OR WAIT 15 SECS
Some companies may be stretching their ERP systems and squandering opportunities for productivity gains.
Enterprise Resource Planning (ERP) systems—every manufacturing organization has one and each of them has invested a good deal of time and money implementing them. Introduced in the 1990s as an evolution of materials requirements planning (MRP) and manufacturing resources planning (MRP II) software, ERP systems were designed to integrate the material and resource management, distribution, and accounting activities; human resources; and financial management activities of the manufacturing enterprise.
PHOTOGRAPHY: EYEWIRE ILLUSTRATION: STEPHEN DUDASH
ERP systems are the latest in an evolution of information systems designed to maximize the productivity of manufacturing organizations by providing controls on costs and the tools to enable companies to better meet the needs and expectations of their customers. For the most part, ERP systems have been very successful in delivering on their promise: improving the commercial operations of small, medium, and large manufacturing companies. In fact, it is unlikely that any such business could exist today without an ERP system.
However, are some companies who own expensive ERP licenses trying to stretch their usability too far? Are these systems being asked to manage business processes that they just don't fit?
This article will outline the real gains that are being made in clinical supply chain management by using information technology tools from many sources and will explain why ERP software may not be the best tool to use to capture those gains.
Q: What is the primary business problem pharmaceutical companies are trying to solve today?
A: Increasing the speed of development while reducing the cost of bringing drugs to market.
This problem becomes more complicated when external factors are considered, factors that seem to work against each other. For example, the pressure from consumer groups and advocates to reduce prices is often out of sync with government policies, the rising cost of the labor pool, and changes in compliance procedures, which add time and cost to the drug approval cycle—contrary to the original intent.
Despite these factors, the business imperatives for pharmaceutical companies to reduce drug development costs and get drugs to market more quickly have been the same for several years. Meeting these imperatives becomes a primary focus and challenge for those faced with managing the clinical supply chain. Specifically, their challenges can be summed up as follows:
These are specific issues particular to the pharmacist, the quality managers, and the materials managers involved in managing supplies for the trial. As such, they need tools designed specifically to address these issues. Clinical trials supplies management and associated systems are built along a business process flow that addresses these and other specific needs, starting with the study design step and progressing through the multitude of functions through final drug accountability. ERP systems are not built to address these process needs, and for good reason.
ERP systems are designed to meet the needs of manufacturing/distribution companies across many industries, from semiconductors to industrial machinery, sneakers, beer, and snack foods. They are built to meet broad requirements for a set of independent commercial functions from industry to industry, and consequently provide mixed results in meeting the needs of any specific industry. ERP systems are designed primarily to ensure that the right amount of inventory is available for the timely fulfillment of changing customer demands—demands typically influenced by seasonality, promotions, and competitive activity.
As such, ERP demand management systems are driven by sophisticated forecasting algorithms that assist the production planners in arranging the manufacture of commercial products. A quick survey of several ERP vendor Web sites shows the business problems they address:
When considering life sciences companies specifically, these systems are designed to:
Although important and practical, these functions are general and for the most part not specific to any industry, let alone to a specialized function within an industry. Further, ERP has little to offer the product development function, as its primary focus is management of activities after the enterprise determines what products it is going to produce (i.e., it manages commercial production). For pharmaceutical companies, this means at the point after the drug is approved. It is at that stage—the commercial activity stage—that ERP systems can be extremely useful indeed.
Rob Pizzie, senior director, Schering-Plough Corporation, articulated it this way: "The difference between commercial drug supply management and clinical trials supplies management is that in commercial you design once and manufacture repeatedly. In clinical trials supplies, you design once and manufacture once, repeatedly."1
Pizzie's observation points to the underlying design of ERP systems. Generally, but not always, ERP systems are designed for repetitive manufacturing operations, where the same products are made over and over. Examples include automotive parts, consumer packaged foods, chemicals, and commercial pharmaceutical products. Clinical supplies management better fits a job-shop scenario, where each job is different from the previous one. ERP systems are not well-suited for this type of an environment.
On a more technical level, a close look at the data model of any of the leading ERP systems will reveal the sophistication of their design that appeals to multiple industries, but it will also highlight their shortcomings in terms of the specific data attributes required for the planning and production of patient drug kits for clinical trials. It is this issue that directly or indirectly translates into most of the challenges experienced by companies attempting to utilize their ERP system as the cornerstone of their clinical trials management function.
This will be discussed in detail later in the article. Let's first consider why companies are looking to their ERP systems to serve a broader purpose than their original design intended.
"We've licensed an ERP system. We should try to fully utilize it." Good point. Most big pharma companies have spent millions of dollars licensing their ERP system and many more implementing it. In most cases, their license allows them to extend the system with no additional license fees or for a modest upgrade fee. In addition, considerable money has been spent ($1.2 million on average2 ) in training users and IT teams on the use of their ERP system. Considering this, it makes good financial sense to use the software already licensed.
