See discussions, stats, and author profiles for this publication at: https://www.researchgate.

net/publication/284000814

Reliability in the Medical Device Industry

Chapter · January 2008

DOI: 10.1007/978-1-84800-131-2_61

CITATIONS READS

25 11,655

2 authors, including:

Krishna B. Misra
RAMS Consultants
132 PUBLICATIONS   3,194 CITATIONS   

SEE PROFILE

Some of the authors of this publication are also working on these related projects:

performability Engineering View project

New Trends in System Reliability Evaluation View project

All content following this page was uploaded by Krishna B. Misra on 20 April 2016.

The user has requested enhancement of the downloaded file.

61

Reliability in Medical Device Industry

Vaishali Hegde
Respironics Inc. Sleep and Home Respiratory Central Facility
Monroeville, Pennsylvania, 15146, USA

Abstract: The medical industry is one of the fastest growing segments of US economy. Increasingly,
medical equipment is being used outside a controlled hospital environment. The complexity and increased
use of medical equipment in non-hospital environments has made the need for safe, reliable products
imperative. Although reliability engineering tools are well established, their application to the medical
industry is fairly new. This chapter discusses the reliability tools and standards applicable to the medical
equipment industry.

61.1 Introduction years. According to US Census Bureau’s

The medical industry is one of the fastest growing
segments of US economy. In 2004, health care
spending in the US reached $1.9 trillion, and was
projected to reach $2.9 trillion in 2009 [1]. In 2004,
the US spent 16 percent of its gross domestic
product (GDP) on healthcare. It is projected to rise
to 20% in the next decade. United States is not the
only country in the world spending large sums of
money on health care. Countries all over the world
are spending more and more on health care every
year. According to the Organization for Economic
Cooperation and Development, health care
spending accounted for 10.9 percent of the GDP in
Switzerland, 10.7 percent in Germany, 9.7 percent
in Canada and 9.5 percent in France [2].
Health care spending is increasing all over the
world due to increased life expectancy and
consequently the increased aging population.
Reduction in mortality during the 20th century led
to large increases in life expectancy. By 2000, life
expectancy in the US was approximately 76.9
projections, by 2030, the older population in US is
projected to be double that of 2000, growing from
35 million to 72 million. By 2050, they project the
older population to be around 86.7 million. In other
words, nearly 1 in 5 Americans will be age 65 and
over in 2030 [3].
In 2000, about 92 percent of people aged 65
and over had made at least one health care visit to a
doctor’s office, an emergency room, or a hospital
during the past year (NCHS, 2003a). Among
people 65 and older, the number of health care
visits increased with age.
Chronic diseases have caused most elderly
deaths throughout the last 50 years. Diseases of
heart, malignant neoplasms (cancer),
cerebrovascular diseases (stroke), chronic
obstructive pulmonary diseases/chronic lower
respiratory diseases and pneumonia and influenza
were the top 5 causes of death for people aged 65
and over, in year 2000 [3]. Treatment of these
chronic diseases requires several different types of
medical devices. All data mentioned above, points
982 Vaishali Hegde
to an explosive growth in the medical device and
medical services market in the world.
When it comes to medical equipment, failure is
not an option. From critical devices such as oxygen
concentrators, lasers, ventilators, MRI scanners,
insulin pumps, implantable pacemakers to
instruments as straightforward as stethoscopes,
injections, and thermometers; medical equipment
must be reliable. The medical industry must have
higher reliability standards than other fields.
However, a high reliability standard is hard to
maintain in today’s environment of intense global
competition, pressure for shorter product-cycle
times, stringent cost constraints, higher customer
expectations for quality and reliability and complex
global and heterogeneous markets.
Medical electronic products are increasingly
being used in non-hospital environments by non-
clinical personnel. This is because sophisticated
medical products are becoming more compact and
portable. For example, a few years ago, an oxygen
concentrator weighed 75 pounds, had to be plugged
into the wall for power, and had to be used in a
controlled clinical environment. Today, a portable
oxygen concentrator weighs only 8 pounds, can
operate on batteries and can be used in a home
environment. Developing and producing a medical
electronic product reliable enough to continuously
repeat certain key functions over and over again
like clockwork and alarming appropriately in case
of a fault condition, is now more important than
ever before.
Today, reliability takes on even greater
significance since medical devices designed for
specific applications are becoming more expensive.
It is not uncommon for cancer treatment or surgery
equipment to cost a few million dollars. While
hospitals may be relatively profitable, they cannot
afford to have too many of these expensive
products. If one such medical system incurs
downtime due to failure, it can adversely affect the
hospital’s bottom line, its reputation and the well-
being of its patients [4]. In some cases, medical
electronics products in use for several years require
updating to increase their reliability as greater
healthcare demands are placed on them.
The financial impact of having an unreliable or
unsafe medical product in the market can be
devastating for a medical device manufacturer.
Since December 13, 1984, the FDA Medical
Device Reporting (MDR) regulations have required
firms who have received complaints of device
malfunctions, serious injuries or deaths associated
with medical devices to notify FDA of the incident.
The MDR regulation provides a mechanism for
FDA, manufacturers, importers and user facilities
to identify and monitor significant adverse events
involving medical devices. The goals of the
regulation are to detect and correct problems in a
timely manner. MedWatch, the FDA’s safety
information and adverse event reporting program is
set up for health professionals and consumers to
report serious adverse reactions and problems
related to drugs, medical devices, cosmetics etc.
MedWatch plays a critical role in FDA’s post
marketing surveillance – the process of following
the safety profile of medical products after they’ve
begun to be used by consumers. The FDA
publishes a weekly FDA Enforcement Report that
contains all enforcement actions including recalls,
field corrections, seizures and injunctions. This
report is published on the internet at
http://www.fda.gov/opacom/Enforce.html. It also
maintains MAUDE (Manufacturer and User
Facility Device Experience database) which
contains reports of adverse events involving
medical devices.
Depending on the severity of MDR or
MedWatch report filed, a defective medical device
may have to be recalled. A recall is a method of
removing or correcting products that are in
violation of laws administered by the Food and
Drug Administration (FDA). Filing an MDR or
recalling a medical device takes a significant
amount of time, effort and resources on the part of
the manufacturer. It can have a negative impact on
a device manufacturers’ reputation in the
marketplace. It may lead to product liability
lawsuits, particularly in the US. All this could
consequently lead to an adverse impact on sales
and profits.
Low reliability can have a significant impact on
service, repair, and warranty costs as well. In fact,
warranty cost is inversely proportional to the
reliability of a medical device. Hence, more and
more manufacturers are willing to invest in

