MD+DI Online is part of the Informa Markets Division of Informa PLC

This site is operated by a business or businesses owned by Informa PLC and all copyright resides with them. Informa PLC's registered office is 5 Howick Place, London SW1P 1WG. Registered in England and Wales. Number 8860726.

Remembering Thomas Burnett Jr.

Originally Published MDDI October 2001


For all of us, the events of September 11, 2001, were deeply personal, regardless of whether we personally knew any of those who died on that day. We at MD&DI mourn the still uncounted thousands who perished, and honor their memory.

But like many in the medical device industry, we have been touched in particular by one of those thousands, Thomas E. Burnett Jr. His story by now is well known. The senior vice president and COO of medical device company Thoratec (Pleasanton, CA), he was coming home that day on United flight 93 to San Francisco. When the plane was hijacked, he, along with Jeremy Glick, Todd Beamer, and other passengers, decided to take action. From phone calls they and others made in those final minutes, we know that they intended to take on the hijackers. Although the plane crashed, it is nearly certain that the actions of these passengers saved many more lives in Washington, DC.

Burnett had spent more than 15 years working in the medical device industry, beginning in 1985 at McGaw Laboratories in Irvine, CA. He subsequently worked at Calcitek (Carlsbad, CA), ultimately as vice president of sales and marketing, before joining Thoratec in 1996.

As remembered by Thoratec president and CEO Keith Grossman, a longtime friend as well as a colleague, Burnett was "well on his way to a truly brilliant career." Like so many others whose lives were cut short by the heartless acts of madmen, Burnett's career potential is now a matter of speculation. But his place in history as one of the heroes of that terrible day is now secure.

Tom Burnett is survived by his wife and three young daughters. To help his family and to honor his memory, Thoratec has established a memorial fund. Donations can be made payable to:

The Thomas E. Burnett Jr. Family
Memorial Fund
C/O CIBC Oppenheimer Corp.
Account #074-17387-10
580 California St., Ste. 2300
San Francisco, CA 94104

Although we never met Tom Burnett, we have come to feel as if we know him now. To paraphrase one of his heroes, Abraham Lincoln, the world will not long remember what we say on this page, but it can never forget what Thomas Burnett and his fellow passengers did on Flight 93. He did not die in vain.

The Editors

Copyright ©2001 Medical Device & Diagnostic Industry

Corporation announces expansion, new focus

. The Parker Hannifin Seal Group (Irvine, CA; 949/833-3000) has announced its new focus on engineered sealing systems following a major expansion and restructuring. The Seal Group nearly doubled in size after acquiring the former Wynn's International Inc. in July of last year. It integrated Wynn's businesses, including Wynn's Precision Inc., Goshen Rubber Co., and Waukesha Rubber Co. into its existing organization. In the process, four new divisions were created, and three existing divisions were renamed. The new North American Seal Group comprises 10 divisions across the United States, from San Diego to Boston. The Group's four European, three Asian, and two Latin American locations have also been realigned to accommodate changes brought about by the acquisition.

Is ISO 9001 Obsolete?

Originally Published MDDI October 2001

With the publication of the 2000 revision of the general ISO quality systems standard, the medical device industry has concluded that it must follow a separate course. A pending update to the industry's own ISO standard, 13485, will formalize the split.

Edward R. Kimmelman

Medical device companies are confronted by a variety of relevant quality systems, standards intended to satisfy manufacturing requirements in addition to customer and regulatory agency objectives. These standards are embodied in regulations, such as the U.S. quality system regulation and the Japanese good manufacturing practices regulation, as well as international standards (e.g., ISO 9001 and ISO 13485). International standards may have been incorporated directly into national regulations or adopted voluntarily by medical device manufacturing organizations.

Manufacturers may face a dilemma, however, in adapting to a changing standards environment. ISO 9001:2000 has instituted significant changes and is not viewed by all as an appropriate foundation for a medical device quality system management standard. Thus, companies may find it advantageous to prepare now to adapt their quality management strategies to evolving requirements.

This article outlines medical device quality system issues created by ISO 9001:2000 and recommends courses of action for individual organizations as they review and possibly revise their quality management systems to meet future requirements.


The mandate of ISO Technical Committee 210, Working Group 1 (ISO/TC 210, WG 1) is to develop international voluntary quality systems standards for the medical device sector. Thus far it has developed three standards:

  • ISO 13485:1996, "Quality systems—Medical Devices—Particular requirements for the application of ISO 9001."
  • ISO 13488:1996, "Quality systems—Medical Devices—Particular requirements for the application of ISO 9002."
  • ISO 14969:1999, "Quality systems—Medical Devices—Guidance on the application of ISO 13485 and ISO 13488."

The titles of these standards reflect the fact that their content and format are based directly on the 1994 versions of the ISO 9000 series of standards. For example, essentially all of the requirements of ISO 9001:1994 are carried over to ISO 13485:1996, with the addition of particular requirements that are relevant to all medical devices or to some classes of medical devices. Many of these particular requirements come directly from existing regulations. WG 1 continues to review and update these standards, based on the changing needs of the medical device sector.

The Global Harmonization Task Force (GHTF) includes representatives from medical device regulatory agencies in many of the industrialized countries, along with representatives from industry. Study Group 3 (SG3) of the GHTF has succeeded in harmonizing the regulatory quality system requirements in the major markets around the world and is working to extend this achievement to other countries.

ISO 13485:1996 is the basis for the GHTF agreement on quality system requirements. Although the quality system regulations in the individual GHTF countries may not follow the format of ISO 13485:1996, the substance of the requirements is consistent with that found in this standard.

In 2000, ISO Technical Committee 176 published a revised version of the ISO 9000 series of standards. This revision included a number of key elements.

  • Adoption of the "process approach" to quality management.
  • Elimination of ISO 9002 and ISO 9003.
  • Refocusing of the primary objectives of ISO 9001 and ISO 9004.
  • Strengthening of requirements within ISO 9001 related to customer satisfaction and continual improvement.
  • Elimination of ISO 8402.

Adopting the "process approach" to quality management resulted in the formatting of ISO 9001:2000 to reflect the key quality system processes (i.e., management, resources, product realization, and measurement). Within these major process areas are subprocesses dealing with quality system management planning, management review, human resources, work environment, design and development, purchasing, production, monitoring and measurement, analysis of data, improvement, and others.

ISO 9002 and ISO 9003 were eliminated through the use of an applications approach (ISO 9001:2000, clause 1.2) that allows an organization to disregard quality system requirements for product realization processes it does not perform (such as design and development).

The refocusing of the objectives of ISO 9001 enabled ISO 9001:2000 to be targeted at a quality management system intended to meet customer requirements. At the same time, ISO 9004:2000 is focused on quality system recommendations for business excellence.

Within ISO 9001:2000, the requirements related to customer satisfaction and continual improvement are stronger than in the previous verison, and requirements related to procedural documentation are weaker. ISO 8402 was eliminated by including relevant quality system definitions in ISO 9000:2000.

Representatives of both ISO/TC 210, WG1 and GHTF, SG3 participated in a number of joint working sessions to consider ISO 9001:2000. They concluded it is no longer an appropriate basis for a medical device quality system standard that is intended to support existing international harmonization of quality system regulation and to be a model for countries developing such regulations.

According to ISO Central Secretariat (ISO/CS) policy, ISO/TC 176 has the overall responsibility to manage the development of quality system standards. The ISO/CS, however, has recognized that the medical device sector, owing to its heavy regulation worldwide, may require the establishment of standards that are targeted specifically at the sector. As a result, the ISO/CS approved the creation of ISO/TC 210 to manage the development of medical device sector–specific documents, with the understanding that ISO/TC 176 would still maintain primacy related to quality system standards in general. One of the key overall objectives of ISO/TC 176 is to avoid the proliferation of sector-specific quality system standards.

Over the last six years there has been direct interaction between the two committees as each has developed its own standards. There is now general agreement that the needs of the medical device sector would be best served by a separate quality system standard based generally on ISO 9001:2000. Some of the ISO 9001:2000 requirements would be removed, however, and a number of requirements would be added. This separate standard will clearly reflect the differences required by the fact that the medical device sector is highly regulated. As a result, the new ISO 13485:200X will bear the title, "Quality management systems—Medical devices—System requirements for regulatory purposes."


The substantive differences between ISO 9001:2000 and ISO 13485:200X will result from the divergence of the objectives of the two standards. ISO/TC 210 seeks to maintain ISO 13485 as an easily understood, baseline standard targeted at meeting customer requirements and maximizing the probability that compliant organizations will produce safe and effective products. Although ISO/TC 176 professes to share that objective, it is obvious from the requirements that have been added that TC 176 intends to move the ISO 9000 standards closer to the business excellence standards epitomized by quality award systems like the Deming and Baldrige models. TC 176 also seeks to simplify ISO 9001:2000 compliance for smaller organizations by reducing the procedural documentation requirements.

It is important to point out that there will be a great measure of agreement between ISO 9001:2000 and ISO 13485:200X. In the current draft, as much as 80% of the actual requirements text of ISO 13485:200X is quoted directly from ISO 9001:2000. In addition, the format of ISO13485:200X is the same. Specifically, it is based on the process model. This model does not significantly change the actual requirements; however, ISO/TC 210 found it to be very useful in explaining the relevance of the requirements because most organizational operations are based on a compilation of processes.

The major substantive differences between ISO 9001:2000 and the committee draft (CD) of ISO 13485:200X relate to continual improvement, customer satisfaction, and procedural documentation level. The key issue related to continual improvement is that ISO 9001:2000 requires actual objective evidence of continual improvement of the quality management system, not just objective evidence of the exercise of quality management processes targeted at determining needed continual improvement and the effective actions based on such determinations.

Clause 8.5.1 of ISO 9001:2000 states that "the organization shall continually improve the effectiveness of the quality management system through the use of the quality policy, quality objectives, audit results, analysis of data, corrective and preventive actions, and management review" (emphasis added). Clause 8.5.1 of ISO/CD13485:200X begins as follows: "The organization shall identify and implement any changes necessary to ensure and maintain the continued suitability and effectiveness of the quality management system through the use of the quality policy. . . ."