Secondly, the leading ERP systems have a successful history of implementation in the commercial pharmaceutical industry; so at the outset, the risk appears low. Furthermore, chief information officers (CIOs) will point to "IT standardization" as another reason—also a valid point.
Finally, and more recently, the powerful integration platforms offered by the leading ERP companies have become compelling reasons to look to the ERP system as the foundation for all the enterprise's business systems.
Financial investment, risk mitigation, standardization, and effective integration—when taken together (and excellent reasons on their own) make a formidable case for proceeding down the ERP path. However, it is worth looking at each of these reasons more closely and analyzing how they have played out with companies that have gone down that path.
At a recent conference for clinical trials supplies professionals, one group of speakers outlined their company's success in customizing their ERP system to suit their clinical supplies and electronic batch record (EBR) functions. Although the project sounded quite successful, it came at a high price.
The cost of modifying the system and implementing the functionality—primarily secondary packaging and EBR—without any licensing costs was quoted at close to $15 million. This is at least twice the cost of buying, implementing, and validating a comprehensive and automated clinical supplies management system, and it took much, much longer to achieve a workable result than a purpose built system would have taken. The difference is so great that it brings into question the fiscal benefits of staying with a legacy ERP system for supplies functions.
This is not to say that this company made a mistake. Given what was available in the way of fully integrated, off-the-shelf clinical supplies management systems at the time, they probably made the best choice they could. However, even in light of the expensive, protracted clinical supplies management projects based on ERP software, there are more recent examples of companies starting similar projects with their ERP system, only to see modest results, lengthy implementation times, significant expenses, and in the end nonvalidated systems. There is a cost for trying to push a square peg into a round hole that some companies do not recognize.
Staying with your ERP system for financial reasons may be valid in some cases, but a survey of the available purpose-built systems, their costs, and implementation track record is necessary in order to conduct a true financial comparison.
To pharmaceutical companies that fit the early pragmatic majority market described in Geoffrey Moore's Crossing the Chasm,3 risk mitigation is important when assessing new project decisions. Your ERP vendor is likely to be a large established company with revenues of several billion dollars and a large installed base of clients. Those are comforting characteristics. However, that is only part of the story.
Several hundred or even thousands of successful ERP implementations do not mean that these vendors have met with success in mission-critical, niche business activities such as managing the clinical supply chain. Working to mitigate this argument are the marketing divisions of large ERP vendors intent on increasing revenues from specific market segments. To help achieve their objectives, these groups engage in the creative marketing and repackaging of standard modules to appeal to specific segments—in this case, life sciences.
In many cases, the repackaging results in valid, useful solutions for specific business functions; in other cases, it does not. The reason for this is the potential impact that the changes required to deliver specific functionality would have on the standard ERP product. The example here is the building, shipping, and tracking of patient kits, randomized and blinded for specific trials. Building that logic into an ERP system has massive rewrite implications on the data model and software design. The obstacles and costs of such a rewrite are usually prohibitive for what is a tiny, niche market to the ERP vendor.
To properly assess the risk of going down either the ERP or purpose-built solutions path, one should obtain a thorough understanding of the requirements of the business process owner, examine the track record of the prospective vendors in that specific application niche, and ask probing questions about timing, cost, results obtained, and lessons learned.
The argument that the large company, both vendor and customer, will not let the project fail for want of resources or money is also commonly heard. But beyond the money spent on a custom project, there is an opportunity cost involved with the additional time taken to get to a point of success over and above the money spent on saving the project. The total cost, including opportunity cost, can be substantial indeed.
All things being equal, IT standardization is an excellent goal. Systems that can be upgraded simultaneously or that are familiar to users, administrators, and IT support teams are far more preferable than an IT environment made up of disparate, isolated applications with different support mechanisms. The caveat remains: all things being equal.
If, however, your standard ERP system meets only 50% of your clinical trials business requirements, then standardization becomes far less important. In addition, IT applications have come a long way in the past few years in terms of adhering to formal or de-facto standards that have commonly emerged, and that make disparate systems a lot easier to work with. For example, it is almost mandatory for any software sold into the pharmaceutical industry to work with Oracle and/or SQL databases. All Web-based applications will adhere to XML for message transfer protocols. These are well-established standards in the pharmaceutical industry.
Beyond this, some ERP providers have opened their application integration platforms to enable easier integration of many different applications (e.g., SAP's NetWeaver). This means that companies can implement special-purpose, cost-effective independent applications that talk to each other while still adhering to the underlying IT standards of the enterprise. So staying with a less-than-ideal software application solely for reasons of standardization no longer needs to be a consideration.