reliability related tasks to try and reap the benefits
in terms of warranty costs. A 5% increase in
reliability focused development costs will return a
10% reduction in warranty costs. A 20% increase
in reliability focused development costs will
typically reduce warranty costs by half and a 50%
increase in reliability focused development costs
will reduce warranty cost by a factor of 5 [5].
61.2 Government (FDA) Control
As the use of medical devices for critical life
supporting functions increases, governments
around the world are regulating their design,
manufacture, quality, reliability, proper use and
disposal. The Food and Drug Administration
(FDA) is the U.S. Government agency that
oversees most medical products, foods and
cosmetics. Within FDA, the Center for Devices and
Radiological Health (CDRH) oversees the safety
and effectiveness of medical devices and radiation-
emitting products.
The FDA is responsible for protecting the
public health by assuring the safety, efficacy, and
security of human and veterinary drugs, biological
products, medical devices, the nation’s food
supply, cosmetics, and products that emit radiation.
The FDA is also responsible for advancing the
public health by helping to speed innovations that
make medicines and foods safer, and more
affordable, and helping the public get the accurate,
science based information they need to use
medicines and foods to improve their health.
Products regulated by FDA include all foods
except meat and poultry, prescription and non-
prescription drugs, blood products, vaccines and
tissue transplantation, medical devices and
radiological products, including cellular phones,
animal drug and feed, and cosmetics. In addition to
setting product standards, FDA regulates the
labeling of products under its jurisdiction. This
information must be valid, well documented, and
not misleading. FDA plays a major role in
protecting consumers and the public health.
The Federal Food, Drug, and Cosmetic Act of
1938 authorized the FDA to take formal or
informal regulatory measures against the
misbranding or adulteration of medical devices.
Reliability in Medical Device Industry 983
Amendments to this act in 1976 empowered the
FDA to regulate medical devices during their
design and development phases. The Safe Medical
Device Act passed in 1990, authorized the FDA to
implement the Preproduction Quality Assurance
Program. This program requires medical device
manufacturers to address deficiencies that lead to
failure during design. Nowadays, similar types of
regulations concerning medical devices are being
followed in other countries. For example, the
Medical Device Directive (MDD) of the European
Union (EU) outlines the requirements for medical
devices in the EU countries.
61.3 Medical Device Classification
The FDA has established three regulatory classes
for medical equipment based on the level of control
necessary to assure the safety and effectiveness of
the device. The class to which a device is assigned
determines, among other things, the type of
premarketing submission / application required for
FDA clearance to market.
Device classification is risk based, that is, the
risk the device poses to the patient and/or the user
is a major factor in the class it is assigned. All
classes of devices are subject to General Controls
which are the baseline requirements of the Food,
Drug and Cosmetic (FD&C) Act.
The three classes and the requirements which
apply to them are:
• Class I - General Controls
Class I devices are subject to the least regulatory
control. They present minimal potential for harm to
the user and are often simpler in design than Class
II or Class III devices. Class I devices are subject
to "General Controls" as are Class II and Class III
devices. Examples of Class I devices include
elastic bandages, examination gloves, and hand-
held surgical instruments.
• Class II - General Controls and Special
Controls
Class II devices are those for which general
controls alone are insufficient to assure safety and
effectiveness, and existing methods are available to
provide such assurances. In addition to complying
with general controls, Class II devices are also
984 Vaishali Hegde
subject to special controls. Special controls may
include special labeling requirements, mandatory
performance standards and postmarket
surveillance. Examples of Class II devices include
powered wheelchairs, infusion pumps, and surgical
drapes.
• Class III - General Controls and Premarket
Approval
Class III is the most stringent regulatory category
for devices. Class III devices are those for which
insufficient information exists to assure safety and
effectiveness solely through general or special
controls. Premarket approval by FDA, is required
for this class of devices. Premarket approval is the
required process of scientific review to ensure the
safety and effectiveness of Class III devices.
Class III devices are usually those that support
or sustain human life, are of substantial importance
in preventing impairment of human health, or
which present a potential, unreasonable risk of
illness or injury.
The reliability activities that need to be
performed during the life cycle of a product,
depends to some degree on the classification of the
product. Class III devices need maximum analyses
and testing because they support or sustain human
life.
61.4 Reliability Program
Reliability engineering tools are well established
however their application to the medical industry is
fairly new. The first step in launching a safe and
reliable medical product is a good reliability
program. What are the hallmarks of a good
reliability program? Developing realistic reliability
goals early, planning an implementation strategy
and then executing the strategy are all key features
of a good reliability program.
The four phases of a reliability program are [6]
• Concept Phase
• Design Phase
• Prototype Phase
• Manufacturing Phase
Several different reliability tools are available for
use in the four phases of a reliability program. In
the concept phase, one can use benchmarking and
gap analyses to develop a reliability program and
integration plan.
In the design phase, one can use reliability
modeling and predictions, derating
analysis/component selection, worst case circuit
analysis, thermal analysis, electromagnetic
analysis, design of experiments, risk management /
FMECAs, fault tree analysis, human factors
analysis and software reliability analysis to
increase reliability of a product.
In the prototype phase, one can perform highly
accelerated life testing (HALT), design verification
testing (DVT), and reliability demonstration testing
(RDT).
In the manufacturing phase, the reliability tools
that can be used are highly accelerated stress
screening (HASS), on-going reliability testing,
FRACAS/CAPA system setup, and end-of-life
assessment.
61.4.1 Concept Phase
61.4.1.1 Benchmarking
Benchmarking is the process of comparing the
current project, methods, or processes with the best
practices in the industry. Benchmarking is crucial
to both a startup company as well as an established
company that is coming out with a new product to
assure the new product is competitive based on
reliability and cost. For example, company A, a
new entrant into the medical manufacturing
market, wants to start selling a new type of dialysis
pump. Company A should check the warranty
information, annual service plan and recommended
maintenance schedule of leading competitor
products to try and benchmark their dialysis pump
with respect to MTBF, failure rate, service interval
etc. against competitor pumps.
61.4.1.2 Gap Analysis
Gap Analysis compares your current capabilities
with what is expected of your product in the
industry. Before performing a gap analysis you
have to set a reliability goal and perform a review
of your current capabilities. Gap = Goals - Current
Capabilities. For example, a ventilator
manufacturer currently has a ventilator on the
market that has a field reliability of 0.988. It is now

