TOWARDS AN INTEGRATED CENTRAL CLINICAL TRIAL UNIT
A shared service model to become a competitive academic research organisation
Student:
Dr G.K. Hovingh (g.k.hovingh@amc.uva.nl)
MBA Health Care Management, Amsterdam Business School Student UvA net ID: 10739734
MBA Supervisor:
Prof. Dr J. Strikwerda (j.strikwerda@uva.nl) Management Consultant and
Professor of organization and change (internal governance) at the Amsterdam Business School of the University of Amsterdam.
AMC Supervisor:
Prof. Dr J.A. Romijn (j.a.romijn@amc.uva.nl)
Head of the depatment Internal Medicine, Academic Medical Center Prof. Dr. E.S. Stroes (e.s.stroes@amc.uva.nl)
Head of the department Vascular Medicine, Academic Medical Center
EXECUTIVE SUMMARY
Clinical Research (“patient-centered research”) is widely performed at different clinical trial units (CTU) in Academic hospitals, such as the Academic Medical Center (AMC) and the Vrije Universiteit Medical Center (VUmc). Currently, at a divisional level, a number CTUs are active, and within the organizational structure of DIVA (division of internal medicine AMC) every sub-department organizes clinical trials within their own CTU organization. The size, impact and organizational structure of these CTUs vary greatly.
The hypotheses for this project were:
- “Products” delivered by a CTU are valuable in light of the overall AMC and VUmc strategy. - An integrative CTU (central CTU (CCTU)) will add value by standardisation and economy of
scale.
- In the phase towards a putative merger, the output (to be defined; see below) of the CTU of the department of vascular medicine (CTU-VAS) can be optimized, which will result in increased production (in EUR) by 20% (>100K/yr).
In order to define whether a merger would be beneficial for the current stand-alone CTU-VAS, parameters of outcomes of the CTU-VAS and other departments were evaluated and a SWOT (strength weakness opportunity threat) analysis was performed to assess the position of CTU’s within the AMC and VUMC.
Based on the company project, I conclude that CTUs indeed deliver an important asset for the overall UMCA strategy and that these merits should be made tangible, visible and quantifiable in order to measure and monitor the net added value.
Different models can be applied to generate a CCTU, and a merger perspective, where mutual benefit is anticipated, is the best model for this purpose. In the pre-merger phase, the aims should be specified and aligned with both the overall UMCA and departmental strategy. Prior to a
merger, CTUs should be organized as self-contained, professional organizations, which will require a transition from its current position as a machinery bureaucracy towards a professional
organization. Such trajectory is started at the CTU-VAS, one of the larger CTUs within the AMC. The optimisation of the procedural efficiency, the focus on value adding activity (the sponsor is defined as the customer) and restructuring based on skill-tasks alignment will result in a 20% increase of benefits in the CTU-VAS.
Based on semi-structured interviews, literature reviews and data concerning trial conduct, two potential mergers are identified: between the CTU-VAS and the CTU of the department of gastroenterology and/or the CTU of the department of cardiology at the VUmc. However, both mergers are currently not to be directly implemented in light of the decision making process about the putative location of the departments in the future UMCA alliance.
The first step to be taken is to form a competence center for CTUs, comprising approximately 5 FTEs who are dedicated to generate the basic prerequisites for any CTU. This team has the task to ensure that:
1) an ICT system is created that will enable a CTU to enroll participants from a larger pool of patients,
2) a financial control system (contract negotiation and management) is in place, 3) all quality and IRB related issues are uniformly being executed in all CTUs. 4) the pharmacy is integrated in the clinical trial conduct process.
This essential backbone should be endorsed by, and paid for by the board of directors given the impact of clinical trial conduct on the overall strategy of the AMC (and VUmc).
TABLE OF CONTENTS
Executive Summary 2
Table of contents 4
1) Introduction and Research Question 6
2) Added Value delivered by a Clinical Trial Unit: 10
The role of clinical trial conduct in the overall strategy of an academic hospital in general, and the AMC-VUMC alliance in particular
3) CTU from a theoretical perspective
3.1 Organizational models applied to a CTU 16
3.2 Competitive advantage; Resource Based Value perspective 19 3.3 The “ideal” CTU: an international survey amongst leaders of 20
successful CTUs.
4) Scientific output generated by CTUs in the deparment of internal medicine 24
5) Evaluation of the status of the CTU-VAS and other CTUs in DIVA 25
5.1 Rationale to analyse CTUs 27
5.2 Organization of clinical trials at the CTU-VAS 27
5.2.1 The value chain 27
5.2.2 The “why?” question 28
5.2.3 The 4P model 29
5.2.3.1 Patient Care 30
5.2.3.2 Publication 32
5.2.3.3 Payment 35
5.2.3.4 Planned trials 38
6) Organizational models for a merger 41
7) Towards a merger; semi-unstructered interviews 41
8) The future of clincial trials; innovation or disruption? 45
9) Conclusions 46
10) Recommendations 48
11) Acknowledgements and addendum 51
1 INTRODUCTION
In this thesis, I describe the results of a company-project, conducted in the framework of the MBA course focused on Health Care Management. The general aim of the company project, as being stated by the MBA dictate was “to describe the roadmap for change in the organization where the student works, in order to optimize the process and, as a consequence, to either earn or save financial resources (total budget impact 100K).”
It is interesting to note that the latter (the strive for a 100K benefit for the organization) was removed from the requirements at a later stage.
From the “change perspective” one could either focus on progression by means of incremental innovation and process optimisation (which is in line with the constant improvement; i.e. Lean) or by a disruptive solution (which is in line with innovations described by i.e. Christensen (2009)). In this thesis both aspects will be discussed.
The overall topic of this company project is focused the organizational structure of the Clinical Trial Unit (CTU) in general, and the current and future position of the CTU of the department of vascular medicine (CTU-VAS) in particular. The first steps and the analyses for this company project were taken in february 2015, and the project ever since became a process that did not remain a theoretical conceptualization, but rather a reality in some aspects, and as such, this company project can be classified as a form of “action research”. The reason to focus on the CTU was driven by my position as head of the CTU-VAS, and I was therefore privileged to not only study the situation, but also to implement some of the systems to improve the quality and quantity of the outcome masures. The CTU-VAS operates as an independent entity within the Department of Vascular Medicine and the scope of activity is to conduct clinical (patient oriented) research, mostly randomized clinical trials (RCTs) on behalf of sponsors (biotechnology and pharmaceutical industries). The CTU-VAS team comprises a total of 13 professionals (8,5 FTE); 1 MD, 1 contract- and database specialist, 1 team leader (who also is active in the actual trial conduct) and 10 research nurses (RN).
When this company project started in February 2015, it was unclear which outcome was
generated by the team’s joined efforts, and until 2015, these outcome parameters were not well defined by the head of the department and myself.
One of the most important and urgent outcome measures for the continuity of the CTU is the financial revenue generated by CTU related activity. While the specific “making money” aim was removed from the requirements for the company project, I feel urged to address the potential financial impact of the proposed changes in working attitude, process handling and organizational structure.