It is important to note that the processes anticipated by both documents to be used to achieve continual improvement are processes that are familiar to medical device organizations because they are required by most quality system regulations. These examples were included in response to comments made by ISO/TC210 during the drafting of ISO 9001:2000.

In early drafts of ISO 9001:2000, clause 8.5.1 also targeted a requirement for improvement of quality system efficiency, but the requirement was removed at the urging of ISO/TC210, because efficiency is more related to organizational excellence than it is to meeting customer requirements.

The requirements in ISO 9001:2000 present a number of issues for medical device organizations. One key issue is that ISO 9001:2000 requires objective evidence that there are active processes for determining whether customers are, in fact, satisfied. During the development of ISO 9001:2000, representatives of ISO/TC 210 convinced ISO/TC 176 to narrow the definition of customer satisfaction to focus it on whether or not an organization is meeting customer requirements. Unfortunately, even this narrowing of the definition does not overcome the concern that there are widely divergent interpretations among professionals assessing quality management systems as to what customer satisfaction really means. In addition, the representatives from ISO/TC210 were not able to get ISO/TC176 to eliminate the use of the term customer perception in clause 8.2.1.

ISO/TC 210 believes that the use of the term customer satisfaction is not appropriate for ISO13485:200X, because it is too subjective to use in a standard that will provide the basis for regulation. As a result, ISO 13485:200X will encourage the use of active processes for gathering customer feedback related to whether or not the organization is meeting customer requirements. It will also require the generation of objective evidence that these systems are being used and the feedback acted upon. Examples of such processes given in ISO 13485:200X include complaint handling, order handling, and contract review.

The key documentation issue is that ISO 9001:2000 has reduced the number of processes that require documented procedures. It specifically spells out a limited number of processes that require documented procedures, and it has left it up to the individual device manufacturer to determine whether procedures are needed for all other quality management processes.

In the regulated world of medical devices, documented procedures are used as objective evidence of control of key quality management system processes. It is inappropriate to allow the flexibility embodied in ISO 9001:2000, especially for processes that could significantly affect the safety and effectiveness of products. As a result, ISO 13485:200X will retain the same documented procedure requirements as in the 1996 version of this standard.


ISO 13485:200X will be a stand-alone standard; it will not make direct references on a section-by-section basis to ISO 9001 as the current version of ISO 13485 does. This approach will go a long way toward disconnecting the two standards and make ISO 13485 less susceptible to future changes in the ISO 9001 standard.

As mentioned above, ISO 13485:200X will quote extensively from ISO 9001:2000. As required by ISO policy, these quoted sections will be designated by a distinctive font or font treatment. For particularly important additions or deletions a note will be added to the ISO 13485:200X clause text, pointing out the difference. If the change requires explanation, that information will be provided in an annex to the standard.

ISO 13488:1996 is the medical device equivalent of the ISO 9002 standard. ISO 9002 was intended to provide quality management system requirements for organizations that performed all activities except design and development of products and the processes for providing them. As indicated above, ISO/TC176 has decided to eliminate ISO 9002.

ISO/TC 210 is left with the decision as to whether or not to extend the viability of ISO 13488, or to follow the lead of ISO/TC 176 and eliminate it. At the present time, there are still plans to publish ISO 13488:200X because a number of medical device regulatory schemes in Europe and North America rely on the existence of separate standards.

The European Community has developed an approach for applying the new version of ISO 9001 to compliance with the conformity assessment requirements of the various new approach directives. They have documented that approach in the foreword of EN/ISO 9001:2000.

WG1 has issued a briefing paper to all participating members of ISO/TC 210, explaining the situation and soliciting guidance as to the appropriate path forward. That decision will be made at the upcoming meeting of WG 1 this month.

ISO 14969:1999 is a standard that provides guidance on the application of ISO 13485. It is based on and refers directly to ISO 9000-2, a guidance document published by ISO/TC 176. Because ISO/TC 176 has decided to eliminate ISO 9000-2, ISO 14969 will be republished as a stand-alone technical specification following the same organizational structure as ISO13485:200X.

Technical specifications are a new kind of ISO document that require significantly less consensus review and can be published more quickly. ISO 14969:200X will contain much of the same guidance language as in the 1999 version of this document, and it also will contain more detailed guidance related to quality planning, design and development, and process validation.

Figure 1 provides a simple time line that points out the quality system transition. ISO 9001:2000 was published in December 2000 (A),with the guidance that the 1994 version of the standard would remain viable until December 2003 (B). Because ISO 13485:1996 refers directly to ISO 9001:1994, WG 1 of ISO/TC 210 is under time pressure to publish the approved new version of ISO 13485 by the end of 2002 or the first quarter of 2003. WG 1 published the CD version of ISO 13485:200X in June 2001 (C) and will consider comments on this draft at its meeting in October 2001 (D). It is anticipated that the draft international standard (DIS) version of ISO 13485:200X will be published during the first quarter of 2002 (E).

Figure 1. Quality system management milestones. (A) publication of ISO 9001:2000; (B) ISO 2001:1994 remains viable until December 2003; (C) committee draft (CD) of ISO 13485:200X is published; (D) comments on CD version of ISO 13485:200X are considered; (E) DIS version of ISO 13485:200X, is published; (F) organizations focus on compliance with ISO 13485:200X, ISO 9001:2000, or both; (G) organizations negotiate with current registrars or notified bodies; (H) organizations internalize the process approach; and (I) DIS provides the basis for analysis of current quality management systems.

Such a publication schedule will enable organizations claiming compliance with ISO 13485 to understand the requirements of the new version in time to inform their employees and make any necessary modifications to their quality management systems before the 1994 version of ISO 9001 disappears. This timing will also allow third-party registrars and notified bodies to adjust their assessment procedures, and to arrange assessment schedules to avoid lapses in registrations and ensure an orderly transition.


It will be important for individual organizations to use their management review processes to discuss and determine their quality management system standards compliance objectives. Should they work for compliance with ISO 9001:2000, ISO 13485:200X, or both (F)?

In the absence of any customer or regulatory requirements to comply with ISO 9001, medical device organizations should seek registration to ISO 13485:200X only. Such registration will provide objective evidence of compliance with quality management system requirements consistent with meeting customer requirements and those of the major regulatory agencies around the world. An organization may choose to adopt some of the customer satisfaction and continual improvement requirements of ISO 9001 because it believes that to be a good business decision. If there are no customer or regulatory reasons for registration to this standard, however, it seems unwise to subject the organization to the trauma of third-party assessment against these requirements.

It is true that some customers and a number of regulatory agencies in smaller countries are not aware of ISO 13485 and may require or request compliance with ISO 9001:2000. Although the GHTF is doing its best to publicize the value of ISO 13485 to regulatory agencies around the world, it might be necessary for individual organizations to educate their customers and the regulatory agencies with which they interact. These organizations should avail themselves of materials that are available from the GHTF and the Association for the Advancement of Medical Instrumentation, which administers the ISO Secretariat for ISO/TC 210 (see below).

It will be important for each organization to negotiate with its current registrar or notified body and with others that may offer programs that are more consistent with the organization's objectives (G). The transition from the 1994 to the 2000 versions of ISO 9001 will likely cause increased demand for quality management system assessments and may lengthen each of these assessments. Getting on a registrar's schedule may be difficult if an organization delays this action.

If an organization seeks registration to ISO 13485:200X alone, it will be necessary to determine if the registrar is or plans to be qualified for ISO 13485:200X for the relevant medical devices. It may be necessary to help the registrar gather the information it requires to plan for assessments and, if necessary, to educate its assessors.

If an organization seeks registration to both ISO 9001:2000 and ISO 13485:200X, it will also be necessary to negotiate with the registrar about the process of assessing for both standards and the costs associated with this process. It will be necessary to determine if dual registration will require more than one assessment or other significant additional costs.

One of the things an individual organization can do immediately is to internalize the process approach by reviewing it with top management, developing training programs for affected personnel, and beginning to modify its high-level documentation to reflect this approach (H). Because both ISO 9001:2000 and ISO 13485:200X have adopted the process approach, there is little chance that such efforts will be wasted.

Once the ISO/CS has published the DIS version of ISO 13485:200X, it is unlikely that significant substantive changes will be made to this standard. At that point, the DIS should be satisfactory to use as the basis for an analysis of an individual organization's current quality management system against the revised requirements (I). One of the objectives of ISO 13485:200X is to maintain the status quo with regard to requirements. It is, therefore, not likely that an analysis of a compliant quality management system will reveal many gaps. In any case, starting this process during the second quarter of 2002 should provide sufficient time for corrective and preventive actions to be taken before the end of 2003.

If an individual organization seeks compliance with ISO 9001:2000, it should start its gap analysis immediately. Such an analysis may uncover deficiencies related to customer satisfaction and continual improvement processes.

The key is for medical device manufacturers to start their quality systems strategy efforts right away, keeping in mind what they learn as they follow the development process for ISO 13485:200X. That should leave enough time for each organization to effect an orderly and timely quality management system transition.


Global Harmonization Task Force:

Association for the Advancement of Medical Instrumentation (AAMI):; Hillary Woehrle, Secretary ISO/TC 210; e-mail,

Copyright ©2001 Medical Device & Diagnostic Industry

Hydrophilic coatings offered for catheters and guidewires.

Two new solvent-free coatings have been developed for manufacturers of medical devices. The coatings, produced by Biocoat Inc. (Fort Washington, PA; 215/654-1377) are based on emulsion polymer technology and adhere well to stainless steel and other metallic and polymeric substrates. Because the coatings contain no organic solvents, they can be applied to plasticized PVC and other polycarbonate substrates normally prone to solvent erosion. In addition, these coatings are lubricious, durable, and biocompatible. They allow easy cleanup, are not flammable, and do not require high-temperature curing primers. Bicoat's Hydak coatings are based on hyaluronan, a biopolymer naturally present in all human tissues and biofluids. As a result, surgeons reportedly see improved performance characteristics when using devices coated with Hydak products.

Using Measurement to Improve Quality

Originally Published MDDI October 2001

The new ISO 9001:2000 standard puts an unprecedented emphasis on measurement. Although companies may find compliance challenging, they will benefit from improved quality systems.

