81(7):56–63.

47. Hammond JS, Keeney RL, Raiffa H. The hiddentraps in decision making. Harv Bus Rev. 1998;76(5):47–58.

48. Sengupta K, Abdel-Hamid TK, Van Wassenhove LN.The experience trap. Harv Bus Rev. 2008;86(2):94–101.

49. European Medicines Agency. ‘‘Best practice guid-ance for the parallel regulatory—HTA scientificadvice procedure’’. 2016. http://www.ema.europa.eu/docs/en_GB/document_library/Regulatory_and_procedural_guideline/2016/03/WC500203944.pdf. Accessed 17 August 2016.

50. Christensen C, Hall T, Dillon K, Duncan D. Knowyour customers’ ‘‘jobs to be done’’. Harv Bus Rev.2016;94(9):54–62.

51. Sullivan W, Hirst M, Beard S, et al. Economic eval-uation in chronic pain: a systematic review and denovo flexible economic model. Eur J Health Econ.2016;17(6):755–70.

52. Hansen MT. Collaboration: how leaders avoid thetraps, create unity, and reap big results. Boston:Harvard Business Press; 2009.

53. Dunlop WCN, Mullins CD, Pirk O, et al. BEACON: asummary framework to overcome potential reim-bursement hurdles. PharmacoEconomics. 2016.doi:10.1007/s40273-016-0427-7.

54. Hansen MT, von Oetinger B. Introducing T-shapedmanagers: knowledge management’s next genera-tion. Harv Bus Rev. 2001;79(3):106–16.

55. Hansen MT, Nohria N, Tierney T. What’s yourstrategy for managing knowledge? Harv Bus Rev.1999;77(2):106–16.

1582 Adv Ther (2017) 34:1572–1583

56. Troy LC, Hirunyawipada T, Paswan AK.Cross-functional integration and new product suc-cess: an empirical investigation of the findings.J Mark. 2008;72(6):132–46.

57. Kim WC, Mauborgne R. Fair process: managing inthe knowledge economy. Harv Bus Rev.2003;81(1):127–36.

Adv Ther (2017) 34:1572–1583 1583