The assumption prior to starting this company project was that the formation of an overall CCTU would be beneficial for all of the merging CTU’s and based on this assumption the following research questions were generated:
- which aspects are pivotal for generating a business model for an integrated internal medicine CCTU (i.e. which parameters should be aligned in light of post merger success factors)
- How should the CCTU be formed in terms of organizational structure? (ie complete new entity, in a horizontal – pan divisional/ departmental structure?)
- How should the resource –revenue balance be created in a format broadly adopted by the departments involved in the novel CCTU structure?
On the roadmap of the ultimate formation of a new CCTU a number of research aims were defined.
I set out to investigate:
1) Whether CTUs generate added value to the AMC/VUmc and, if so, how the aims of the CTU fit into the overall VUmc-AMC strategy (i.e. in the alliance strategy) (legitimacy of the organization)
2) What kind of organization a CTU is, and what is considered best practice in terms of organizational design (based on literature search)
3) Which outcome parameters are to be analyzed in light of the strategic focus described under 1)
4) Whether differences exist between DIVA departments in terms of generating these outcome parameters; (focus, efficiency and efficacy)
5) Whether a merger of clinical trial units from different departments is feasible and desirable. The latter criterium is analyzed and evaluated from the perspective of the different stakeholders and compared to figures generated in literature. (i.e. economy of scale). “benefit” in terms of outcome parameters is defined and analyzed in semi-structured interviews with different stakeholders.
6) The implications of a merger on the organization (re)structuring (mandates) while taking into account the implicate undesignability paradigm (Strikwerda (2014))
7) The advantages and disadvantages of the merger scenarios based on key success indicators, described in literature.
8) The pro’s and cons of different merger scenarios based on the AMC –Vumc situation. Data for this analysis were generated by semi-structured interviews with different stakeholders.
The clinical trial conduct process was analyzed by the CTU-VAS team in order to optimize the working process, which would enable the CTU-VAS to become a stable partner in future mergers. The time course taken in this analysis follows the roadmap (see below) towards a potential merger, which would ultimately result in a CCTU:
1) evaluation of the status of CTU-VAS from a strategic and organizational perspective (comparator: organization structures described in literature)
2) evaluation of the status of the CTU-VAS and other CTUs from a process perspective (based on data with respect to “best practice” of trial conduct)
3) recommendations and outcomes of interventions to strengthen the position of the CTU-VAS in a potential merger.
Figure 1) the roadmap towards a CCTU
Lastly, a summary of recommendations for the assumed merger is provided and different scenarios are given.
The methodology used for the analyses in this thesis are multi-faceted and range from extracting data from academic (peer reviewed) literature, publically available documents from the AMC, VUmc and UMCA, interviews with stakeholders in non AMC-CTUs (i.e. CTU in UMCG, UMCU and private trials organisations) and interviews with internal stakeholders (department of internal medicine, cardiology and gastroenterology in AMC and VUmc). Data from these sources will be described in order to generate a decision making model with regards to the ultimate goal; the formation of a centralized clinical trial unit.
2 ADDED VALUE DELIVERED BY A CLINICAL TRIAL UNIT
THE ROLE OF CLINICAL TRIAL CONDUCT IN THE OVERALL STRATEGY OF AN ACADEMIC HOSPITAL IN GENERAL AND THE AMC-VUMC ALLIANCE IN PARTICULAR.
Clinical research in the AMC and VUmc is commonly conducted by a CTU team, and these CTUs vary in size and operational activity. The scope of all of the CTUs is to conduct patient oriented research on behalf of sponsors (biotech and pharmaceutical companies; contract research) or on behalf of investigators (Investigator initiated research). For contract research mostly phase II and phase III studies are performed, with the specific aim to quantify the effects of non-registered medical interventions (meaning that these drugs are not readily available to clinicians in the field) in patients with different forms of disease. These patients are recruited from the outpatient or inpatient clinics and, upon written consent, are randomized to either an active drug or a placebo. This randomization is mostly done in a double-blind fashion, meaning that both the research team and the patient are unaware of the treatment being given.
In general, a CTU research team comprises:
- a PI (principal investigator; a medical doctor, who is responsible for all of the clinical trial conduct),
- a MD (directly involved in medical aspects of the trial conduct; ie physical examinations) - a research nurse (RN, who mostly performs blood withdrawals, takes questionnaires and
hands out the medication).
In large-scale CTUs a legal and financial officer as well as a secretary are mostly added to the team. The sponsor generates the financial resources and as such should be regarded as “the customer” for the CTU. Clinical trials should be executed while adhering to quality related prerequisites (see below) and the protocol, and data derived in this context is perceived as “the outcome measure” that holds value for the customer (principles of “fitness for use” and “total quality” described by i.e. Juran (1988), and for the CTU-VAS described in MBA quality-paper oct 2015). The need of the customer is well defined and evaluated by the sponsor, the CRO and CTU.
The sponsor is responsible for the design of the trial and for the IT infrasctructure. In RCTs, source data from documents are uploaded into a web-based (inter)national database). A CRO (clinical research organisation) is hired (by the sponsor) to perform interim quality checks of the data. The latter is crucial and mandatory given the fact that novel medication, with a potential harmful effect, is investigated in most clinical trials. The responsibilities and mandates are all clearly
marked and are subject to different legal regulations (i.e. Good Clinical Practice Guidelines, Wet Bescherming Persoonsgegevens (WBP), Wet Medisch Wetenschappelijk Onderzoek met Mensen (WMO), EU Clinical Trials Directive).
Once a trial is designed and about to be launched, CTUs are contacted (by the CRO or sponsor) with a request to participate in the trial. Upon signing a CDA (confidentiality agreement, checked by legal department in AMC/VUmc), the PI inventorizes whether
a) the patient category (for the specific trial) is accessible,
b) the trial is feasible in terms of resources (i.e. imaging modalities present and accesible?) and c) whether the fee per patient is sufficient.
The latter can also be performed by the financial manager, in case present in the CTU team.
The financial structure of clinical trials conduct is relatively straightforward. The sponsor makes an estimate about the number of individuals to be enrolled in a clinical trial (power calculations are based on the extent of baseline risk/disease state, and anticipated beneficial effect during a predefined follow-up) and based on this number the sponsor negotiates with the CRO about the financial aspects of the clinical trial. In some trials, the full trial program is contracted by the CRO and in this case, the CRO acts as the “subcontractor” that subsequently contracts CTUs worldwide; the sponsor outsources all logistics to the CRO. The CRO in this case is considered the customer for the CTU.
A sponsor can also directly negotiate with CTUs and this is mostly the case in relatively small-scaled trials, requiring a small number of enrolling CTUs. The sponsor also directly contracts the CTU for trials where a particular rare type of patients should be enrolled; i.e. patients suffering from rare metabolic diseases, whose care is centralized in an academic hospital. Under such circumstance, the sponsor either negotiates with a CTU at a local level (i.e. sponsor Pfizer
negotiates with CTU-VAS AMC) or at a national organization level (i.e. sponsor AMGEN negotiates with VRN (vascular research network), an organization representing a number of CTUs in the Netherlands).