Daniel P. Olivier and Paschal Dwane

With the official publication in December 2000 of the latest version of the ISO 9001 quality system standard (ISO 9001:2000), an unprecedented number of changes were introduced. Among the most significant of these is the process approach to the definition of quality systems.1 In this approach, an organization is defined as a system of processes, and quality management consists of identifying, analyzing, and controlling these processes and their interactions. This approach also emphasizes the understanding of requirements, the need to assess the value of processes, and the measurement of the processes to support evaluation and continual improvement.

The concept of processes is familiar to most companies through the use of flow chart techniques, among other things. The concept of measuring processes, however, gives rise to questions that are new to many. What processes should be measured? What measurements should be collected? Are there industry standards for measurement? To what extent should the collection of these measurements be automated?

Analysis of quality processes is critical for process optimization and determining the effectiveness of process changes. In the absence of appropriate measurement, attempts at process improvement often fail. Effective measurement is essential to ensuring that any process changes introduced realize the desired quality and productivity benefits.


Measurement was not an essential component of the ISO 9001:1994 standard. Why has it become one in ISO 9001:2000? The answer involves the process approach to quality systems. ISO 9001:1994 was somewhat prescriptive regarding the definition of processes, specifying 20 subelements. But in ISO 9001:2000, manufacturers are responsible for defining their own processes. Therefore they must also define their own standards for assuring that those processes generate quality products efficiently.

Along with this responsibility to define internal processes comes a need to define measurements that ensure the effectiveness of those processes. Measurements of processes are essential to ensure the final quality of the product. In other words, to the extent that design and manufacturing processes have a low incidence of errors, the finished product is likely to be of higher quality. Measurements can also identify process inefficiencies such as scrap rate, effort devoted to rework, and excessive cycle times. Each of these measurements also provides insight into business efficiency and profitability.


Figure 1. A hierarchy of process and product measures associated with the implementation and refinement of a measurement system.

The science of measurement is not well established for most organizations. Figure 1 outlines a hierarchy of measurements that corresponds to the evolution most companies go through as they implement and refine a measurement system. This evolution reflects that of the ISO 9001 standards themselves. Before the introduction of the first ISO 9001 standard in 1988, industry focused primarily on quantity measurements. The 1994 and 1998 standards gave rise to optimization measurements, which are aimed at corrective and preventive actions. The ISO 9001:2000 standard requires a third generation of measurements to establish processes that strive for continual improvement.

Quantity Measurements. As companies implement a quality system, they focus first on measurements of quantity. Such measurements reflect a single quantitative dimension. Examples include the volume of products manufactured, the total revenue generated, or the number of new product designs released. These measurements have typically been used for production forecasts and inventory management.

Optimization Measurements. In contrast to quantity measurements, optimization measurements provide more detail. While a quantity measurement may include the volume of products produced, an optimization measurement would address factors such as products produced per hour of labor, cycle time required for each product, or external resources consumed per product. Optimization measurements can be used to evaluate efficiency and, therefore, to assess the benefits of process improvement efforts.

Process Improvement Measurements. Although optimization measurements help show how to optimize a given process, they do not show how to redesign a process to achieve new levels of productivity. A process improvement measurement can lead to a root-cause assessment that can identify the source of recurring quality problems. This can lead in turn to identification of new product and process designs, such as the redesign of a printed circuit board to consolidate components and increase reliability or of an automated production line to reduce manufacturing process steps.


Even when the benefits of measurement are understood, the question remains which measurements should be taken. A list of candidate measurements used by many organizations is provided in Table I. These are based on recommendations provided in ISO 9004:2000 and on the best practices of industry leaders.2

Candidate Measurement
Type of Metric
Volume of units produced per month
Process capability
Inventory turnover rate
Throughput yield
Defect rate per month
Analysis of manufacturing defects    
Time to market for new development projects
Efficiency of review and test activities in finding defects  
Productivity of engineering teams  
Defect rate per month per project  
Analysis of product defects related to design    
Percentage of defects found in supplier products
Percentage of certified suppliers
Percentage of dock-to-stock materials
Defect rate per supplier
Analysis of supplier defects to determine root cause    
Customer Satisfaction      
Number and categories of customer complaints
Number of warranty failures
Customer loyalty as measured by repeat purchases  
Customer satisfaction trends  

Table 1. Candidate measurements and metric types.

Manufacturing. The most common measurements used for the manufacturing process are related to production planning and productivity evaluation. Other measurements include evaluation of process optimization factors such as yield (yield data include the number of failures found in each step of the manufacturing process). Even though production capacity may be acceptable, a low yield points to process inefficiencies that may cause a high scrap level and require significant rework. Mikel Harry describes these inefficiencies as a "hidden factory," representing the wasted capacity spent producing bad product.3 Process measurements identify inefficient processes. In addition, analysis of defects (nonconforming products) found during manufacturing can reveal opportunities for process improvement. Tracking manufacturing trends such as process capability values over time also can be helpful in evaluating the long-term effectiveness of process improvement efforts.

Engineering. Engineering is often neglected when process measurements are discussed. The engineering process, the argument goes, is a creative one and should not be subject to the types of measurements used in manufacturing. But while it may be true that manufacturing measurements are not appropriate for engineering, it does not follow that no measurements should be performed.

For most companies, the biggest demand placed on engineering is to reduce time to market for new products. This can be accelerated by selecting the appropriate measurements. Just like the "hidden factory" in manufacturing that wastes resources, engineering resources are often wasted because the customer requirements are not well defined before the design process begins, or because the requirements change after significant design progress has been made. Using measurements to recognize the factors that contribute to delays is the first step in improving the engineering process. In addition to a time-to-market measurement for each project, many organizations also use productivity measurements for engineering teams (such as engineering change orders completed per month or tested software releases per month).

The most beneficial measurement to help improve the engineering process comes from an analysis of design defects found during the engineering process and by customers after release. Another key process improvement measurement involves assessing the product development life cycle to identify the phase where the majority of time is spent. Improvements in this phase offer the greatest potential for shortening the overall development schedule.

Suppliers. With the increasing complexity of today's products, companies are becoming more and more reliant on suppliers. With this increasing reliance on suppliers comes heightened risk that inferior products or services from suppliers may hurt product quality and profits. One appliance manufacturer, for example, traced 75% of warranty failures to the inferior quality of components from suppliers.4 Effective measurement of suppliers is essential not only to ensure quality of products but also to improve productivity. Traditional measurements for suppliers include the number of suppliers that are certified, the number of components that can be accepted without inspection (dock-to-stock), and the number of defects related to supplied components. In addition, analysis of the cause of defects that occur in supplied parts can be shared with suppliers as the basis for implementing a process improvement program. Supplier trend data also helps in selecting suppliers for long-term contracts and preferred contract awards.

Customer Satisfaction. Customer satisfaction is specifically identified as requiring measurement in ISO 9001:2000 section 8.2.1. Although such measurement is widely used as a gauge of product quality, it is also an excellent measurement of profitability. Customer satisfaction studies have shown that a 5% increase in customer retention can translate into a 20% improvement in profits.5 For this reason, trending of customer satisfaction and analysis of customer complaints to identify the root cause of problems have become widespread performance indicators.


Figure 2. A sample Pareto analysis by defect type. Pareto graphs and charts rank the frequency of defects by category. The ranking can be used to determine priorities and choose corrective actions.

This discussion of measurements has focused primarily on evaluating conformity to established quality and productivity goals. These measurement systems can also provide the basis for defining new, enhanced processes. Process measurement systems that address general causes of defects can highlight the vital few defects that are the source of the majority of quality problems.6 (This strategy of ranking defects in order to decide which quality improvement projects to pursue is also known as the Pareto principle, see Figure 2.) For most companies, this is one of the most significant benefits of a measurement process. More detailed knowledge of the cause of quality problems provides a better understanding of the best strategies for correction. As a process matures, the attributes that identify failures can be expanded so that the remaining problems and their causes may be addressed.

Another method for conducting a more detailed analysis of measurements to better define quality improvement programs is a cost-of-quality calculation. The cost of quality represents the difference between the actual cost of a product or service and what the cost would be without product failures, manufacturing defects, or substandard service.7 The cost of quality is calculated as the sum of the cost of external failures (customer complaints), internal failures (failures in manufacturing and engineering), appraisal costs (costs for reviews and testing), and prevention costs (costs for process improvements and training). These costs can be calculated by assessing the average cost for each category and multiplying that cost by the number of instances of the failures or prevention activities reported (cost of appraisal is normally a fixed cost based on the resources dedicated to the activity). Measurement of the cost of quality is an excellent way to gain understanding of the factors that contribute to product quality and production costs. Calculation of the cost of quality frequently clarifies the significant benefits that can be realized from investment in preventive action programs.


With the emphasis on the measurement of processes there is a risk that many ineffective types of measurements may be created. The result can be processes that become more complicated instead of more efficient. The following are some guidelines on how to define and use measurements in a way that minimizes risk and maximizes benefits.

  • Take only measurements that can be used effectively for decisions on process improvement and product evaluation. Measurements that do not add value should be eliminated. The fewer and simpler measurements are, the better.
  • Measurements should be derived from published quality policies and objectives.
  • Measurements should be trended or summarized in a meaningful way to show process stability and effectiveness. Trend charts are most useful in showing the effectiveness of quality improvement programs. Charts should be structured to highlight which activities are effective at achieving target results.
  • Data collected for measurement must be reviewed and verified to confirm correctness.
  • Measurements should be automated to the maximum extent possible to simplify the collection process.
  • Procedures for the collection of measurements should cover the consistent definition of terms, who is responsible for performing the measurements, and when the measurements are to be taken. Training should be provided that emphasizes quantitative decision making and continual improvement based on measured data.
  • Common measurement categories and presentation strategies should be implemented across business processes.
  • Measurements should evolve over time as processes evolve and more meaningful or directed measurements can be established.
  • Whatever is measured will become a focus of attention. Be careful not to introduce measurements that result in unwanted side effects (for instance, measuring lines of code produced for engineering can result in inefficient and excessive amounts of code being produced).
  • Because the purpose of measurements is to identify opportunities for process improvement, management must be willing to invest resources to act on those opportunities. If the measurements are not used to make process change decisions, the credibility of the metrics program will be destroyed.