planning to add oxygen blending feature to this
ventilator. They would like the new ventilator to
have a reliability goal of 0.9999 to meet customer
expectation. The process of determining the current
capability (0.988), the goal (0.9999), and the gap
(0.9999-0.988 = 0.0119) is called gap analysis.
The output of this phase is the reliability
program. The program generally quantifies out of
box failure rate, reliability within warranty period,
and reliability throughout life of the product. It also
defines a schedule of the different activities that
will be performed to achieve your reliability goal.
61.4.2 Design Phase
61.4.2.1 Reliability Modeling and Predictions
A reliability prediction is a method of calculating
the reliability of a product by assigning a failure
rate to each individual component and then
summing all of the failure rates. Standards such as
MIL-HDBK-217, Bellcore, PRISM, CNET, HRD5
etc can be used for reliability predictions. A
reliability model presents a clear picture of
functional interdependencies and provides the
framework for developing quantitative product
level reliability estimates to guide design trade-off
process. Models are helpful for identification of
single points of failure, making numerical
allocations, evaluating complex redundant
configurations, and showing all series-parallel
relationships.
61.4.2.2 Derating Analysis / Component Selection
In MIL-STD-721 derating is defined as using an
item in such a way that applied stresses are below
rated values. Limitation of electrical, mechanical
and environmental stresses is critical to high
reliability. One common rule of thumb used in
electronics is that 50% derating can decrease the
failure rate by 30%. Telcordia SR332 (Reliability
Prediction Procedure for Electronic Equipment)
standard gives derating information. Component
selection must be based on the basis of derating
analysis. Design engineers should select
appropriate component ratings to avoid
overstressing them. In case of implantable medical
devices, biocompatibility and reliability of the
Reliability in Medical Device Industry 985
material and components should be considered as
well, while making component selections.
61.4.2.3 Worst Case Circuit Analysis
Worst case circuit analysis (WCCA) is an analysis
technique which, by accounting for component
variability, determines circuit performance under a
worst case scenario, i.e., under extreme
environmental or operating conditions. The output
of a WCCA allows an assessment of actual applied
part stresses against rated part parameters, which
can help ensure the application of sufficient part
stress derating to meet design requirements. One of
the most critical steps involved in completing a
meaningful WCCA is the development of a part
characteristic database. This database should
contain a composite of information necessary for
quantifying sources of component parameter
variation. Once these sources have been identified,
the database can be used to calculate worst case
component drift for critical parameters.
61.4.2.4 Thermal Analysis
Temperature is one of the important variables that
impacts system reliability. Therefore, the thermal
design of a system must be planned and evaluated
carefully. Thermal Analysis techniques can be used
to evaluate the circuit board, casing, and junction
temperatures and check for components that
exceed their temperature limits. The analysis can
also be used to identify the hot spots and
components on the board that designers may
modify by various means to eliminate thermal
problems. Typical solutions are adding heat sinks,
thermal pads, and thermal vias, using thicker
ground or power planes, local conduction pads,
thermal screws, changing of radiation emissivity by
coatings, or relocating components.
61.4.2.5 Electromagnetic Analysis
Electromagnetic compatibility, or EMC means that
a device is compatible with (i.e., no interference is
caused by) its electromagnetic (EM) environment
and it does not emit levels of EM energy that cause
electromagnetic interference (EMI) in other
devices in the vicinity. A medical device can be
vulnerable to EMI if the levels of EM energy in its
environment exceed the EM immunity (resistance)
986 Vaishali Hegde