The total fee for the CTU consists of:
1) a fixed part (irrespective of the number of patients enrolled), which comprises a start up fee, IRB (internal review board) and pharmacy fee (both pass through; invoice directly to
2) a flexible amount, based on the number of patients screened and randomized.
Fixed fee budgets are in the order of 10K per study, while the variable component can vary from 5K to over 300K.
Almost every subdivision in the AMC and VUmc (i.e. hematology, gastroenterology, cardiology) organizes trials in a CTU organization and all of these CTUs are approached by the same CROs and sponsors for participation in clinical trials. The sizes of these CTUs vary from 1 RN to teams of over 10 individuals. The CTU-VAS is the largest CTU in the AMC and operates as an independent entity within the department of Vascular Medicine. The team members are hired by the AMR ( AMC Medical Research, the research BV at the AMC) on behalf of the head of the department of Vascular Medicine. Financial and human resources are outsourced to the AMR and the
department of vascular medicine pays a yearly fee, which is based on the sum of the salaries of the CTU (fixed percentage). A total of 13 people are working at the CTU (total of 8,5 FTE) and the annual running costs is approximately EUR 730,000 (of which 95% is related to salaries). An average of 20-30 studies are conducted at the same time and approximately 6 studies are in the active recruitment phase at any given moment. The recruitment is done by the PI, who is assisted by a team of PhD students who typically enroll 2-40 patients per study. The financial status of the CTU-VAS is monitored by the AMR, the head of the CTU and the head of the department of vascular medicine (prof. E.S. Stroes).
The CTU operates within the context of the total organization; the AMC. As a consequence, one of the key questions to be addressed is whether “the product” generated by a CTU adds value in light of the strategy defined by the AMC in particular and an academic hospital and medical school in general. In other words: “are the aims of the CTU and the overall strategy of the academic hospital alligned, and if so, in which fields of focus?”
In broad terms, Dutch university hospital are involved in: 1) patient care (the hospital perspective),
2) education (i.e. medical students, nurses) and 3) research
The focus on- and resource allocation to these three main fields is largely driven by strategic decission making. Alignment of the needs of- and financial resources generated by stakeholders (i.e. patients, insurance companies (patient care), government (education) funding agencies
(research)) with the potential within the organization (healthcare providers, educational staff and researchers and facilities) is crucial in defining this strategy.
It is of particular interest that in recent years, the AMC and the VUmc are progressing towards a complete merger into an UMCA. In light of this perspective, a strategic alliance program
committee has been formed. In their 2014 report (“op weg naar excellentie”), it was stated that the new merger should strive to serve as a top-referent care center, with a close connection to fundamental and translational research (Alliantie AMC-VUmc (2014, p.6)). The translational research is described to become organized in “centers of excellence”. The latter are units of extended tertiary care and research, underlining the academic perspective and nature of the institution.
The outcome parameter aimed for is defined as ”to become the structural number 1 in the
Netherlands, focussing on research where a number 1 or 2 position can be obtained in terms of H-index and impact factor.” The H-factor, or Hirsch factor, is invented by Jorge E. Hirsch, and is commonly accepted as an author level metric to quantify both the productivity and citation impact of manuscripts generated by a scientist (described in chapter 4)
The overall aim of future UMCA is “not only to become a national leader, but also to be an
international competative academic center”. It is pivotal to analyze the role of the CTUs in general and the CTU-VAS in particular in light of these strategic aims. The current (2015) position of the UvA (=AMC) in the international QS ranking list of medical schools is 46, while the VUmc is not ranked in the top 100 (website: QS score (2015)). This QS score is a widely adopted ranking based on 36 subjects, including academic reputation, employer reputation and research impact. On the QS ranking list, both AMC and VUmc by themselves are thus not meeting the criteria stated in the mission statement; in most fields the Erasmus Univerity in Rotterdam is ranked on a higher position, while in some specific areas, both the VU and UvA are not even in the national top-3 position. The question arises whether a merger per se would result in the top 1 or 2 position. It is imperative to strategically focus on developing areas on which these rankings are based. It is interesting to note that in the Times Higher Education ranking list UvA and VU are ranked 47 and 65, respectively (website: Times Higher Education Score (2015)). The latter is based on the aggregate of markers of teaching (total students/staff, PhD/staff, reputation survey; 27.5%), research volume (income and reputation; 27.5%), citations per paper (35%), international outlook (7.5%) and income from industry (2.5%). This clearly shows that scientific output (which is
determinant and the strategic focus of the intended UMCA should therefore be (partly) directed towards gaining a stronger scientific position in the strive to become an international successful competitor.
In terms of the legimacy of the CTU within the academic merger one should first address which “product” is delivered by the CTU and whether and to which extent this has relevance in the overall strategy. The “product” delivered by the CTU is clinical research, and once published, this CTU outcome does add value to the overall strategy of the UMCA. The relative magnitude of this added value is in terms of publications is analyzed for the department of internal medicine by the author and the outcomes of this studies will be addressed in chapter 4. It is interesting to note that in recent literature it was stated that “clinical trials units provide essential academic functions, which should be incorporated into routine practice.” (Chapman (2015))
Apart from the scientific output, CTU activity also generates a financial income. These potential profits are realized by two different financial value streams. The traditional perspective on the financial outcome generated by a CTU is to calculate the net benefit based on the difference between income (the sum of start-up and enrollment fees) and cost associated with CTU-related activity (95% FTE related cost). A grosly neglected financial aspect of the CTU activity is, however, based on substitution. The benefit of free drug administration within the clinical trial (substitution) should also be regarded as “net income”, especially when this lowers the expenses made in
medication that are to be paid for by the “academic component” part of the total budget. The latter could have a huge effect, once the cost of medication is exceedingly high. One example of such situation is the treatment for patients with hemophilia or specific forms of cancer. The budget impact of this form of “hidden revenue” is discussed in chapter 5.2.3.3.
Importantly, academic hospitals tend to be a tertiary referal center for patients with complex diseases and/or patients with common diseases combined with a number of comorbidities. This is of particular interest given the fact that in recent years we do observe a tendency to strategically focus on “academialisation”, a process by which an academic hospital aims to select for- and focus on patients meeting the criteria for tertiary care. This “focus on core business” is driven by the market; an academic center, due to large overhead is too expensive as a care provider for many of the “non-complex” diseases. The clinical trials conducted in academic hospitals are largely focused on patients with “hard-to-treat” clinical entitities such as cancer. By participating in clinical trials,
the academic hospital underlines the focus on finding medical solutions for specific (i.e. rare, or hard to treat) patients, which fits the strategy of “academialisation” in a clincial perspective.