There are no established standards for acceptable measurement levels for the majority of quality system processes. Although some candidate measurements have been published for productivity and quality levels, these measurements must be considered in light of the target industry and manufacturer. The best answer to measurement standards is benchmarking other companies, preferably the leaders in the industry. Benchmarking is an excellent way not only to establish measurements for excellence but also to learn new process improvement techniques.8

ISO 9001:2000 puts the onus on each manufacturer to define its internal processes and to develop measurements that ensure that these processes are effective. While adjusting to the new standard will be challenging for many manufacturers, the end results will pay them back many times over.


1. ISO 9001:2000, Quality management systems—requirements (Geneva: International Organization for Standardization, December 13, 2000), v.
2.ISO 9004:2000, Quality management systems—Guidelines for performance improvements, (Geneva: International Organization for Standardization, December 13, 2000), 34–45.
3.Mikel Harry and Richard Schroeder, Six Sigma: The Breakthrough Management Strategy Revolutionizing the World's Top Corporations (New York: Doubleday, 2000), 79.
4.JA Donovan and FP Maresca, "Supplier Relations," in Juran's Quality Handbook, 5th ed. (New York: McGraw-Hill, 1999), 21.5.
5.Frederick F Reichheld, The Loyalty Effect, The Hidden Force Behind Growth, Profits, and Lasting Value (Boston: Bain, 1996), 13.
6.JM Juran and FM Gryna, Quality Planning and Analysis, 3rd ed. (New York: McGraw-Hill, 1993), 47.
7.Jack Campanella, Principles of Quality Costs, 3rd ed. (Milwaukee: ASQC Quality Press, 1999), 5.
8.Robert C. Camp, ed., Global Cases in Benchmarking. (Milwaukee: ASQ Quality Press, 1998), 7.

Copyright ©2001 Medical Device & Diagnostic Industry

Custom polyurethane thin films available

to medium production runs can be easily accommodated. When compared to conventional thermoplastic films, thermoset polyurethane films offer improved tensile strength, tear strength, and abrasion resistance. Potential applications include use in abrasion-resistant coatings and transparent coverings for switches, self-adhesive tapes, gaskets, diaphragms, and more. The films can be laminated to other materials (e.g., PTFE) or coated with adhesives to create advanced composite film structures with multiple performance and manufacturing capabilities. Opaque, colored, translucent, and nearly clear films are available. The films are also offered with Mearthane's patented conductive technology to provide static dissipation in applications with resistivity ranges as low as 1 x 107 (omega)-cm.

Reliability Technology for Manufacturers: Engineering Better Devices

Originally Published MDDI October 2001

The medical device industry is among the wide cross section of industries that benefit from reliability engineering. Four methods presented here offer manufacturers concrete techniques to improve the effectiveness of their medical devices.

B.S. Dhillon

The history of reliability engineering goes back as far as World War II, but the serious application of reliability engineering concepts to medical devices is a much more recent development. The latter part of the 1960s are generally regarded as its real beginning, and many publications concerning medical device reliability appeared during this period. 1–3

Since then, several methods and techniques have been developed to analyze and ensure the reliability of engineering systems.4 Many of these methods and techniques can equally be used to ensure the reliability of medical devices and systems. This article presents widely used reliability analysis methods that can be used to improve the effectiveness of medical devices and equipment.

The scope of reliability engineering is extremely wide, encompassing many areas of engineering technology, from ensuring the success of space missions to delivering a steady supply of electric power in a variety of applications. It is useful as a means to improve the reliability and effectiveness of medical devices and systems as well. The four methods of reliability engineering most widely used in the industrial sector are failure rate estimation, fault tree analysis, the Markov method, and failure mode and effects analysis.


The failure rate estimation method is widely used in the industrial sector to estimate the failure rates of electronic equipment.5 During bid proposal and early design phases—when it is often referred to as the parts-count method—failure rate estimation is used to provide a quick estimate of a system's potential failure rate. The information needed to use the method includes generic part types and quantities, the equipment's use environment, and part quality levels. For a single use environment, the equipment or device failure rate is expressed by

Equation 1

where the following variables apply:

  • λd is the device failure rate, expressed in terms of failures per 106 hours.
  • n is the total number of different generic component classifications in the device under consideration.
  • qi is the quantity of generic part i.
  • λgp is the generic failure rate of generic part i, expressed in failures per 106 hours.
  • fq is the quality factor of generic part i.

Tabulated values for λgp and fq of various parts can be found in MIL-HDBK-217, Reliability Prediction of Electronic Equipment.5

As a design matures, more information becomes available, and the failure rates of device parts are estimated individually. Usually, MIL-HDBK-217 is used to estimate the failure rate of electronic parts. The part failure rates are then added to obtain a total-equipment failure rate. This value provides a clearer picture of the actual failure rate of the device under consideration than does one obtained by using equation 1.

An equation of the following form is typically used to estimate failure rates of electronic parts.5

Equation 2

where λp is the part failure rate usually expressed in failures per 106 hours, Θe is the factor that accounts for the influence of environment, Θq is the factor that accounts for part quality level, and λb is the base failure rate normally defined by a model relating the influence of temperature and electrical stresses on the part under consideration.


Assume that a piece of electronic medical equipment is being designed and that it is composed of four different types of parts: 20 8-bit bipolar microprocessors, 10 8-bit MOS microprocessors, 50 varactor diodes, and 5 photodetectors. The equipment is to be used in a ground-benign environment. The generic failure rates are as follows:

    • Bipolar microprocessors have 0.028 failures per 106 hours, with a corresponding quality factor of 0.25.
    • MOS microprocessors have 0.048 failures per 106 hours, with a corresponding quality factor of 2.
    • Varactor diodes have 0.012 failures per 106 hours, with a corresponding quality factor of 0.5.
    • Photodetectors have 0.011 failures per 106 hours, with a corresponding quality factor of 0.7.

To calculate the medical equipment's failure rate, insert the given data into Equation 1, as follows:

λd = (20) (0.028) (0.25) + (10) (0.048) (2) + (50) (0.012) (0.5) + (5) (0.011) (0.7) = 1.4385 failures per 106 hours.

Thus, a quick estimate for the medical equipment failure rate is 1.4358 failures per 106 hours.


Fault tree analysis (FTA) is one of the most widely used methods to analyze engineering designs with respect to their reliability in the industrial sector, particularly in nuclear power generation. FTA is event oriented, as opposed to the failure orientation of the failure mode and effects analysis (FMEA) method described subsequently. Furthermore, FTA is more expensive than FMEA.

FTA starts by identifying an undesirable event—called a top event—associated with a system. Events that might cause the top event are generated and connected by logic operators such as AND and OR. The AND gate provides a true (faults) output if all inputs are true (faults). The OR gate provides a true output if one or more inputs are true.


A solid-state relay used in a medical equipment failure rate, λp (from reference 5), is given by

Equation 3

The following values are specified for the relay:

λb = 0.029 failures per 106 hours, Θe = 3.0, and Θq = 1.9.

Calculating the solid-state relay failure rate by inserting the data into Equation 3 yields the following:

λp = (0.029) (3.0) (1.9) = 0.1653 failures per 106 hours.

Figure 1. Fault tree symbols: (a) OR gate, (b) AND gate, (c) reluctant event, and (d) basic fault event.

The construction of a fault tree proceeds by the generation of events in a successive manner until the events (basic fault events) need not be developed further. The fault tree itself is the logic structure relating the top event to the basic events. These relationships are depicted through the use of a large number of symbols.4,6 The four basic symbols used in the construction of fault trees are shown in Figure 1.

In the figure, the circle represents a basic fault event (e.g., failure of an elementary component); the basic fault-event parameters are failure probability, unavailability, and failure and repair rates. The rectangle denotes the resultant event that occurs from the combination of fault events through the input of a gate such as AND or OR.

The basic steps involved in performing fault tree analysis are as follows:

  • Define the system, the assumptions involved in the analysis, and the events or states that would constitute failure.
  • Establish a system block diagram indicating inputs, outputs, and interfaces if simplifying the scope of analysis is desirable.
  • Establish the top-level fault event.
  • Use fault tree logic and fault event symbols and apply deductive reasoning to identify what could cause the top-level fault event to occur.
  • Continue developing the fault tree by identifying causes for intermediate fault events (i.e., the fault events that can cause the top-level fault event to occur).
  • Develop the fault tree to the desired lowest level, that of the most basic fault events.
  • Analyze the completed fault tree qualitatively as well as quantitatively.
  • Identify appropriate corrective measures.
  • Document the analysis and take appropriate measures to rectify problem areas.

To obtain the output probability of failure of the OR and AND gates, the following two equations can be used:

For OR:

Equation 4

and for AND:

Equation 5

where the following are true: F0 is the OR gate's output fault event probability of occurrence; FA is the AND gate's output fault event probability of occurrence; n is the total number of independent input fault events; and Fi is the occurrence probability of the ith input fault event, for i = 1, 2, 3, . . ., n. (The method is described in detail in reference 7.)


Assume that a magnetic resonance imaging (MRI) machine can be broken down into the three independent subsystems A, B, and C. Subsystems A and B are themselves composed of two parts each: a1 and a2 and b1 and b2, respectively. The machine can fail if any one of subsystems A, B, or C fails. Subsystem A fails if both parts a1 and a2 fail, and subsystem B if either b1 or b2 fails.

If the probabilities of failure of items C, a1, a2, b1, and b2 are 0.01, 0.02, 0.03, 0.04, and 0.05, respectively, then a fault tree for the top event MRI Machine Failure would look similar to the figure below.

The fault tree shows the probabilities of occurrence for basic events and the calculated probability values using equations 4 and 5 for intermediate and top events.

The probability of occurrence of MRI machine failure (the top event) is 0.0977. In other words, there is approximately a 10% chance that the machine will fail.

A fault tree for MRI machine failure.


The Markov method is a powerful reliability evaluation technique that can generally handle more cases than any other method. One important application is reliability analysis of repairable systems. The technique can also be used when the components are independent, or for systems involving dependent failure and repair modes.