As indicated in the discussion on standardization, integration of IT applications has come a long way in the past few years. Integration platforms inherent in ERP systems or made available by specialized integration vendors are ubiquitous in most sizeable IT shops today. This means that the "best-of-breed" approach to niche solutions is now more viable than ever. This is key when looking at the business process environment of the extended clinical supply chain.
As a result, if you are using an integration platform, you can choose between one of your ERP vendor's modules for a particular function or another vendor's solution if that solution fits your needs better—and you can integrate it using the ERP integration platform already in place. Unless there is an excellent fit with your ERP vendor's proposed software module, there is no longer any advantage in selecting that solution for every business process in your organization. Most ERP vendors have embraced this concept, accepting that they will promote co-existence with applications where their solutions are not a strong fit or where they do not have solutions to offer.
Although the initial arguments outlined here seem to be good reasons to go down the ERP path, the business of clinical trials supplies management is driven by pharmacists and clinical experts who know their complex business needs very well. Therefore, to rely solely on IT teams and not include these experts in key decisions about the selection of systems used to manage the clinical trials supply chain is to invite significant implementation and acceptance challenges.
As mentioned, many of the common and necessary functions required for the effective and compliant management of clinical supplies are not accommodated by ERP systems in their standard form. To provide these functions, significant data schema changes and customized logic would need to be undertaken—changes that could potentially have substantial consequences on the parent product in terms of rewrite, updates to the whole ERP client base, and the update effort.
Further, pharmaceutical companies make up less than 2% of the manufacturing/distribution companies4 constituting the ERP suppliers' target market. The clinical trials supply chain is an even smaller market within this niche segment. Simply put, the complexity and complications involved in providing all the mandatory functions required for clinical trials supply management is simply a nonstarter for ERP vendors.
An example of such a complex, mandatory design feature is the availability of an integrated kit and patient randomization function and the inherent structure needed to accept randomization from other established, approved systems such as SAS or Oracle. Applying randomization and maintaining blinding through all aspects of the function touches many parts of the supply management system. These are not features found in off-the-shelf ERP systems. Any software system that purports to meet the needs of the clinical supplies function that does not have the structure to provide or accommodate randomizations falls far short of the needs of the business.
Another example includes a study where dosage or strength changes during the trial. This type of study becomes very difficult to accommodate while maintaining ease of traceability and reporting capability associated with the trial when using the standard ERP approach of managing inventory as tightly specified stock keeping units (SKUs). Specialized systems are required for this type of study and other complex study designs, including maintaining traceability down to the blister-well level throughout the trial.
These clinical trial specific features, or the absence of them, can be very problematic. Michael A. Arnold, senior director, supply chain logistics, Pfizer Inc, explains:
"ERP systems just don't have what we refer to as 'protocol aware' capabilities, which are essential to monitoring and tracking clinical supplies at the patient level. After numerous attempts to retrofit ERP systems to clinical supplies, it became clear that the time, effort, and cost associated with ERP software redesign were too prohibitive to the clinical supply organization and unattractive to the ERP software companies, whose primary focus was on serving the broader needs of commercial manufacturing and distribution processes. Going down the specialized systems path, we are well on our way to globally automating clinical supply activities for all phases of development for clinical studies."5
Technically, there is no reason why an ERP provider cannot build systems specific to the needs of the clinical supply functions of the enterprise. There are cases where in fact this is happening. But these subsystems will likely be independent of and loosely coupled with the main ERP product rather than configured into the main product. In essence they will be a related but separate product from the ERP core.
Nevertheless, the approach may be an attractive one for some companies for whom the advantage of dealing with a single vendor is important. In these cases, however, it is still prudent to compare any offered customized solution in terms of functionality, full cost, and time to implement against the proposals from established, purpose-built system vendors.
The validation of systems also has implications for systems managing the transactions involved with clinical trials. Validation of these types of systems is likely the most important, the most involved, and the most costly of all the implementation functions. There are companies that have built custom systems that remain nonvalidated after several years, and yet these companies continue down their chosen path.
Validation scripts and methodologies should be part of any system built or purchased. When effectively designed, they have reduced the cost and reduced the amount of effort associated with validation from many months to a number of weeks, substantially decreasing those systems time-to-value. Such a specialized, focused validation approach is very doable with special-purpose systems designed for clinical supply chain management processes.
Michael Dragoon, manager, worldwide clinical supplies, Procter & Gamble Pharmaceuticals, sheds light on the future:
"With the ever-increasing complexity and global reach of clinical trials, the days where companies can support individual stand-alone information and material management systems are numbered. Future successful companies will find ways to connect these mission-critical systems between Clinical Operations, Clinical Supplies, and Service Providers for streamlined clinical trial management. Technology and business need to finally intersect, providing companies with awesome possibilities to improve efficiencies while also reducing stress."6
Until this point, the discussion has been concerned with identifying where ERP systems apply in an organization. Clearly, they are not aimed primarily at incorporating all of the internal and external parties and functions involved in clinical trial management. The many companies involved in the multiple functions of clinical trial management include labeling companies, packaging contractors, and interactive voice recognition (IVR) companies.