Lastly, clinical trial conduct can be used as an educational model for (future) healthcare providers. In clinical trials interventions with a potential harmful effect on patients are being performed, and this not only requires a thorough understanding of risk-benefit thinking, but also requires strict procedural and quality control measures. It is of note that sponsor initiated research needs to comply with all of the quality measures (grounded in i.e. the Helsinki declaration and the Good Clinical Practice guidelines), while in investigator initiated research these quality measures are, in some cases, not as well guarded due to the absence of a monitoring system. While aiming for “top-quality” research, both from a safety, as well as from an outcome (valid and controlled data) perspective, clinical trial conduct might serve as “best practice”, that should be implemented in all types of research. This would imply that “clinical trial conduct” should be more incorporated in the curriculae for medical trainees.
To summarize; clinical trial conduct does generate financial resources, scientific output, and patient-oriented solutions, and could play a role as a teaching model for high standard research conduct. All of these outcomes fit the overall strategy of an academic hospital in general and the UMCA acclaimed strategy in general. A detailed view on the different input and output
parameters is warranted to come to a tangible decission about whether or not to invest in CTUs in general and a centralized CTU in particular. Although the outcomes of the CTU are in line with the overall UMCA strategy, cost-benefit can not be judged based on these bold outcome measures.
The “added value for the organisation” argument is true once:
Total investment (EUR) < Financial revenu (EUR) + scientific output + patient outcome + educational outcome.
The latter 3 outcomes should be translated in financial revenue (EUR) (=validation), once the net financial income is below the total investment. No formal guidelines are in place on how to financially validate these 3 outcome measures and, as a consequence, resource allocation under such circumstances is part of strategic decission-making. The basic strategic question is “do we invest XX EUR in a CTU that generates XX-Y EUR, for reasons of scientific excellence, patient or educational outcome, or do we rather invest these XX EUR differently?” The answer to this
question is provided by the head of the department, who is mandated and empowered to decide about the investments to be done.
The equation is studied in more detail for the CTU-VAS in the next chapters.
3 THE CTU FROM A THEORETICAL AND EMPIRICAL PERSPECTIVE
3.1 ORGANIZATIONAL MODELS APPLIED TO A CTU
In his organizational configurations framework, Mintzberg describes a model comprising different organizational elements (Mintzberg (1979)). Companies differ based on the composition and relevance of these elements and upon combining the organizational elements and coordination mechanisms, Mintzberg came to the following organization-types;
-
simple structure,-
machine bureaucracy,-
professional organization,-
diversified organization,-
innovative organization (adhocracy).A paradox is identified while comparing the characteristics of the academic hospital and a CTU with these types of organizations;
-
Mintzberg describes a hospital as a professional organization, characterized by a democratic form of decision making (which is restricted to certain layers within the organization). In this type of organization professionals rely on “knowledge”, which is embedded in their profession.-
A CTU is, in general, a machine bureaucracy, where specialistic procedures are conducted at a routine basis by “technical skilled” employees (research nurses, medical doctor). A machine bureaucracy is relatively insensitive to external influences. One of the largest pitfalls in a machine bureaucracy is de-motivational aspect of this organization. For a CTU, the input and output is fully standardized, partly due to the stringent prerequisites build in the quality-standards that have to be met. This repetitive and “strictly-bound” perspective in terms of procedures could be a basis for the currently observed lack of exploration capacity among the CTU-VAS team.It is interesting to note that these two types of organizations are both embedded in the
whereas in the professional organization mutual adjustment and standardisation of norms might be present, direct supervision (based on quality and procedural measures) is more likely to be the type of coordination that is required in a machine bureaucracy. Although Mintzberg’s perspective is not widely perceived as “scientific proven”, some of the characteristics of organizations
described can be recognized.
Galbraith (Galbraith (2002)) described that organization design can be defined as “simply redesigning the organization based on roles and structures”, or, and this was perceived by Galbraith as a more effective approach, by aligning the strategy (direction), structure
(specialization, shape, power distribution), process (information), rewards (motivation), and talent (people, skills). A full attention to all of these organizational elements, which are commonly
depicted as the “star model” representing the 5 items as key elements, is crucial to be
competetive. It is of note, that no dominant model for organization design is currently available, and different types of methods in organization design can be identified (intiutive, rational, rule following, procedural, based on familiar organization forms and emulation).
The internal organization is suggested to be a coherent system (Strikwerda (2014)), based on its function of coordination. A model to describe the different aspects in an organization is visualized in figure 2. The governance structure was provided by Merchant & van der Stede (2003) and this model was further evolved by Strikwerda (2014). While this model by itself does not provide a method for organizational design, it gives insight iun the different aspects of the total system. While focusing on the CCTU one could use this model to generate a design model, while taking into account the different organizational elements. The model is based on two axes; the imposed versus self coordination axis, and the implicit versus explicit coordination.
Overview of coordination mechanisms
Imposed coordination Self coordination Explicit coordination Implicit coordination Communicated Strategy Budgets/ Target setting/ Resource allocation Performance review Task-structure (personal, BU’s) Management control Information Cause-effect relations (process-) teams Professional standards Mission statement Company values Recruitment criteria Pre-organizational Socialization (vocational/ Professional training) Organizational socialization Stories Experiences Policies Standards Instructions Decision criteria Programmed decisions Assessment Remuneration promotion Training Management Development Building Loyalty Decision rights Reserved powers
Peer group control Procedures Routines Trust Exemplary behavior Operational Planning Organization culture Physical layout Dominant logic Confidence Psychological climate Open standards
3.2 COMPETITIVE ADVANTAGE; RESOURCE BASED PERSPECTIVE
A lot of the knowledge and insights of the Resource Based View (RBV) on strategy can be exploited in the evaluation of the CTU, and this knowledge can be used to define / generate a competitive strategy. Most executives will mainly focus on the interplay between environmental (external) opportunities and threats on the one hand, and the weaknesses and strengths of the
firm/organization on the other. A SWOT (Strengths, Weaknesses, Opportunities, and Threats) analysis can be readily applied to explicate the actual status of the CTU organization with regards to these parameters, and, based on the analysis, one can subsequently create a prospective strategic plan. The analysis tools to assess the threats and opportunities have advanced to a larger extent than the tools to do the same analysis for the strengths and weaknesses of the internal organization. To close this gap, Resource Based View (RBV) has been developed by, amongst others, Barney (1995). The central paradigm of the RBV model has been described: “When a firm's resources and capabilities are valuable, rare, and socially complex, those resources are likely to be sources of sustained competitive advantage”.(Barney (1995), Besanko (2007)). While taking this statement into account, one can conclude that an organization will only be viable and sustainable (in other words: be profitable over time) once the resources are, to some extent, protected and of additive value for the customer. Apart from being unique and valuable, resources should have some form of protection against imitation and a resource can be defined as anything with an impact on the end product delivered by the organization. A total of four resource indicators have been defined as crucial in determining the sustained competitive advantage:
Valuable: Is the resource valuable to the customer; is the delivered service/product perceived to represent a value exceeding the cost of production?