The method proceeds by the enumeration of system states. The resulting differential equations are then solved to obtain various reliability measures. The only serious problem with the method is that as the number of system states increases, the calculations can often become unmanageable.

The Markov method is based on the following assumptions:

  • All transition rates (e.g., failure and repair rates) associated with the system under consideration are constant.
  • The transitional probability from one system state to another in the finite time interval Δt is given by αΔt, where a is the constant transition rate (e.g., failure or repair rate) from one system state to another.
  • All occurrences are independent of each other.
  • The probability of more than one transition in the finite time interval Δt from one system state to another is negligible—i.e., (αΔt) (αΔt)→0.


Assume that an x-ray machine's failure and repair rates λ and µ are 0.0005 failures per hour and 0.008 repairs per hour, respectively. The machine's transition diagram is shown in the figure below. Given that, it is possible to calculate the probability that the x-ray machine will be available for service during the next 200 hours of operation.

X-ray machine transition diagram.

The numerals in the figure's box and circle denote the x-ray machine state—that is, when it is operating and failed. Using the Markov method, the following differential equations for the diagram apply:

Equation 6

Equation 7

where Pi(t) is the probability that the x-ray machine will be in state i at time t, for i = 0, 1. At time t = 0, P0(0) = 1, and P1(0) = 0.

Solving equations 6 and 7 yields the following:

Equation 8

Substituting the given data into Equation 8 yields

Equation 9

Thus, there is 95.19% chance that the x-ray machine will be available for service during the 200 hours of operation.


Failure mode and effects analysis (FMEA) was developed in the 1950s and is used to evaluate designs at their early stages in terms of reliability.8,9 This criteria is also very useful to highlight both the need for and the effects of design changes. The method involves listing all possible failure modes for each component with their effects on the device subsystems.

FMEA requires that the following steps be performed:4

  • Define the boundaries of the system under consideration and its associated detailed requirements.
  • List all system components and subsystems.
  • Identify and list each component's failure modes, including a clear description.
  • Assign a failure rate or failure probability to each component failure mode.
  • List each failure mode effect on the subsystem, system, and plant.
  • Enter remarks for each identified failure mode.
  • Review critical failure modes and take appropriate corrective measures.

There are many benefits of performing FMEA: it provides a systematic approach to classify hardware failures, lowers development time and cost, reduces engineering changes, and is easy to understand. It serves as a useful tool for more efficient test planning and highlights safety concerns. Furthermore, this method can improve customer satisfaction and serve as an effective tool to analyze small, large, and complex systems. The method is described in further detail in reference 7.

Perhaps most importantly, FMEA provides a safeguard against repeating the same mistakes in the future and improves communication among design interface personnel.10 The application of this method during the initial stages of medical device design can be very useful.


1. BS Dhillon, "Bibliography of Literature on Medical Equipment Reliability," Microelectronics and Reliability 20 (1980): 737–742.
2. BS Dhillon, Reliability Engineering in Systems Design and Operation, (New York: Van Nostrand Reinhold, 1983).
3. BS Dhillon, Medical Device Reliability and Associated Areas (Boca Raton, FL: CRC Press, 2000).
4. BS Dhillon and C Singh, Engineering Reliability: New Techniques and Applications (New York: John Wiley, 1981).
5. Reliability Prediction of Electronic Equipment, MIL-HDBK-217 (Washington DC: Department of Defense).
6. RJ Schroder, "Fault Tree for Reliability Analysis," in Proceedings of the Annual Symposium on Reliability (1970), 206–210.
7. BS Dhillon, Design Reliability: Fundamentals and Applications (Boca Raton, FL: CRC Press, 1999).
8. JS Countinho, "Failure Effect Analysis," Transactions of the New York Academy of Sciences 26 (1964): 564–584.
9. BS Dhillon, "Failure Mode and Effects Analysis-Bibliography," Microelectronics and Reliability 32 (1992): 719–731. 10. P Palady, Failure Modes and Effects Analysis (West Palm Beach, FL: PT Publications, 1995).

Copyright ©2001 Medical Device & Diagnostic Industry

Outsourcing: Striking the Proper Balance

Originally Published MDDI October 2001


While outsourcing design and manufacturing processes, OEMs should stay involved and maintain in-house expertise.

William Leventon

Outsourcing manufacturing operations enables companies to eliminate overhead and convert fixed production costs to variable costs.

Medical device companies all over the country are disintegrating—but it's not a bad thing. In fact, it may be giving these companies an edge in today's competitive marketplace. "When companies look at their vertically integrated manufacturing operations, they see that they're not as efficient as they could be," says Tom Thompson, executive vice president of business development for Avail (Dallas), a contract manufacturing firm. "So one way to be more competitive is to outsource manufacturing."

Outsourcing is becoming increasingly popular among medical device companies, according to Thompson. "The trend is toward dis-integration, as we've seen before in the automobile and computer industries," he adds.

Manufacturing isn't the only job that companies like Avail are taking on. Many contracting firms now offer a comprehensive menu of services, including design, packaging, shipping, and repair. Outsourcing these tasks and others can save medical device companies time and money. But there are also pitfalls that companies must avoid if their outsourcing ventures are to be successful.


Why should a company outsource? "Manufacturing can really be a pain in the neck," says Al LaVezzi, president of contract manufacturer LaVezzi Precision Inc. (Glendale Heights, IL). It is not uncommon for companies to purchase a number of specialized machines to make a single product. They must then find and hire highly skilled machinists to operate the machines. And when new technology makes the machines obsolete, they have to purchase new ones.

"It's very expensive to maintain in-house manufacturing expertise," notes John Pfaff, vice president and general manager of The MedTech Group (South Plainfield, NJ). By using a contract manufacturer, he adds, "you're distributing the manufacturing cost across a broader base, which should result in a better return on investment than you would get if you did the manufacturing yourself."

This leaves medical device companies with more money to devote to areas such as R&D, marketing, and sales. "Companies are finding that it makes more sense for them to concentrate on their core competencies and outsource to companies the tasks that are our core competencies," says Sandra Schneider, marketing manager for Lake Region Manufacturing Inc. (Chaska, MN).

By outsourcing, companies eliminate the cost of manufacturing overhead. "We convert fixed manufacturing costs to variable costs," explains John Nussbaum, president of Plexus Corp. (Neenah, WI). "Customers are charged a variable rate, depending on how many units they need." When there is no need for production, there are no manufacturing costs to pay.

Both large and small medical companies are outsourcing their manufacturing operations. As they recognize the advantages of the practice, large firms are losing their attachment to proprietary manufacturing technology, according to Clark Briggs, vice president and general manager of Sparton Medical Solutions (DeLeon Springs, FL).

Contract manufacturers now provide full-service outsourcing, such as final assembly and test.

Sometimes, these companies decide that the switch to outsourcing is a good opportunity to "raise the bar" on product quality. A firm that has manufactured a product for a long time may not make changes even if some customers complain about the product. "But we've seen that if they decide to outsource it, that opens the door to tweaking the specification," says Thompson. "They say, 'Let's get six sigma, let's tweak the AQL levels to reduce field failures.' So they're driving us to be better than they were."

As for small medical firms, many simply don't have the money to manufacture their own products. So they use the "virtual company" model common in the networking and telecommunications industries.

"Most medical start-ups outsource everything. It's part of their business plan," notes Jack Fulton, sales manager for Specialized Medical Devices (Lancaster, PA). "For them, the cost of bringing all the necessary capital equipment in-house would be astronomical."

Without contract manufacturing facilities, "it would cost a tremendous amount of money just to get to the starting point, let alone build your product," says Gerald Sanders, managing director of California-based San Francisco Science. "Typically, a start-up company would never realize the economies of scale that would justify putting together that type of facility." Start-ups also benefit from the experience and expertise of their outsourcing partners, Sanders adds.


Both large and small customers have been attracted by capabilities that many contractors have added to their core manufacturing operations. "When people come in, a lot of times they'll look around and say, 'Wow!'" Nussbaum reports. "They're surprised and impressed by what we have to offer."

The evolutionary process that has resulted in today's full-service contractor can be attributed both to increased customer expectations and the contractors' desire to attract new customers. In addition, the expansion of services has followed a certain logic. "We've got the product in our possession, so while we've got it, why not attach this other thing, do this other task, etc., to add more value to the basic fabrication and assembly process?" Thompson says.

Typically, full-service contractors can take a medical company's concept, turn it into a product, and get the product into the hands of customers. Services usually include design, engineering, manufacturing, assembly, packaging, sterilization, and distribution. In recent years, contractors have also been taking on product validations and verifications, vendor audits, and even product repairs.

Contractors have also been expanding their manufacturing capabilities. For example, Specialized Medical Devices has recently added a finishing department, which can handle such tasks as vacuum heat treating, passivation, electropolishing, and laser marking.

Contractors have expanded in some cases by simply bringing in new personnel and equipment. In other cases, they have acquired or partnered with other companies that have capabilities they lack.

By consolidating a variety of services within one company, contractors gain more control over product quality, notes Don Sherratt, director of business and technology for medical devices at the ETL Semko division of Intertek Testing Services Ltd. (Boxborough, MA).

"When different parts of a job are contracted to different companies, you can lose sight of the quality trail," says Sherratt, whose firm works with contract manufacturers. "This makes it extremely difficult to pin down who's responsible for what."

But there can also be downsides to consolidation. "A company might buy a machining operation so they can say, 'Now we can do it all,'" says LaVezzi. "But if it's not a good machining operation, you're stuck with it if you use that contract manufacturer."

Despite such pitfalls, contractors are expected to continue to expand their services. "Our customers are expecting more," Thompson explains. "And I think their expectations will keep escalating."


In response to customer demands, contractors are now deeply involved in product design. "To call us a contract manufacturer is really a misnomer, given the huge amount of design input we give our customers," Nussbaum says.

A contractor's design engineers can assist customers in developing their products. Or they can take a customer's concept and design the product themselves. More customers are asking contractors to handle all design and manufacturing tasks, Pfaff reports.

In developing products for customers, some contractors will add their own intellectual property to the design. "We have access to technology that can make their products work," says Tom Kleist, vice president of sales and marketing for Lake Region, which holds nearly 30 patents for cable and wire technology.