to which the device was designed and tested. The level. There are several different types of FMEAs.
different forms of EM energy that can cause EMI For example, design FMEAs, process FMEAs,
are conducted, radiated, and electrostatic discharge functional FMEAs, system FMEAs etc. The basic
(ESD). EMI problems with medical devices can be approach for all the FMEAs is the same. A sample
very complex, not only from the technical section of a FMEA worksheet is given in Figure
standpoint but also from the view of public health 61.1.
issues and solutions. The Center for Devices and
Radiological Health (CDRH) encourages
manufacturers of electro-medical equipment to use 61.4.2.7 Fault Tree Analysis
the IEC 60601-1-2 standard, a widely recognized A fault tree analysis is a systematic, deductive
standard issued by the International methodology for defining a single specific
Electrotechnical Commission, Geneva, undesirable event and determining all possible
Switzerland. The standard provides various limits reasons or failures that could cause that undesirable
on emissions and immunity. event to occur. The undesired event is the top event
61.4.2.6 FMECA in the fault tree and is generally a catastrophic or
complete failure of the product. Fault trees use
Failure modes, effects and criticality analysis concepts of logic gates to determine the overall
(FMECA) is a reliability evaluation and design reliability. In medical devices, patient harm is the
review technique that examines the potential most common undesired top event. The results of
Fig. 61.1: Sample Section of a FMEA Worksheet

failure modes within a system or lower indenture FTA can be used as a troubleshooting tool during
level, in order to determine the effects of failures service calls, after the product is available in the
on equipment or system performance. FMECA market. A sample section of a fault tree is given in
uses a “bottom up” approach. This approach begins Figure 61.2.
at the lowest level of the system hierarchy and
traces up through the system hierarchy to
determine the end effect on system performance.
The criticality portion of this method allows us to
place a numerical value or rating on the criticality
of the failure effect on the entire system or user.
After the initial analysis, one can prioritize the
failure modes, provide mitigations, reanalyze,
rescore and confirm that risks are at an acceptable
 Reliability in Medical Device Industry 987

rates [7]. The Mills model was developed by H.D.