Rare: Is the resource (or capability) a rarity; are the assets scarce?
Imperfectly imitable: What is the barrier to producing / developing the resource to imitate, and will there be significant cost disadvantage to a firm trying to obtain, develop, or duplicate the resource/capability?
Organization: Is the firm organized and able to exploit the resource/capability? And capable to appropriate the value created?
VRIO indicator Indicator met at the
CTU yes/no Comments
Valuable yes
the direct access to (ie. genotyped) patients is unique and provides the CTU-VAS with a strong position in
(FH (familial hypercholesterolemia)) trials. While focusing on clinical trial coduct the sponsor represents the customer and the patient represents
“the pivotal part of the product”. The product deliverd (clinical trial in specific patients) is valuable to
the customer.
Rare yes
the combined imaging, infusion and intervention potential is unique compared to a non-academic setting. The translational and multi-departmental focus (ie collaborations with dept of radiology) is an
asset.
Imperfectly Imitable yes
The genotyping in selected patients is expensive and absolutely not available for a large number of centres.
The CTU-VAS can therefore enroll fast, and reliable in FH trials.
Organization no
CTU behaves as a reactive organisation, at best, making decisions not on “value to be delivered”, but
on current capacity. Half of the teammembers are reluctant to change. Output criteria are limited to numbers of visits. Profit and loss figures are in an infant state, the value and importance of the high standard of generated publications is yet to be determined. The quality of work delivered does meet
the (inter)national accepted standards.
The above-mentioned outcome of the VRIO analysis in the setting of the CTU-VAS is suggestive of the presence of a sustained competitive advantage, on the indicators V, R and I. Potential
competitors are other academic hospitals and private CTUs, competing for enrollment capacity (speed of enrollment is a key determinant for net profit). However, given the “niche” patient categories, hardly any Dutch or international CTU is a serious threat for the activity at the CTU-VAS. The CTU-VAS is consistently among the top-recruiters for studies in patients with FH, and similar figures are seen for studies related to other clinical diseases (specifically referred to the dept of vascular medicine).
For the organizational part, however, there is considerable concern, and the organizational restructuring is warranted to increase the viability of the CTU-VAS.
Moreover, this analysis only holds true for the current situation, where specific (mostly academic-type) patients are being enrolled. Combined with the fact that the academic centers are typically equipped to perform time consuming and intensive studies, new competitors are not likely to start
the context of clinical trials will change (see chapter 8); the number of studies to be conducted in these specific patients (one of the key VRIO indicators) is likely to decrease. As a consequence, the VRIO indicator will be subject to change and novel factors that meet all VRIO criteria should be identified and incorporated. This concept of dynamic capability has been described by Eisenhardt (2000) and Teece (2007).
3.3 STARTING FROM SCRATCH; DESIGN OF A CTU MODEL An international survey amongst leaders of successful CTUs
A literature search was performed to analyze whether data is available about the “best practice” in terms of the organization structure, performance and requirements of a CTU. Quite obviously, no prospective analyses has been performed to address which management style and
organizational structure is generating the best outcome in terms of scientific publications, patient satisfaction and/or financial revenue.
However, a number of studies have been conducted to analyse whether and to which extent organizational aspects impact these outcomes.
One of the most relevant manuscripts in this field is written by Gluud and Sorensen (1995), who performed a structured interview among 43 excellent organizations and units involved in clinical trials in Europe. It is of note that one of the interviewed individuals is heading the Oxford CTU (Rory Collins), and this CTU is well known for its large scale and excellent trial conduct. In light of the above mentioned strategic aims for the UMCA it is noteworthy that the university of Oxford holds the number one position in the Times Higher Education list for the last 4 years in the field “clinical, preclinical and health” subject ranking list, and could serve as a “best practice” for clinical trial conduct. The recommendations derived from the analysis are depicted below (underlined and in quotation marks) and a brief comparison is made with the current CTU-situation on a national and AMC level is depicted in italic.
Aims of the CTU: “A CTU should 1) facilitate, coordinate and perform scientifically relevant
Randomized Controled Trial (RCT)s within prevention, diagnosis, therapy and care. Such trials can be initiated by investigators, drug and device industry or the unit itself. 2) Perform systematic
reviews of RCTs (Cochrane). 3) Participate in the development of trial methodology. 4) Teach at the graduate and postgraduate level about methodology of RCTs”
NOTE: it is interesting to see the overall aims that have been put forward by the expert panel; at the AMC and most other national CTUs, the role of a CTU as a teaching facility is grosly neglected. None of the CTUs of which the author interviewed the PIs stated that this was an aim or asset of their CTU. Given the experience with clinical trial conduct in a structured, well defined and high-quality environment, a CTU could be a perfect “teaching facility” within the academic setting. Since education is a one of the key tasks of an univeristy hospital, the CTU could also add value in this domain. Educational activities are reimbursed by the university and, as such, have impact on the budget of the CTU.
Ownership: “profit and non-profit CTUs both can be viable. The latter is owned by the university or health services, and is anticipated to have competitive advantage due to the lower cost, since profit trial organizations are thought to spend resources on marketing, housing and staff, whereas in the univeristy driven CTU these costs are considerably lower (i.e. more flexibility for staff; see below)”
NOTE: a small number of for-profit CTUs are currently active in the Netherlands, and it has been shown that it is hard to make profit from clinical trials in such organizations. In 2014-2015 one of the large CTU organizations has been closing a significant number of affiliations (5 out of 7
centers), for reasons of non-profitability. The huge investment in generating databases comprising eligible patients for future studies, the premature closing (due to either inferiority or futility of the trial), and postponement of a number of clinical trials were reasons for this banktruptcy.
The financial (pro)position of the CTU-VAS will be discussed in 5.2.3.3. It can be concluded from the financial analysis that there is a significant financial risk of conducting clinical trials. Given the direct connection between risk and profit, however, the potential net profit is considerable.
Board of directors: “A CTU should be run by a board of directors comprising 5-7 internationally renowned trialists, who should meet at least yearly to advice on future development activities of the CTU.”
NOTE: none of the AMC CTUs is organized as such, principal investigators are directly supervising the trial conduct, and are kept responsible for both the quality and quantity of the clinical trials, as well as for the financial revenue to some extent. The implications of this organizational model of
Affiliation: The panel suggests that any CTU should be afflialated with an academic center to “contribute and utilise” the academic resources. The technical and methodological expertise present in a university setting is considered to be crucial for success of a CTU.
NOTE: The Dutch “for-profit” CTUs are not directly linked to an academic center or a hospital. These centers have been shown to run the risk of disconnection with front-running clinical trial- insights. However, in a number of large, non academic hospitals CTUs have been shown to be very successful (i.e. MCA and CWZ, personal communication). Albeit not directly linked with an
academic center in a formal and contracted constitution, the PIs working in these centers hold personal contact with PIs within academia, and partly via these contacts, they are directly connected to current and future sponsor-driven studies.