Large and small medical companies alike are outsourcing various manufacturing functions, such as this cable assembly operation.

Nevertheless, Sanders warns medical device companies to be careful about using a contractor's designs and technology. "You could end up with a device that's three-quarters your intellectual property and one-quarter the contract manufacturer's intellectual property," he says. "And you don't necessarily have access to the contract manufacturer's intellectual property. So in effect, you're held hostage by the contract manufacturer. Or worse, you may have to pay them a royalty to use their intellectual property."

To prevent such problems, Sanders advises medical device companies to hire a lawyer who is knowledgeable about agreements between contractors and customers. When the agreement is drafted, lawyer and client should make sure that intellectual property developed for the product in question belongs to the medical device company, and that the medical device company gets an exclusive, paid-up, worldwide license to use any of the contractor's intellectual property used in the product—without being tied to that contractor.

If the proper precautions are taken, Sanders believes that medical device companies could find it economical, efficient, and timesaving to let contractors design and develop their products. At the very least, a company would probably be wise to give the contractor input on design manufacturability. "With their experience, contract manufacturers can often show you different ways to design your product that will make it easier and less expensive to manufacture," says Sherratt. He adds that such changes could reduce the cost of a part by up to 30%.


One of the main reasons contractors have been adding design teams and other services is to get products to market more rapidly. "It's all about speed," Fulton says. "With all these capabilities in-house, we can control our lead times from start to finish. Once you reduce lead times, you're far more successful in attracting and retaining customers."

Why the need for speed? In the past, Nussbaum explains, many medical products had fairly long life cycles. Today, however, new designs and technologies are constantly replacing old ones, making speed to market crucial to the success of medical products.

In the race to turn out new products, contractors claim they have an edge over medical device companies. When it comes to manufacturing products, "we can be quicker because that's where we live," Schneider says. Contract manufacturing plants are filled with state-of-the-art equipment, such as machine tools that combine many operations into one.

Another tool that is speeding up production is the Internet. Contractors report that e-mail has greatly improved communication between people in different locations. Drawings, images, and other types of information can reach their destination in seconds rather than days. The Internet also allows virtual "meetings," where individuals at different points on the globe can review design drawings together and swap information.

Plexus uses the Internet to connect various printed circuit board (PCB) design sites in several different countries. "If we have a rush PCB design project, we can literally work around the clock," Nussbaum says.

Using the Internet, a medical device company fed CAD data on a part directly into a machine tool at LaVezzi Precision. The company "didn't even have a drawing," LaVezzi reports. "They just dumped their program into our system and it made the part for them."

In addition, the Internet can connect one of LaVezzi's coordinate measuring machines to similar equipment in a customer's plant. With the right software, the customer can see exactly what's being checked at the contractor's plant and how it's being done. The customer can also inspect a part in his own facility at the same time the contractor is checking it.

Sparton customers can also use the Internet to keep tabs on a project. From their own facilities, they can access Sparton's database to check the status of a project, make engineering changes, and submit approvals.

Briggs believes that, in time, contractors will offer customers Internet-based tools similar to those provided by major computer manufacturers. "Customers can go to their Web sites, build a custom computer by clicking different boxes, click the order button, check in daily to see the order status, see when it ships, and know the day it's going to show up on their doorstep. We're not quite there yet. But there have been experiments in that area."

Already, Briggs says, the Internet has made it possible for contractors and customers to merge into a "virtual company" in which everyone is connected despite distance. "Customers can come right into your company as though they were down the hall—even if they're in another country."


By tearing down distance barriers, the Internet has made it easier to move manufacturing operations to Mexico and other locations outside the United States. Recently, a medical device company closed its Midwestern manufacturing plant in order to use an Avail-operated facility in Mexico. Avail has four offices in the United States; however, most of the company's employees work in plants in Tijuana.

Put simply, the reason for the move was cost. By manufacturing in Mexico, where wages are much lower than they are in the United States, "we can make our nickel and still save customers a lot of money," Thompson explains.

On the other hand, a medical device company's anxiety about outsourcing can increase with its distance from the manufacturing process. "Any time you outsource, there's a feeling that you're losing some visibility and control because you're not there," Briggs says. "There's a certain comfort in being able to see your product being made every day. And the farther away you are from the manufacturing process, the less comfortable you might be."

A contractor with distant facilities is probably not a "one-stop shop," which is supposed to be desirable because it implies a coordinated work effort. According to Schneider, "one-stop shop" has become a buzzword among contractors, who sometimes advertise themselves as such. She warns, however, that medical device companies ought to investigate such claims. "Is their shop really one stop? Or is it six stops across the country under one [company] name?"


Investigating potential outsourcing partners can be a painstaking process. "It's important for the OEM to get into the contractor's [facility] and do some severe diligence," Thompson says. "Once, a large OEM told us almost apologetically, 'We hope you don't mind us wanting to look at your financials. But we do almost as much diligence in picking a contractor as we would if we were looking at a company for acquisition.' But it makes sense, because if we mess up, you're messed up. So you want to choose your partner very carefully."

During such investigations, contractors recommend checking the company's quality system, pricing structure, design and manufacturing technology, and especially its experience in medical device manufacturing.

Manufacturers should not be satisfied with impressive presentations or fancy certificates hanging on the walls. Sanders recommends conducting audits of the contract manufacturer before signing on the dotted line. "Ultimately, you are the party that's responsible for the product," he tells manufacturers. "So you have to make sure that [the contractor] will keep its commitments."

At Sparton, potential customers are conducting ever-more-detailed audits of the company's facilities. Sometimes, says Briggs, large companies will send teams of people who spend a week at his facility—just to see if Sparton is qualified to bid on their project.

Briggs believes that some of these companies are being extra careful because they have been burned by foreign facilities and small contractors that lacked an adequate manufacturing infrastructure. Whatever the reason, he's actually grateful for the increased scrutiny, because Sparton has upgraded its facility in response.


Once a contractor has been selected, the next step is hammering out an agreement. Fulton recommends that the agreement clearly identify who is responsible for which tasks, what the deadlines will be, and how communications between the contractor and customer will work. In Sanders' view, the agreement should also limit turnover of manufacturing personnel, which can cause production mistakes and delays.

Contracts should also assign responsibility for product quality. "Often, contracts leave the OEM with all the liability," Sherratt says. "You don't want that kind of contract, because then there's no pressure on contract manufacturers to produce a good product. If they make it and it's wrong, they must take responsibility for making things right."

After the contract is signed, the company's engineering, quality, and manufacturing departments can just be eliminated, right? Wrong, says Pfaff. "You can't outsource to the point where you have no in-house expertise," he warns. "Though you don't have to keep the whole department, you still need people in your organization who understand engineering, quality, and manufacturing. These people will interface with the contract manufacturer to make sure your requirements are met."

Contract services can include certain product support functions, such as failure analysis and component-level traceability services.

To make communications as simple and clear as possible, Sherratt suggests that medical device companies have a single point of contact within the contract manufacturing organization. Sanders recommends finding the people who are working on the contracted product and staying in close touch with them.

Although a company may not be doing the actual work, the firm can't just hand off large chunks of the project and move on to other things. "It's a mistake to think that you'll just give instructions to a [contract manufacturer] and a good product will come out of the pipeline," Sanders says. "You can't just turn your back on the process."


Today's full-service contract manufacturers can take on all of the tasks necessary for bringing a medical device to market. These companies save large medical device OEMs time and money, and they're essential to the existence of small medical start-ups. There is more to successful outsourcing, however, than just turning a project over to another company. To prevent problems and disappointments, contractors and customers alike recommend that potential outsourcing partners be carefully investigated and that manufacturers remain involved in the process after the contract has been signed.

Copyright ©2001 Medical Device & Diagnostic Industry

Using Computer-Aided Design to Enhance Product Development

Originally Published MDDI October 2001


Using Computer-Aided Design to Enhance Product Development

More than a design tool, CAD can improve communication, help control the design process, and reduce time to market.

Gregg Nighswonger

Developing A Fingertip-Sized Implantable Heart Pump
Computer-aided designs, such as this model of a fingertip-sized heart pump, can reduce the number of physical prototypes that must be made.

Computer-aided design (CAD) is a tool intended to facilitate the evolution of a complex set of concepts into a completed product. Like all tools, CAD is an extension of the user, reflecting that individual's skills and experience. Within the span of a few decades, the use of pencils, pens, and paper has given way to computers and software. Use of CAD has become commonplace in the high-tech industries. But the transition has not been without its pitfalls.

A former tool designer in the aircraft and aerospace industries likes to share the story of his company's shift to CAD. When the decision was made to fully embrace the new technology, the firm's engineers, drafting personnel, and designers were offered lucrative early retirement options to make room for the new CAD team. But within two years, a large number of these retirees were offered equally generous fees to act as consultants. "The company wanted us to come back and give refresher courses in basic drafting and design," he explained. The new CAD tools enabled the designers to render beautiful drawings at a rapid pace. "But," he added, "the new team kept designing things that, in this world, were impossible to machine or build. So they needed help with using the basic tools and concepts."

This critical relationship between a tool and its user is echoed by Bruce Christie of I.N. Inc. (Los Alamitos, CA), "Did the word processor make everyone an author? Did the Paint program make everyone an artist? No. [CAD does] make you a more productive engineer and it makes you able to tackle programs that perhaps you couldn't tackle with your pencil and paper. But whether you get the job done right, and if the quality of the result is acceptable to the client, depends on the quality of the engineer who's doing it."

Rex O. Bare, president of product development firm Omnica Corp. (Irvine, CA), agrees. He contends that "if the person using the tool doesn't know what he or she is doing in the design process, the tool won't make the design better. It might make it look more interesting, or a little more presentable, but it still goes back to the competency of the person using the tool."

CAD offers diverse benefits in the design process. Among these are more-effective communication, greater control over the design process, and savings in development time. Each of these are factors in achieving an improved flow of ideas from initial concepts through design processes to the finished product. Use of CAD tools also allows a broader range of design factors to be considered.


Product success is influenced significantly by the level and extent of communication achieved during the development process. CAD software is being used to help create an environment in which each participant can make effective contributions to this process. In addition to designers, participants can include experts in human factors and other specialties, as well as representatives from management, marketing, and manufacturing.

Bare explains that the use of CAD "encourages the participation of people who might not normally have been able to make meaningful contributions based on looking at drawings. The marketing guys, the managers, the people who would not have been interested in or been able to deal at the level of an engineering drawing can absolutely look at full-color, fully shaded images and understand what they're looking at. So they can participate, and they do."

From the design consultant's view, the level of communication allowed by CAD systems also addresses another problem. Says Earl Robinson, vice president of industrial design at Omnica, "The 'not invented here' syndrome is real. There are territorial problems. We're consultants, and when we go into companies to work for them, quite often people feel threatened."

Robinson believes that such attitudes can be addressed effectively through broader participation in the design process. "When you can do this as a team, and you get the sense of authorship from everybody at the table, and everybody can give their two cents worth, then it moves a lot faster. We get consensus quicker using CAD. Period."

The current generation of CAD systems also provides a degree of functionality that can enhance the communication of complex concepts. Says Bare, "By being able to look at all sides of the thing, with the section views, with the fit, we're able to solve a lot more of the problems and look at more potential solutions or approaches than we could ever do with the older kinds of systems. And once you have that information in a CAD form you come to realize that there are so many things you can do with it—animations, photorealistic renderings, educational arrows pointing to features to communicate with marketing people or customers, or putting that information into reports so that somebody reading the report knows what they're looking at. You are able to go through all of those things faster, better, and with better communication."

Another advantage of involving more key individuals in the early phases of the design process is that it helps to increase understanding. Says Bare, "One of the reasons it used to be so hard to get consensus was that people weren't willing to commit unless they understood. And they weren't willing to admit they didn't understand, so they dragged their feet so it wouldn't be obvious they weren't understanding what they were looking at. Here that's really not an issue. They may not like what they see or they may react to it and tell you what they don't like or whatever, but at least they understand what they're looking at. And that's fundamental because it's all about communication."

Echoing this idea about communication, Robinson says, "Most of the time what we find is that if you ask the customers or the marketing guys what they want, they really can't tell you very well. But if you show them something and ask them if they like it, they can tell you that."

Ron Sully, Omnica product development, says, "I was talking with our head designer the other day about drawing and his drawing skills, and he said 'These days a lot of times I'll model something because I literally can't draw it.'" Bare adds, "CAD helps designers visualize things. A lot of times it's difficult to visualize something well enough to talk about it. You can kind of begin to visualize this thing and you can talk about it in the abstract, but you couldn't yet draw a picture of it. But if your part's going to be cylindrical and it's going to have some notches here, you can create that in a solid image and then you've got it. You know where you go from there. So the tool itself can help pull you through some design hurdles that would have been really difficult to figure out how to illustrate with the old line drawings."

CAD tools also allow larger teams to tackle design challenges. Says Christie, "The 3-D CAD systems allow a greater degree of collaboration. You can assign multiple engineers; at one point we had as many as four people working on various parts of one product, all working in parallel and knowing that they're not going to get into problems because the CAD system helps enforce the interfaces between the individual parts and keeps everything fitting. So what that did is this. [Although CAD didn't] have as much benefit in reducing the total number of engineer hours, it had a tremendous benefit in reducing the number of weeks to completion."


When I.N. Inc. designers worked with Masimo Corp. (Irvine, CA) to develop the Radical SET pulse oximeter, a gold-medal-winning entry in the 2001 Medical Design Excellence Awards (MDEA) program, CAD was an essential part of the product development process. "I think it was crucial," says Christie. "And the reason is that Masimo was anxious to get the product to market, and it was a fairly complicated product. If you look at the parts breakdown, it's a fairly complicated, involved product, with a lot of functional characteristics. It has a lot of interfaces between parts because of the fact that it has a separate desktop and handheld parts that snap together, with a swivel connector on it. And all these functional aspects are things where computer modeling certainly helps you get to a functional design much faster."

He adds that Masimo was concerned that the device have an attractive yet professional appearance. "The ability to use CAD to implement the industrial design of that product was a critical factor," he says. "Without a high-end CAD tool, it would have been very difficult to achieve the appearance that we accomplished on that. So for both functional and appearance reasons, CAD was very important."

I.N. Inc. sales and marketing manager Jeff Herbert adds, "A lot of the sexiness in the overall appearance of the design was driven by the industrial designer in trying to get an ergonomic feel on the product. We're talking about a handheld product that snaps into a base and charging station. So the ergonomics, the human factors needs, drove a lot of the overall design and then had to be captured with the sophisticated CAD tool that we used."

Omnica also contributed to the development of an MDEA-award-winning product—the HELiOS personal oxygen system from Tyco Healthcare Puritan-Bennett (Indianapolis). Use of Solid Works CAD software enabled Omnica's team to consider a large number of possible solutions and examine many design details, says Bare.

Commenting on this aspect of CAD, Robinson adds, "You can look at so many more details earlier on that are options. You would only consider a few of those options perhaps in years past, where now you can look at 15 or 20 possible approaches and you can mix and match features and carry along a lot more of these options through the design process, without them slowing it down or adding so much complexity that you can't carry them along." Robinson warns, however, "Sometimes that's good and sometimes it's bad. It can lead to what we call creeping elegance, which is 'while we're at it, let's throw in this feature' and nobody asked for it or wanted it. And you have to fight that tendency."


A recent example of how the use of CAD can significantly shorten the time needed to develop new products is the Cardionove implantable heart pump (see sidebar). According to Cardionove Inc. CEO Conrad Pelletier, the use of CATIA software from IBM significantly shortened the product's development time by reducing the number of physical prototypes that were needed.

Pelletier explains that the software allowed the designers to rely more heavily on modeling than on creating prototypes. He says,"It's easier on a computer model to modify any one of the parameters that you believe will improve the design, and you can see the results immediately without actually having to cut the prototype itself—which is a lengthy procedure. Just to machine the prototype is a matter of almost a month. Then when you study the prototype on the bench, it's another two or three weeks before you have results. So if you can model these results on a computer, it's so much faster. Within the same length of time you can develop two to three prototypes rather than only one."

He adds that the software was also used to analyze how the device would interact with human blood. He explains, "We've also worked on trying to predict the zone of hemolysis, which was a surprise to us as a possibility. So on the computer model we could look for areas of nonlinear flow, which is where the destruction of red blood cells occurs mainly. We could then modify the design of the prototypes so as to keep hemolysis, the destruction of red blood cells, to a minimum. This was also a discovery that we made after we started using the models."

The software also provided several key advantages in transferring the device design to actual production. Rishi Madabusi, business development manager for IBM Lifecycle Management, adds, "One was being able to take the virtual model and create the kind of manufacturing instructions needed to create physical prototypes if you wanted."

Madabusi adds, "We're getting to places we couldn't otherwise. And doing it more thoroughly and looking at more options. We might have been able to get one or two solutions carried through with the older methods and arrive at something like the same point, but we're going to get there with a lot more thorough treatment and a better understanding of more things about the product when we get there. It may still have taken us six months to work our way through the mental process to get there, but six months out, with the CAD system, we're going to have looked at it much, much more thoroughly than with older systems." He adds, "In that sense it absolutely shortens the time that we're taking to look at all those options. But there is kind of a minimal time involved with any concept, from the early stages to where it's beginning to be a real workable product. I don't know that we've shortened that by too much."


The use of CAD systems has evolved substantially since it required aerospace designers to emerge from retirement. It now enables engineers and designers to explore many possibilities before bringing solid ideas to the table, to interact more effectively with key departments and individuals at earlier stages of development, and to make better use of time and manpower during critical phases of product development.

CAD software enables designers to better visualize concepts.

These are the key advantages now being offered by CAD: more-effective communication in setting design parameters, the ability to exercise a greater degree of control over the process, and more rapid development time. Of course medical manufacturing is not the aerospace industry. Robinson explains, "The medical industry isn't built on urgency. It's built on thoroughness and quality and thought processes. There are other industries, like automotive and the movie industry, where things are totally built on deadlines. The medical industry isn't built that way. It's more about doing it right. And I think there have been instances where we have done things in time periods that would have been impossible without CAD because we did figure out all the problems early. We solved the fits, we took all the spec sheets off the Internet for the internal components, and when it was built, it was right. We nailed it the first time around. It's not what you do commonly, but it can happen."

Copyright ©2001 Medical Device & Diagnostic Industry

Answering the Call for Harmonization of Medical Device Alarms

Originally Published MDDI October 2001


Industry sounds off on a new document aimed at standardizing the design of alarm systems for use in medical equipment.

Michael E. Wiklund and Eric A. Smith

To many medical device manufacturers, the notion of a medical device alarm standard is, well. . . alarming. That the draft of a standard jointly developed by the International Electrotechnical Commission (IEC) and ISO has grown to more than 60 pages in length is more alarming still. After all, one might ask, shouldn't alarms be simple? Is there really that much to say about designing an effective alarm system? Couldn't the basics be covered in just a few pages, and the rest be left to the designers?

Designing an effective alarm system is not so simple, though the end result should be. For starters, there are numerous design options to consider, such as auditory alarms, visual alarms, and combinations of the two ( e.g., a flashing light accompanied by a beep). One can even annunciate a warning with a tactile cue, similar to the way a pager vibrates to draw attention in a noisy environment or to avoid distracting others.

To complicate matters further, designers must consider the varying degrees of alarm required to alert users to conditions ranging from minor concerns to deadly threats. There are alarms that signal active problems, such as a dangerous arrhythmia, and those that indicate such a problem has occurred and has already righted itself.

When one factors in the wide variety of medical devices currently in use and the needs of a multidisciplinary, multicultural, global user population, the development of a standard of any length seems like quite an accomplishment.

The emergence of an alarm standard draft reflects the medical device industry's overall desire for harmonization, fueled by anxiety about clinical mishaps. One could draw an analogy to the standardization of traffic signals and road signs: Americans are well served by the red, yellow, and green traffic-light convention, as well as the common stop and yield signs. If these visual cues varied from one state to another, one city to another, or even one intersection to another, chaos and more traffic accidents would likely ensue.

Just as harmonization makes drivers' lives easier, so should harmonized alarms make medical workers' lives easier. And, in contrast to the variation in the style of traffic lights and signage from nation to nation, the harmonization of alarms will be global, theoretically enabling manufacturers to develop one alarm system for all markets.

Again, the standard under development is an initiative of ISO and IEC. The document, which is expected to constitute the eighth collateral standard to IEC 60601-1: Medical Electrical Equipment—Part 1: General Requirements for Safety, is titled "General Requirements and Guidelines for the Application of Alarms in Medical Electrical Equipment." A joint working group of ISO Technical Committee 121, Subcommittee 3, and IEC Technical Committee 62, Subcommittee 62A, began work on the collateral standard in 1997. Current plans call for the final standard's release in another year or so.

The group's second working draft, circulated in December 2000, generated an unusually large number of comments from the medical device development community. Many of the comments were highly critical of the first-of-its-kind standard, and, in some cases, indicated the need for major reorganization. The group met for a third time in Vancouver this past July, and again this September in Andover, MA, to continue working toward a revised draft for public comment. At press time, the next meeting was scheduled for March of next year, in London.


The need for an alarm-system standard is strong. A general inconsistency among devices, coupled with the human-factors shortcomings of some alarm systems, presents major headaches for users. The draft standard's introduction highlights these problems, stating, "Surveys of healthcare personnel have indicated widespread discontent with alarm signals. Problems include difficulty in identifying the source of the alarm signal, loud and distracting alarm signals, and high incidences of false-positive or -negative alarm conditions. . . Often, alarm signals are more confusing than enlightening. Many operators respond to alarm signals by disabling the alarm system or by adjusting a[n] alarm limit to a[n] extreme value that effectively disables the alarm system."

Currently, medical workers must learn the characteristics of each medical device's alarm and be able to differentiate between them. For example, one device may announce an alarm condition with a rapidly flashing red light and a repeating beep, while another might emit a steady amber light accompanied by a set of ascending tones. Workers may also struggle to understand the nuances of the more sophisticated alarm systems. The distinctions between "alarm silenced," "audio inhibited," and "audio suspended," for example, are difficult to discern.

Alarm-system design shortcomings are more than a mere nuisance. Inadequacies can lead directly or indirectly to patient injury and death, as can habits developed by workers to avoid the distraction of false alarms, such as turning the alarm system off entirely. While making alarm systems better may not head the list of remedial actions taken to save lives, it certainly stands to make a positive difference.


The alarm design standard evolved in part from guidance provided in three existing ISO documents: ISO 9703: Anaesthesia and Respiratory Care Alarm Signals—Part 1:1992: Visual Alarm Signals, Part 2:1994: Auditory Alarm Signals, and Part 3:1998: Guidance on Application of Alarms. It covers a wide range of alarm-system design characteristics, including:

  • Activation states.
  • Prioritization.
  • Annunciation.
  • Auditory signaling.
  • Visual signaling.
  • Remote signaling.
  • Limit setting.
  • Alarm resetting.
  • Data logging.

Portions of the draft standard are quite detailed, leaving little to designer discretion. For example, one particular part of the document states that alarm systems should include at least four auditory alarm-signal harmonic components within the range of 300 to 4000 Hz. It recommends that a high-priority alarm indicator light be red and flash at a frequency of 1.4 to 2.8 Hz, with a duty cycle of 20 to 60%. Other parts of the standard are more open to interpretation in that they establish basic requirements but leave the details up to the designers.


The large number of comments on the draft standard was not surprising, considering the significant impact the standard will have on medical device design practices. Manufacturers spend considerable time and money developing and validating their proprietary alarm systems, some as brand-specific as the company's logo. As a result, device companies are naturally protective of their approaches to alarm design and what they perceive to be good for customers and for gaining a competitive advantage.

Carl Pantiskas, a clinical engineer at Spacelabs Medical (Redmond, WA) who has reviewed and commented on the draft standard, concurs. He states, "Spacelabs Medical has put a lot of thought into the design of alarm systems for [its] patient monitors. I would expect any organization that has a good solution—one that has been validated through years of effective clinical use—to be somewhat reluctant to change it to something that may not have gone through the same rigorous validation. You'd have to question whether [the standard was] actually improving things." That said, Pantiskas acknowledges the overarching value of alarm-system harmonization, adding, "I fundamentally believe that an alarm standard is necessary, just as long as the standard is realistic and does not restrict technological innovation."

Pantiskas's specific concern is that the draft as it is currently written does not fully account for the entire range of medical device use scenarios, including the use of devices in the home, and the technological options available to alarm designers. He points out, for example, that the standard should take into consideration the needs of the hearing impaired. He also anticipates that new technologies and applications, such as those found in telemedicine, may introduce alarm-system design issues that transcend the device-focused guidance in the standard. Moreover, Pantiskas worries that the guidance might not be relevant to networked devices or devices warranting advanced alarm-system designs.

Pantiskas's concerns match several others in the body of comments on the standard, so those responsible for the draft—who may feel that the standard already addresses a wide variety of use scenarios—will need to reopen their discussion. Such deliberation is representative of the nature of any collaborative effort to develop and refine a common standard. The creative tension between professionals with differing and sincere viewpoints ensures that the final standard will reflect myriad compromises but will ultimately be good for the industry.


Patterns emerge within the voluminous comments on the draft standard. Some reviewers find the document complex and hard to follow, which suggests the need for extensive reformatting. Some argue that the alarm-system requirements are laden with overly specific design recommendations. Others see the draft stand-ard as being too focused on medical equipment that is closely attended—devices used in the operating room, for example—at the expense of equipment that is typically attended less vigilantly, like that used in acute-care settings. Still others are concerned with the standard's attention to usability design characteristics considered by some to be unrelated to patient safety. A few believe that the standard is better suited to the characteristics of complex devices than those of simpler ones. Another criticism is that the drafters of the document try too hard to "teach" readers about the human-factors engineering of alarm systems.

Within the committee itself, there reportedly is considerable debate on the value of adding an informative annex to the standard that prescribes a set of eight alarm sounds. The idea behind the appendix is that alarm conditions associated with medical devices can be divided into logical categories, each with its own unique alarm sound. The categories, as outlined in the appendix, are:

  • General.
  • Cardiac.
  • Perfusion.
  • Ventilation.
  • Oxygen.
  • Temperature/energy.
  • Drug or fluid.
  • Equipment or supply failure.

As such, a ventilator alarm would sound different than an infusion pump alarm, and a clinician would know instinctively where to look, once the disparate sounds become familiar.

In the appendix, each category is assigned a series of musical notes (a, b, c, d, e, f, and g) designed to be easily distinguishable from one another. High- and medium-priority alarms would be annunciated by a series of five and three notes, respectively.

A high-priority cardiac alarm, for instance, would be annunciated by the series: c, e, g, g, C (the small c and large C denote the same note, one octave apart). The alarm-sound designers intend users to map the syllables of the phrase cardiac alarm to the notes. Accordingly, the mapping would be as follows: c = car; e = di; g = ac; g = a; C = larm. The overall sound would be comparable to a trumpet call or the National Broadcasting Company's familiar television chime.

Whether this kind of mapping will produce real benefits, particularly among medical workers who do not speak English, is uncertain. Yet it seems reasonable to assume that the operators of the equipment would come to associate tones with their appropriate equipment type.

Stephen Dain, MD, a practicing anesthesiologist at the London Health Sciences Centre (London, ON, Canada) who has contributed substantially to the draft standard, firmly believes in the syllable-mapping approach to annunciating an audible warning. But he acknowledges another significant concern about the eight-tone concept, one that has caused intense committee debate about the appendix and could ultimately lead to its exclusion from the standard. Dain states, "The recommended sounds have not been sufficiently validated. We don't know for sure if they are the best possible tones or whether or not users will be able to differentiate between them and correctly associate them with specific equipment. We need to do the user research. But, if we don't take this approach, alarm systems will share the exact same tones, making it more difficult to isolate a particular alarm."

Figure 1. A sampling of recommended alarm symbols.

Similar concerns persist regarding a set of recommended alarm symbols, such as those shown in Figure 1. When the authors of the draft standard conducted an industry-funded study of potential device users to determine the symbols' reliability, they found that many clinicians experienced difficulty understanding and differentiating the symbols—a finding that was passed along to the technical committee. The study underscored the importance of validation testing as a prelude to standardization.


When IEC releases the final standard, it might be incorporated directly into the third edition of IEC 60601. More likely, it will be issued as a collateral standard to the second edition of IEC 60601. Either way, the final document will wield the power of a voluntary standard that reflects the industry's consensus on best practices. It will not have the power of government regulations, and, as such, manufacturers will not necessarily be required to comply with it. Rather, manufacturers can continue developing products that incorporate proprietary alarm systems, but they will do so at their own risk. One such risk for a device manufacturer is becoming the target of a product liability claim that points to a noncompliant alarm system as the cause of injury. There is also the risk that customers will gravitate toward compliant devices—and away from noncompliant ones—to establish alarm-system consistency within their institutions.

According to common practice, existing medical devices would likely be grandfathered, meaning their alarm systems would not be subject to the new standard. The principle of grandfathering is somewhat complicated, however, by the fact that alarm-system behavior is often controlled by software, a device component that is updated over time. This means that some existing medical devices could be updated to comply with the new standard. Thus, one can expect some manufacturers to make the updates to existing medical devices, if only to maintain product-line consistency.

Interestingly, the draft standard permits manufacturers to replace the recommended alarm sounds with any number of alternative sounds. The only stipulation is that the alternative sounds be validated; however, the committee has yet to validate even the recommended sounds.


Clearly, the alarm-system standard needs further refinement before it is ready for public release. According to industry reviewers, it must be made easier to use, it must focus more on general alarm characteristics than on integrated solutions, and it must apply to a broader range of devices. In addition, its recommendations need to be validated through user testing.

When they are complete, however, the guidelines set forth in the standard will be valuable to medical device designers, manufacturers, clinicians, and patients alike. Designers will have set guidelines to follow in developing application-appropriate alarm systems, which in turn will save manufacturers time and money. Clinicians will benefit from alarms that are more informative, enabling them to perform better and enhancing patient care.

Copyright ©2001 Medical Device & Diagnostic Industry