Mills [8] by reasoning that an estimation of faults
remaining in software could be made through a
seeding process that assumes a homogeneous
distribution of a representative class of faults.
Before starting the seeding process, fault analysis
is performed to determine the expected types of
faults in the code and their relative frequency of
occurrence [7].

61.4.3 Prototype Phase

Testing is an important part of any product

Fig. 61.2: Sample Section of a Fault Tree
development cycle. It is performed to verify
performance, understand failure modes and
weaknesses of the product, benchmark your
61.4.2.8 Human Factors Analysis product against competitor products, determine
reliability and safety of the product, and study the
Human factors analysis should be performed with
impact of stress on performance of the product.
safety, manufacturing and maintainability in mind.
Reliability and safety testing is crucial in medical
Human beings are famous for not reading and
devices.
following instructions. Surrounding distractions is
Testing can be performed in two different
a major concern in the hospital environment.
modes: standard or accelerated. In standard mode,
Medical device manufacturers have to make sure
tests are performed in ambient temperature, at
that hospital staff is able to use the medical
typical operating stress conditions. In accelerated
equipment correctly and efficiently in spite of all
mode, parameters such as temperature, voltage,
the other surrounding hospital distractions. They
current, cycling, are increased well above their
also have to make sure that symbols used on their
normal levels to reduce test times.
medical devices meet standards approved by the
In the prototype phase, one can perform highly
FDA. AAMI (Association for the advancement of
accelerated life testing (HALT), design verification
medical instrumentation) has published a technical
testing (DVT), and reliability demonstration testing
information report on graphical symbols for
(RDT).
electrical equipment in medical practice.
61.4.3.1 HALT
61.4.2.9 Software Reliability Analysis
Highly Accelerated Life Testing is a quick and cost
Software is an important component of medical
effective tool used for discovering design issues
devices. A software life cycle can be divided into
and improving design margins. Thermal stress and
design, coding, and testing. If good practices are
vibration stress are the two main stresses applied
followed in all stages of the software life cycle a
during HALT. Generally a product is subjected to a
reliable software product will be ensured. The
cold thermal cycle, a hot thermal cycle, a fast
USAF Model and the Mills Model are two
thermal transition cycle, a vibration cycle, and a
software reliability models that are generally
combined thermal-vibration cycle. A root cause
considered useful for medical devices. The USAF
analysis is performed at the end of the test to
model was developed by the U.S. Air Force
determine the cause of failures. Design and
Laboratory in Rome, NY. to predict software
processing changes can be made based on HALT
reliability during the initial phases of the software
findings. A verification HALT can be performed to
life cycle. The model begins by developing the
ensure that problems are fixed and new problems
predictions of fault density and then transforming
have not been introduced. For example, during a
them into other reliability measures such as failure
988 Vaishali Hegde

HALT test, a ventilator with an operating range of 61.4.4.1 HASS

5C to 40C, provided incorrect therapy above 35C
Highly Accelerated Stress Screening is used to
and shut down at 65C. The root cause of failure
reduce out of box quality returns, decrease field
was determined to be the temperature rating of a
service and warranty costs, and detect and correct
proximal pressure flow sensor. The standard sensor
process defects. Typical failures that HASS will
was changed to an extended temperature range
find are soldering defects, bent IC leads, socket
sensor to solve the problem. On the same
failures, incorrect components or component
ventilator, the internal battery was disconnecting at
placement and programming errors. It is
very low vibration levels. The battery connector
recommended that a HALT be performed prior to
was changed to a positive latch, mini lock type
performing a HASS. HALT will help to identify
connector to prevent disengagement of the battery.
the operating and destructive limits of a product.
Both the sensor issue and the battery connector
This information is essential while developing a
issue would not have been discovered before
HASS process.
product release if HALT had not been performed.
61.4.4.2 ORT
61.4.3.2 DVT
On going reliability testing is performed to get an
Design Verification Testing is performed to verify
indication of the reliability of the product before it
product functionality. Generally, this testing is
is shipped to customers. In this test, a sample of
performed under normal operating conditions; no
product is taken off a production line and tested for
external stresses are applied. The results of design
a period of time, adding the cumulative test time to
verification testing are generally required during
ensure that the reliability target will be met. The
pre-market notification 510(k) submittal to the
samples are rotated on a periodic basis to get an
FDA.
on-going reliability. One must take care not to wear
61.4.3.3 RDT out any components because these units are
shippable units and you cannot risk taking
Reliability Demonstration Testing is used to
significant life out.
validate reliability prediction analyses and gather a
measure of confidence that the released product 61.4.4.3 FRACAS/CAPA system setup
will achieve a certain reliability target. In this test,
Failure reporting and corrective action systems
a sample of units is tested at slightly accelerated
(FRACAS) or Corrective and Preventive Action
stresses for several months. The stresses are higher
systems (CAPA) provide a framework for
than normal operating stress conditions but lower
controlling corrective action processes. At its core,
than HALT stress conditions. The stresses are held
FRACAS is a closed loop corrective action system
constant thus enabling you to calculate the
which enables you to collect failure data, analyze
acceleration factor for the test. MTBF (mean time
data and determine root cause, document the
between failures) can be obtained from test results.
corrective action and implement controls to prevent
This test is generally performed once before
the reoccurrence of the failure. An integral part of
product release.
CAPA is a Failure Review Board (FRB). The
Board is typically a cross-functional group
61.4.4 Manufacturing Phase
representing Quality, Reliability, Engineering,
Manufacturing, Regulatory and other departments
In the manufacturing phase, the reliability tools
depending on the nature of the business. The Board
that can be used are highly accelerated stress
is responsible for reviewing and approving
screening (HASS), on-going reliability testing,
recommended corrective actions and evaluating
FRACAS/CAPA system setup, and end-of-life
effectiveness of corrective actions after
assessment.
implementation. A FRACAS system is essential
for medical device manufacturers because of the

FDA Medical Device Reporting (MDR)
regulations.
61.4.4.4 EOL Assessment
End of Life assessment is performed to determine
when a product is starting to wear out, whether the
preventive maintenance strategy is effective and
whether the predictions performed during design
phase were accurate. EOL assessment uses the
Weibull plotting technique to figure out where the
product is on the bathtub curve. Field failure data is
used to determine the number of days before
failure.
61.5 Reliability Testing
There are many different reasons for performing
reliability testing. The main reasons are to induce
failure modes as well as detect unanticipated
failure modes so corrective actions can be
implemented, to determine if items or system meet
reliability requirements, to compare estimated
failure rates to actual failure rates, to monitor
reliability growth over time, to determine the safety
margin in a design, to estimate MTBF or MTTF
values, and to identify weaknesses in the design or
parts [9].
O’Connor states that reliability testing is a part
of an integrated test program involving statistical
testing, to optimize the design of the product and
the production processes; functional testing, to
confirm the design; environmental testing, to
ensure that the product can operate under the
projected environments; reliability testing, to
ensure that the product will operate for its expected
life; and safety testing, to ensure safety of the
product for use by humans, animals or property
[10].
Developing a test plan is the first step of an
integrated test program. The Reliability Toolkit
provides a test plan outline which includes the
following steps: define purpose and scope, list
reference documents, define test item facilities, test
requirements, test schedule, test conditions, test
monitoring, test participation, failure definitions,
test ground rules, and test documentation [11].
Reliability in Medical Device Industry 989
Test procedures are required for the proper
execution of the test plan. Test procedures must
address calibration and proofing the test
equipment. They must have a detailed description
of the tools, parts, adjustments, hook ups,
datasheets, tools and materials required for the test.
The Reliability Toolkit provides a reliability test
procedure checklist that includes equipment
operation, on/off cycles, operation modes,
exercising methods, performance verification
procedure, failure event procedure, and
adjustments and preventive maintenance schedule
[11].
There are several different types of reliability
tests that can be performed over the life of the
product cycle. Reliability Development/Growth
Tests (RD/GT), Reliability Qualification Tests
(RQT), Performance Tests, Screening Tests and
Probability Ratio Sequential Tests (PRST).
61.5.1 Development / Growth Test
This type of test is run to determine if there is a
need to change the design to achieve reliability
specification, or to verify improvements occurring
in design reliability after changes have been made.
These tests are generally run in the prototype or
development phase. Estimates of MTTF or MTBF
can be obtained from these tests.
61.5.2 Qualification Test
The objective of this type of test is to determine if a
design is acceptable for the intended function.
Qualification tests in many cases include aspects of
vibration, shock, temperature cycling and other
environmental considerations the product will see
in use.
61.5.3 Acceptance Test
Acceptance testing is statistically derived to
determine if an item is to be accepted or rejected
for use, either individually or on a lot basis. This
method of testing does not provide MTBF, MTTF
or any other quantifiable measure.
990 Vaishali Hegde
61.5.4 Performance Test
Performance testing is conducted on completed
designs and normally manufactured items to verify
the reliability predictions and test results in the
preproduction phase. This testing provides a
benchmark for comparison of previous activities to
see if they are effective in delivering a design that
meets reliability requirements.
61.5.5 Screening
Screening tests are performed with the intent of
eliminating the infant mortality period. By
eliminating infant mortality period, the yield of
finished product is improved, and a higher
reliability is achieved out in the field.
61.5.6 Sequential Testing
Sequential testing for products designed to operate
for some period of time is outlined in MIL-STD-
781. These tests called Probability Ratio Sequential
Tests (PRST) are based on the ratio of an
acceptable MTBF which should have a high
probability of acceptance to an unacceptable
MTBF which should have a low probability of
acceptance. Items are placed on test, and failures
that occur are plotted against test time. A decision
is made based on the plotted point. The decision is
one of three, accept the items as meeting the
required MTBF, reject the items as not meeting
acceptable MTBF, or continue testing.
Sequential test plans provide the least amount
of time to make a decision when product is either
very good or very bad as compared to reliability
requirements. When the product is borderline,
testing can be continued for an indeterminate
period of time, which is an unacceptable situation.
To prevent the tests from continuing indefinitely
there are specific truncation points designed in
these tests. The test will terminate at either a given
amount of failures or a predetermined test time, if a
decision has not been made.
61.6 MTBF Calculation Methods in
Reliability Testing
MTBF (mean time between failures) is used
frequently to assess reliability of a medical
product. There are five methods to estimate MTBF
in reliability testing [7].
• Time terminated, failed items replaced
• Time terminated, failed items not replaced
• Failure terminated, failed items replaced
• Failure terminated, failed items not
replaced
• No failures
All the methods assume exponential distribution
(constant failure rate). Each method is described
below.
61.6.1 Time Terminated, Failed Items
Replaced
In this method, testing is terminated at a
preassigned time, and all failed items are replaced
immediately after their individual failures. MTBF
can be calculated using
MTBF = MTd n (61.1)
where Td is the test duration time, M is the number
of items placed on test and n is the total number of
failures. One does not need to record the time to
failure of each individual unit on test. For
example, a medical device manufacturer places
fifteen identical handheld oximeters on test, for
2400 hours. Six of the fifteen oximeters failed in
the 2400 hour time period. All failed oximeters are
replaced as soon as they fail. An estimate of the
MTBF of the tested oximeters can be given by:
MTBF = (15)(2400) 6
≡ 6000hours
Therefore, the MTBF of the oximeters is 6000
hours.
 Reliability in Medical Device Industry 991

61.6.2 Time Terminated, Failed Items not 61.6.4 Failure Terminated, Failed Items not
Replaced Replaced

In this method, testing is terminated at a In this method, testing is terminated at a

preassigned time, and the failed items are not predetermined number of failures, and the failed
replaced. MTBF can be calculated using items are not replaced. MTBF can be calculated
using
 n 
∑
MTBF =  Ti + ( M − n)Td } n
 i =1 
(61.2)  n
∑

MTBF =  Ti + ( M − n)Td } n (61.4)
 i =1 
where Ti is the time to failure, M is the number of
items placed on test; Td is the test duration time
where Ti is the time to failure, M is the number of
and n is the total number of failures. One needs to
items placed on test; Td is the test duration time
record the time to failure of each individual unit on
and n is the total number of failures.
test.
For example, a manufacturer started a test with
five identical defibrillators. Two of the five 61.6.5 No Failures Observed
defibrillators failed after 75 and 110 hours. The
failed defibrillators were not replaced. The test was In this method, MTBF cannot be calculated
terminated at a predetermined time of 500 hours. because no failures are observed. However, a lower
What is the MTBF of the tested defibrillators? one-sided confidence limit can be calculated to
state the minimum limit of the MTBF for a
MTBF = {(75 + 110 ) + (5 − 2 )(500 )} 2 specified confidence level.
≡ 842.5hours
LOCL = (2Tt ) χ α2 , m (61.5)
Therefore, the MTBF of the defibrillators is 842.5
hours.
61.7 Reliability related Standards and
61.6.3 Failure Terminated, Failed Items
Replaced Good Practices for Medical
Devices
In this method, testing is stopped at a
predetermined number of failures, and each failed Nowadays, compliance to certain safety, quality,
item is replaced. MTBF can be calculated using and usability standards is a pre-requisite to gaining
approval from FDA to market a medical device in
MTBF = MTd n (61.3) the USA. There are over 700 medical standards
available on different topics for different types of
where Td is the test duration time, M is the number
devices. Some standards directly or indirectly
of items placed on test and n is the total number of
related to medical device quality, reliability, safety,
failures. The equations for time terminated and
and usability are given in Table 61.1.
failure terminated are equivalent when failed items
are replaced.
992 Vaishali Hegde

Table 61.1. Reliability Standards

Standard Reference No. Standard Title / Description

ANSI/AAMI/ISO 14971:2000 Medical devices – Application of risk management to medical
and 14971:2000/A1:2003 devices
ANSI/AAMI ES1-1985 Safe current limits for electro medical apparatus
ASTM F1100 – 90 Standard Specification for Ventilators Intended for Use in Critical
Care
ASTM F1246 – 91 Standard Specifications for Electrically Powered Home Care
Ventilators, Part 1 – Positive Pressure Ventilators and Ventilator
Circuits
ASTM F1463 Specification for alarm signals in medical equipment used in
anesthesia and respiratory care
ASTM F0792 Guide for computer automation in the clinical laboratory
AAMI EOTP-2/85 Performance evaluation of ethylene oxide sterilizers – EO test packs,
good hospital practice
AAMI TIR 12:2004 Designing, testing, and labeling reusable medical devices for
reprocessing in health care facilities: A guide for medical device
manufacturers
AAMI TIR 32:2004 Medical device software risk management
AAMI HE-1988 Human factors engineering guidelines and preferred practices for the
design of medical devices
Guidance for the content of Pre-market Submissions for Software
Contained in Medical Devices Issued May 11, 2005 by CDRH
ISO 10651-6 Lung ventilators for medical use – Particular requirements for basic
safety and essential performance
ISO 10993-1 Biological evaluation of medical devices – Part 1: Evaluation and
testing
ISO 15001 Anesthetic and respiratory equipment – Compatibility with oxygen
ISO 8185:1997 Humidifiers for medical use – General requirements for
humidification systems
ISO 9000 series ISO 9000 series of quality standards
ISO 1348 Quality systems – medical devices – particular requirements for the
application of ISO9002
IEC 60601-1-1 Medical Electrical Equipment Part 1: General requirements for safety
– Collateral Standard: safety requirements for medical electrical
systems
IEC 60601-1-2 Medical Electrical Equipment Part 1: General requirements for safety
– Collateral Standard: electromagnetic compatibility.
IEC 60601-1-6 General requirements for safety – Collateral Standard: Usability
 Reliability in Medical Device Industry 993

Standard Reference No. Standard Title / Description

IEC 60601-1-8 General requirements for safety – Collateral Standard: General
requirements, tests and guidance for alarm systems in medical
electrical equipment and medical electrical systems
IEC 68-2-64 Environmental Testing Part 2:Test methods Test Fh: Vibration,
broad-band random (digital control) and guidance
IEC 1123 Reliability testing compliance test plans for success ratio
IEC 605 Equipment reliability testing
UL544 Standard for safety of medical and dental equipment
UL 2601-1 Medical Electrical Equipment, Part1: General Requirements for
Safety
MIL HDBK-338 Electronic reliability design handbook
MIL HDBK 217F Reliability prediction of electronic equipment
MIL-STD-1629A Procedures for performing a failure mode, effects and criticality
analysis
MIL HDBK-781 Reliability test methods, plans, and environments for engineering
development, qualification and production
MIL-STD-2155 Failure reporting, analysis and corrective action system (FRACAS)

References [5] Rand Corporation. The cost and benefits of

[1] Borger C, et al., Health spending projections
through 2015: Changes on the horizon. Health
Affairs Web Exclusive W61 2006; :22February,S.
[2] Pear R., US health care spending reaches all-time
high: 15% of GDP. The New York Times 2004;
January 9; 3.
[3] Wan H, Sengupta M, Velkoff V, DeBarros K. 65+
in the United States: 2005. U.S. Census Bureau.
[online], 2005;1-70. Available:
http://www.census.gov/prod/2006pubs/p23-
209.pdf
[4] Khan Z. Medical Electronic products call for
higher reliability. ECN, 3/1/2005. Available
http://www.ecnmag.com/article/CA508377.html
reliability in military equipment, 1988.
[6] Schenkelberg F. Reliability integration 3-day
training course. Ops A La Carte LLC, Oct, 2004.
[7] Dhillon BS. Medical device reliability and
associated areas, CRC Press, 2000.
[8] Mills HD. On the statistical validation of computer
programs. IBM Federal Systems Division,
Gaithersburg, MD, Report No. 72-6015, 1972.
[9] CRE Primer 1998, Quality Council of Indiana.
[10] O’Connor PDT. Practical reliability engineering.
(3rd edition). John Wiley and Sons New York,
1996.
[11] Systems reliability division. Rome Laboratory
Reliability Engineer’s Toolkit: New York. 1993;
April.

View publication stats