Dimension limits of a CTU and staff: “A CTU should be provided with a core staff and a flexible staff, the numbers of the latter will depend on the number of studies performed as well as their requirements. Belonging to the core staff, a CTU should be staffed by at least two senior medical trialists, one of whom should be the head of the CTU. These trialists should have a broad clinical experience, substantial experience in conducting RCTs. A skilled data managedprogrammer should be in charge of developing and maintaining the systems for randomisation and data management. One administrative secretary with knowledge of finances is essential for taking care of
administrative aspects, calculation of the costs of trials, running the office systems, etc. At least one project coordinator/secretary is needed in the start-up phase. The core staff should comprise 6-7 persons, who would be able to mount and launch 6-7 multicenter RCTs”
NOTE: The total FTE to number of active trials are much lower compared to most of the figures for the academic CTUs. For successful non-academic CTUs (such ast the MCA and CWZ CTU) this ratio is more in line with the proposed figure (much smaller number of clinical trials conducted by similar or marginally lower number of FTEs).
The differentiation of tasks to be done by specialized individuals is somethig that most of the CTUs do not adhere to, due to the high cost. The CTU-VAS implemented this differentiated
responsibility metric in 2012, which turned out to have a beneficial effect on the working process.
Cost of a CTU:
The anual cost of salaries was estimated to be approximately 280.000 GBP (in 1998). The total cost for a well functioning CTU was estimated to be 450.000 GBP.
NOTE: The actual cost of salaries has increased over the last 17 years and cost of FTE comprises approximately 90% of the total expenses. This means that the cost of production is almost completely driven by the efficiency of the research team.
The recommendations are very much in line with those stated in the European collaboration paper where it was suggested that clnical trials should preferably be performed in “European Centres for Clinical Research”, which should serve as “centres of excellence” (CAN MED report (1998)). It is noteworthy that over 2 decades after the publication of this consensus paper the UMCA strategy is referring to “centres of excellence” as a main goal to achieve a market leader position.
Clinical trial conduct is crucial to register any drug, and in the field of cardiovascular risk reduction, 1000s of patients have to be enrolled in clinical outcome trials (based on power-calculations, see above). As a consequence, many of the CROs and big pharmaceutical sponsors are seeking for novel CTU sites and India is considered to be a a potential booming country to conduct clinical trials. This is also driven by the fact that India (and the Far East) are considered “next level markets”, and local authorities do require test results obtained in the “local” population prior to drug approval. Currently the number of CTUs in India is rather limited, and the quality of the trial conduct does not reach the western standards in a number of these CTUs. In light of this, it is interesting to observe that Uma and co-authors described a “best-practice” framework for organizing a CTU in their manuscript “Best practices sharing: Setting up a professional clinical research unit in India” published in 2015. The authors summarize the prerequitsites for
conducting trials and concluded that “an effective organization is needed to develop and maintain a positive relationship among the sponsor, site and trial subjects.”
4) SCIENTIFIC OUTPUT GENERATED BY CTU’S IN THE DEPARTMENT OF INTERNAL MEDICINE
A large scale publication analysis was performed to quantify the relative contribution of clinical trials to the scientific output of the department of internal medicine.
In the overall aims for the UMCA merger, it was stated that “citations” and H factor were key determinants. The former means how many a given manuscript is cited by others, while the Hirsch factor (“h factor”) is an individual-author level metric quantifying the productivity and citation impact of a given author. (The h-index of an author means that the individual has published n papers, each of which has been cited in others at least n times (example: the author of this thesis has published173 publications, of which 37 have been cited at least 37 times; resulting in a H-factor of 37). Typically, the H-H-factor increases over time, irrespective of the actual production in recent times, and the H factor is therefore sometimes thought to be a mere reflection of time, and not as much on productivity. “Rising stars” are not identified by H-factor per se; it is the speed of increase in H-factor compared to peers that makes them “a talent”.
To measure productivity and success generated by CTU activity, the total number of publications by each individal medical doctor working at the department of internal medicine were retrieved. The number and impact of the papers were analyzed after stratification for CTU (read” “clinical trial related”) based-, and non CTU oriented activity.
This type of analysis is also not a valid reflection of the scientific output, since not all journals are equal. The relative importance/weight of journals is reflected by the impact factor (IF) of each journal, and therefore the total impact factor points derived by CTU versus non-CTU activity was also analyzed.
The results were obtained for a total of 69 specialists working at the DIVA (list of names and
subspecialties is available upon request) in the time period from 1-1-2014 until 31-08-2014. A total of 1491 publications were retrieved from www.pubmed.org for the department of internal
medicine of which 205 (13.7%) were based on clinical trial activity (fig 3A).
The total number of publications represented a cumulative impact factor score of 10193, of which 16.6% was derived from papers about clinical trials (fig 3B).
The net impact on IF of papers about clinical trials is larger then the impact on number of publications. Based on this observation one can conclude that the mean IF of a paper about a clinical trials is larger than the IF derived by a non clinical trial paper.
figure 3A) (left) relative distribution of publications generated by clinical trial (red) and non clinical trials (blue) figure 3B) (right) relative ditribution of IF derived by publications generated by clinical trial (red) and non clinical trials (blue)
5 EVALUATION OF THE STATUS OF THE CTU-VAS.
5.1 RATIONALE TO ANALYSE THE CTU-VAS
One of the initial steps on the roadmap towards a potential large merger between a number of CTUs was to analyze the CTU-VAS. The reasons to evaluate the working process and outcome of CTU-VAS (in comparison with other departments) were to:
1) investigate possibilities to generate more output by optimizing the work-process at the CTU-VAS. The overall aim of this optimization trajectory, which encompasses a number of aspects from the Lean-philosophy (appendix 1) was to become a growing and
learning (=healthy) organization
2) generate a “healthy” organization that would subsequently be a strong partner in a potential overall merger.
While analyzing the (scientific) output of all CTUs, insight is obtained as to which position the CTUs have at the sub-departmental level. The assumption is that departments with a large scienitific output generated by CTU based activity would match, since this is likely not only to reflect their focus, but also their success in terms of generating CTU-scientific output. This perspective of a
alignment of focus of potentially merging CTUs is provided by Epstein (2004), who described 5 key success drivers for post merger success.
These are: 1.Integration strategy 2.Integration team 3.Communication 4.Speed 5.Aligned Measures
Prior to the merger, aligned measures should be defined, and, as a consequence, this was the focus for the interviews described in chapter 7. The other 4 factors are of greater importance once the strategy for the merger is defined and a decission to merge is made. Epstein states that it is crucial to generate a coherent integration strategy that reinforces the ‘‘merger of equals’’ rather than an acquisition. This equality should be based on the parities being comparable to each other in terms of size, power, process and outcome measures (i.e. scientific output), that clearly define and articulate the drivers of success.
In this perspective it is eminent to create appropriate measures of success, and for the CTU-VAS this has been mainly classified as “scientific output”. Unlike for an acquisition or conglomerate, in a merger these aligned measures should be a balance between financial and non-financial
measures (see appendix 2). Moreover, “these measures should track revenue and cost synergies and comprise both process as well as result measures.”