The extended supply chain, depicted in Figure 1, extends beyond the sponsor to the investigative sites; the manufacturing, packaging, and distribution contractors; the IVR vendors; and more. Currently, much of the work in this area involves the collaboration of pharmaceutical companies, supply chain software developers, contract research organizations (CROs), IVR vendors, and clinical trial management systems (CTMS) vendors.
Figure 1. The parties involved in the extended clinical supply chain.
The goal is to automate the information flow and displace paper-based systems. The sponsor is given full visibility of the status of supplies and changes in requirements of the study regardless of where the drug, kits, or information resides. To understand the challenges in achieving this goal, it helps to look at the process flow.
There may or may not be external parties involved, but in the case of most pharmaceuticals today, there is a growing level of outsourcing for some clinical trial management work. Following is a description of a small subset of the external transactions that may be involved and how they are managed between various systems.
Using the forecast (driven by the study design), work orders and shipment orders can be created and routed internally, or through a secure messaging data exchange or portal to the internal systems of external contractors where CROs are being used. By means of the contractor messaging back to the sponsor, the sponsor can track exactly where the drugs are in the supply chain, either in real time or near real time.
Site replenishment as a drug is dispensed, which is managed by the IVR system, is also integrated in real time to the supply management system. The manual interface between the IVR message and the picking and dispatching of the kit can now be eliminated. The request simply feeds into the supply management system and the required kits appear on a pick list, all tracked in an audit trail.
Real-time updates to approvals or withdrawal of approvals of sites or investigators are also needed. Integration with CTMS such as Perceptive's IMPACT provide that real-time visibility. If a site becomes unapproved and that status is entered into the CTMS, the integration automatically and immediately updates the status in the supply management system; the logic of the supply management system then prevents shipments to the unapproved site.
The need for visibility beyond the walls of the enterprise is increasingly being raised by more and more pharmaceutical companies. And with the use of standardized message exchange and portal technology, it is becoming a reality. As many pharmaceutical companies continue to expand their drug development outsourcing activities, the need for control and security of their drug supply outside their walls is growing.
In response to this growing demand for control and visibility to data, no matter where it resides, an open system data exchange to facilitate specific message transfers between sponsors and their CROs, IVR companies, and Investigators and to automate the complete clinical trials management process is now in the early stages of implementation. Figure 2 outlines this data exchange. A few pharmaceutical companies have committed to pioneering the concept and are bringing their partner CROs and IVR vendors into their development activities.
Figure 2. The data exchange engine & portal identifying the external players.
A growing number of pharmaceutical companies are recognizing the gains to be made in managing the complete clinical trials function not by looking for a single system that claims to do it all, but by pulling together specialized systems of their extended clinical supply chain, each with their own niche focus and domain expertise, and opting for an open-system integration approach. It is in this area, an integrated solution to the entire clinical trials process, where some of the special-purpose system vendors and their clientele are investing.
With the small size of the life sciences submarket and the focus of ERP research and development dollars on horizontal applications that cross multiple industries, there is a low probability that the ERP vendors will make any significant investment in building the special-purpose functionality required and integrate it into the broader ERP platform.
As the specialist companies continue to invest in pulling the players in the extended clinical supply chain together, the gains will be many and will come from unexpected places.
One pharmaceutical company that has implemented a sequencing feature for distribution through a contractor has found annual savings in excess of $1 million in distribution charges each year—savings that go a long way toward a significant ROI for their systems.
But this is only half the story. Special purpose, integrated, task-specific solutions today are enabling the streamlining of functions involved in managing clinical trials from several weeks to only days. It is in these time savings that those following this approach will realize their biggest gains.
ERP systems play a key role in the commercial operations of large pharmaceutical companies today, but by overextending these systems, companies are almost certainly missing opportunities to achieve considerable gains in the clinical trials management process—gains made available by specialized systems providers serving this complex submarket.
Greg Turner, BE, is senior director of sales and business development with Aptuit Informatics, 29219 Canwood Street, Suite 120, Agoura Hills, CA 91301, (978) 809-0286, email: Greg.Turner@aptuit.com
R. Pizzie, Schering-Plough Corporation, IIR Conference, Las Vegas, NV, January 2005.
2. Nucleus Research, "ROI Comparisons Report: Portals," research note D85, 2003.
3. G. Moore, Crossing the Chasm (HarperCollins, New York, 2001).
4. U.S. Department of Labor, SIC Division Structure.
5. M.A. Arnold, Pfizer, Interview with Author, July 2005.
6. M.G. Dragoon, Procter & Gamble Pharmaceuticals, Interview with Author, August 2005.