Moreover, the operational practices, organization structure and culture was investigated for the CTU-VAS, in order to potentially come to a “true-north” for the CCTU. One should first learn about these aspects within the own organisation before trying to align and design a CCTU organization.
In Harvard Business review, Ron Ashkenas (2013) wrote about prerequisites in the process of an acquisition, a model that shows some similarity once the CTU-VAS would incorporate a very small sized CTU (i.e. 1 RN from a different department). It was stated that one should first analyze the own organisation and to subsequently “create a high-level picture of what you want a combined company ideally to look like one year after a successful integration — not just in terms of finances, but also in regard to operational practices, strategic initiatives, organizational structure, and culture. This thought-process will smoke out your assumptions about how much change you think the company needs. More importantly, it will give you a basis for dialogue with other managers
about their expectations for change, which might be different than yours. In fact, one of the reasons that integrations falter is the lack of alignment among managers about what will actually happen.” (quoted from Ashkenas (2013)).
5.2 CLINICAL TRIAL CONDUCT AT THE CTU-VAS
5.2.1 THE VALUE CHAIN
The process of trial conduct at the CTU-VAS was analyzed to identify crucial steps and, where deemed applicable, to optimize the process. This value chain was created by the whole team; we discussed all steps that have to be taken in the process from first contact with the CRO until the final close-out. The overall aim of the analysis was to increase the quality of the product and add value for the customer. Each team-member was asked what he or she perceived to be “the customer” for the service provided by the CTU-VAS and both “the patient” as well as “the sponsor” were mentioned. The sponsor is the main customer, given its role as the money-generating client, who asks for the service, which is, in broad terms “clinical trial conduct
according to pre-definied protocols within a given time frame”. The fact that “the patients” were considered “customers” reflects the product-based attitude of the team; the patient is, in fact, the product that is processed during the trial conduct. The value chain is depicted below (figure 4).
While analyzing the process it became clear that we suffered a considerable time delay, due to the fact that we were not able to control the pace of pre-initiation phase process of a trial. As a
consequence, planning of resources (FTE) was troublesome; a Gannt chart could not be made since the “document tuning, handling and submission” step was performed by an external
company, that charged the sponsor directly for this activity. We, however, were dependent on the quality and speed of this company and we noticed that this dependence had a large effect on the speed of initiation. Since “fast first enrollment” is one of the KPIs set by the sponsor, we hired a dedicated administrative worker, who processed all of the paperwork (no hand-offs; full control of the process in-house). This was the first “vertical merger”; we incorporated this part of the
process in-house. The “IRB submission service” was brought into the budget, and sponsors were found to be willing to pay for this service. In fact, by removing one external partner, the process indeed became more transparant and faster, and we observed that this “in house service” was much appreciated by the customers.
A number of steps in the process of the trial conduct were changed according to what the team decided to be “best practice”. By standardisation of a number of procedures, the overall level of stress (measured as a team defined important outcome) was found to decrease. The predictability of the workflow was increased, and by diminishing the number of potential “unpleasant surprises” stress was overall reduced.
5.2.2 THE “WHY?” QUESTION
In team meetings it became clear that the “why are we doing clinical trials?” question could not be answered by many of the team members. This is probably partly due to the fact that the CTU is a machinery bureacracy, and this structure is driven by the standard and strict procedures that have to be followed in clinical trial conduct. The pursuit for procedural excellence has likely resulted in waning of the realisation of the reason why these trials are being done. Simon Sinek claims that being able to answer the “why?” of doing things is a key to success (Simon Sinek (2011)). The other questions that most companies are very well capable of answering are the “how (procedure)” and “what (actual activity)?” questions. The inability of the team to answer the “why” question clearly exemplifies the disconnect of the team with the overall mission of the CTU-VAS, which is likely to be related to the fact that the moral values and strategic perspective of the
organization were not well embedded and not well comunicated by the head of the department/ PIs. As such, the overall mission of the department (“to conduct clinically meaningfull trials to enhance the quality of life for our patients”) should be brought back into the organizational design (see above; the Strikwerda (2014)) and alignment of the strategy with the motives of the team should be strived for.
5.2.3 THE 4P MODEL
The proposition of the customer (product to be delivered by the CTU: valid data with regards to a novel clinical intervention; a crucial factor in generating market share) has been discussed
previously. The CTU, however, is not driven by the market share of the sponsor; the rationale of the CTU to participate in a trial is driven by other incentives.
The PI and head of the department defined a number of outcome parameters (incentives) while analyzing the business propostion of a VAS. These parameters were simplified into the CTU-VAS we “4P model”. Unlike the common 4P model for marketing strategies (product, price, place and promotion), the CTU-VAS 4P model is based on the following parameters
1) Patient care 2) Publication 3) Payment
4) Planned trials with impact on 1-3)
The 4P model is used to quantify the added value of each clinical trial. The 4P model has been introduced to the CTU team in order to be transparent and consistent in communicating about the quantifation/relevance of these putative outcome measures. Once being approached for
participation in a trial, a collective decission making process is started where the final decission is made by the head of the CTU-VAS, while making an estimation of outcomes in these 4P areas. While prior to this company project the decission making was solely done by the head of
department and the PI(s), the team is now more involved in this process, which has resulted in a better understanding of the motives for participation by the overall CTU team. Moreover, the 4P model is discussed with the team responsible for a given trial, which not only sets the sense of urgency, but also provides the team with the “driver” for the CTU-VAS. Once a team is informed about the added value, culture will change into the direction of “ownership” of the process, which will ultimately result in pro-active changing behavior that will optimize the net outcome. It should
only on the Publication outcome measure will have a positive effect; it is the best advertisement for the CTU as a company; many sponsors select CTU-VAS for participation in clinical trials for this reason.
5.2.3.1 PATIENT CARE
The CTU-VAS participates in a large number of clinical trials for specific patient populations, such as patients with homozygous and heterozygous familial hypercholesterolemia, high lp(a) levels, low HDL-C, statin intolerance and venous thrombosis durign pregnancy. The focus on these types of clinical trials is driven by the observation of “clinical need”; additional therapies and
interventions are warranted for these patients. These diseases are considered relatively rare (mainly the first 3 categories) and/or require an intense and interdisciplinary approach for care. These characteristics make these disease to be classified as “eligible for referal to an academic hospital.”
Moreove, the PIs working at the deparment of vascular have been active in terms of generating international guidelines for these diseases and giving lectures on national and international fora. This has resulted in the fact that the AMC is now considered a “center of excellence” for these types of patients. The focus on these patient categories was a strategic choice and is in lign with the overall strategy of the AMC-VUmc to treat patients who are deemed elegible for referal to a “tertiary” hospital. The focus on particular patient categories, however, meant that the available resources had to be re-directed, and this clearly meant that a decision had to be made about what NOT to do (ie; care for diabetes patients is outsourced). This is in line with the statement by Porter about strategy; “Strategy is not as much a matter of making choices what to do, it is more about what not to do” (Porter (1979)).
The departmental strategy has resulted in a shift in referals to the outpatient clinic; we notice an increase of the number of referrals of the abovementioned types of patients, and the proportion of patients specifically referred for participation in a clinical trial is rising.
As mentioned before; we have to acknowledge that the patient is not “the client” in the value proposition; the end-user is the sponsor of the trial. Patients, however, should be treated as the most valuable asset within the value chain; the CTU-VAS is completely dependent on patients willing to participate in clinical trials.
The propostion of the patient has been grosly neglected, however, at the CTU-VAS. Participants in clinical trials have never been interviewed about their experience during and after the trial. In the near future a structured interview will be taken from all of the participants in clinical trials.
Moreover; educational activities will be planned for participants and their family members. It is clear that the largest threat for the CTU-VAS is the “resource of patients” and in 5.3 the consequences for low enrollment are described. An infrastructural change has to be made in order to increase the number of potential participants; the IT structure within the AMC and VUmc should enable recruitment at a wider scale (selection of patients by electronic case finding), and strategic alliances with other hospitals should be considered. To ensure current and future success in trial conduct it is warranted not only to assess which patient categories are considered to be of interest for future trials (a slow transition model), but also to consider whether clinical trials will be conducted by the same “face-to-face” trial conduct model in the (near) future. It might very well be the case that the business case for a CTU will change dramatically (shortly discussed in chapter 8)
Patients eligible for clinical trials are largely derived from AMC inpatient clinics. However, an increase in enrollment capacity can be generated by collaborations with other institutions where similar patient categories are being seen. (see figure 5)
Currently a geographical map of potential resources for putative participants has been made and collaborations with these providers are being explored.
The departments depicted in green are either already collaborating in the referal system, or have expressed their intent to do so. The partners in grey are potential partners in the exploration phase. Some partners are (considering) creating their own CTU; the CTU-VAS can provide knowledge to these partners and a reciprocal referal from the AMC-vascular medicine can be generated to CTUs focusing on a specific category of patients (ie IVUS trials in VUmc cardiology).
5.2.3.2 PUBLICATION
Scientific output could be one of the prime professional incentives for a PI to participate in a clinical trial, since PIs (co-)author many of the clinical trials. One of the KPIs for PIs in academia is “publishing (clinical) research”. Publications also have an effect on grant generating efficacy; the likelihood to generate financial resources for research is increased by an extensive publication track record. The number of (co) authorships is one of the key criteria in the evaluation process of personal grant applications. In other words: publishing studies will generate a professional
recognition and a possibility to generate funds for additional research.
The scientific output generated by the CTU-VAS and other departments within the DIVA was analyzed according to the method described in chapter 4. In short; studies published between 1-1-2014 and 31-08-2015 were retrieved from pubmed, (www.pubmed.org) on an individual medical staff member level. The Impact factors of these studies were used, since these are a well accepted reflection of the “weight of the publication”. Only studies of physicians were retrieved, under the assumption that this would enable identifying patient-related studies.
Aggregates were made on an individual MD and departmental level in order to identify “top trialists”. Studies were classified as “trial” once classified as such by pubmed, and/or were intervention studies in humans (randomisation design).
Analysis 1: Differences in mean IF per MD in each department.
In order to correct for the differences in FTEs per department mean IF output per MD per
department were calculated and shown in figure 6. A difference in output was observed; which is likely to be a reflection of the difference in strategic focus and tasks at a departmental level.
Figure 6) Mean IF per staff member
Analysis 2: Total IF generated vs IF derived from clinical trials.
The ratio between total IF : trial IF was calculated for each department (figure 7).
It can be appreciated form this figure that the largest total impact from clinical trial publications was generated by the department of vascular medicine.
Figure 7) IF total and IF from clinical trials
Analysis 3: Relative IF outcome derived by clinical trials
This analysis shows that the department of oncology generates almost one third of their
publication impact from clinical trials (figure 8). In absolute terms, however, the total IF generated by trials by the department of oncology is less compared to the total IF generated by the
Figure 8) IF by publications about clinical trials; proportion of total IF
Analysis 4: top trialists at an individual level
The total impact factor derived by clinical trials were analyzed at an individual MD level to identify “top trialists” among the total cohort of 69 MDs. The following MDs generated 100IF points or more with clinical trials in the period 1-1-2014 until 31-08-2015:
Prof J. Prins (229, “infectieziekten”), prof P. Reiss (125, “infectieziekten”), prof H. Buller (125, “vasculaire”), prof J. Kastelein (287, “vasculaire”), prof E. Stroes (100, “vasculaire”) and dr. G.K. Hovingh (202, “vasculaire”). The IF generated by prof dHaens (gastroenterology (GE)) was also evaluated, given the fact that he is heading the CTU at the GE department (total IF by clinical trials: 135). In the department of Cardiology at the AMC, 3 MDs generated more than 100 IF points by (co)authorships on papers about clinical trials.
Based on this analysis one can conclude that a limited number of individuals are generating a considerable IF output based on clinical trial conduct. Moreover, these individual PIs are not evenly distributed at the departmental level, which shows that there is a skewed distribution of PI outcome within the overall DIVA.
Analysis 5: IF per published paper.
The aim of a PI is to generate impact by means of scientific-publication output. In the previous analyses the focus was on the total IF. However, there is an interesting balance between quality and quantity of published work. Once being asked how to publish work, the overall consensus among PIs is that publishing one paper with a tremendous impact factor (ie Lancet; IF 39) on a relatively good authorship position will greatly outweigh a large number of publications with an IF
below 4 (ie 15 publications in the Neth J Med, IF 2.2). In other words; not only the total IF is important, but also the mean IF per paper published. In analysis #5 the mean IF per paper published was calculated for every department, and we evaluated whether more impact was generated by papers derived from clinical trials compared to papers published based on non-clinical-trial data. The mean IF per paper are depicted in table 1)
Department Mean IF per paper
generated by clinical trial
Mean IF per paper generated by non clinical
trial
Added value by Clinical trial conduct (delta IF per paper)
algemeen 3,6 8,5 no (-4,9) endocrinologie 4,4 6,1 no (-1,7) geriatrie* - - - hematologie 5,4 7,2 no (-1,8) infectiologie 8 6,7 yes (+1,3) nefrologie 4,5 4,2 yes (+0,3) oncologie 9 5,3 yes (+3,7) reumatologie 6,2 6,7 no (-0,5) vasculaire 15,4 9,0 yes (+6,4)
Table 1) Mean IF per paper from clinical trials vs non clinical trials in the departments of DIVA. (* number of papers published by department “geriatrie” was limited; no conclusions could be drawn)
The departments “infectie ziekten”, “nefrologie”, “oncologie” and “vasculaire” do generate more IF per paper by conducting clinical trials, while for the other departments a higher IF is generated by non-clinical trial related research. This difference might be a reflection of:
