Pediatric Anesthesia ISSN 1155-5645
`
`ERRORS: REVIEW ARTICLE
`
`Medication errors - new approaches to prevention
`
`Alan F. Merry & Brian J. Anderson
`
`Department of Anaesthesiology, University of Auckland, and Auckland City Hospital, Auckland, New Zealand
`
`Keywords
`children; anesthesia; medication errors;
`prevention
`
`Correspondence
`Professor Alan Merry,
`Head, Department of Anaesthesiology,
`University of Auckland School of Medicine,
`Auckland, New Zealand
`Email a.merry@auckland.ac.nz
`
`Section Editor: Charles Cote
`
`Accepted 18 March 2011
`
`doi:10.1111/].1460-9592.2011.03589.x
`
`Summary
`.
`.
`~
`.
`~
`.
`.
`.
`Medication errors in pediatric anesthesra represent an important risk to
`children. Concerted action to reduce harm from this cause is overdue. An
`
`understanding of the genesis of avoidable adverse drug events may facili-
`tate the development of effective countermeasures to the events or their
`effects. Errors include those involving the automatic system of cognition
`and those involving the reflective system. Errors and violations are distinct,
`but violations often predispose to error. The system of medication adminis-
`tration is complex, and many aspects of it are conducive to error. Evi-
`dence-based practices to reduce the risk of medication error in general
`include those encompassed by the following recommendations: systematic
`countermeasures should be used to decrease the number of drug adminis-
`tration errors in anesthesia;
`the label on any drug ampoule or syringe
`should be read carefully before a drug is drawn up or injected; the legibility
`and contents of labels on ampoules and syringes should be optimized
`according to agreed standards; syringes should always be labeled; formal
`organization of drug drawers and workspaces should be used; labels should
`be checked with a second person or a device before a drug is drawn up or
`administered. Dosage errors are particularly common in pediatric patients.
`Causes that should be addressed include a lack of pediatric formulations
`and/or presentations of medication that necessitates dilution before admin-
`istration or the use of intravenous formulations for oral administration in
`
`children, a frequent failure to obtain accurate weights for patients and a
`paucity of pharmacokinetic and pharmacodynamic data. Technological
`innovations, including the use of bar codes and various cognitive aids, may
`facilitate compliance with these recommendations.
`Improved medication
`safety requires a system-wide strategy standardized at least to the level of
`the institution;
`it is the responsibility of institutional leadership to intro-
`duce such strategies and of individual practitioners to engage in them.
`
`Introduction
`
`Medication errors are common in anesthesia practice
`(1) and the perioperative setting (2—4). Many cause lit-
`tle or no harm, but some have devastating conse-
`quences for patients and, on occasion, for practitioners
`(1,5—7). In this study, we define error and differentiate
`it from violation; we present a brief introduction to
`the frequency and nature of medication errors in pedi-
`atric anesthetic practice. We discuss the nature of
`error. We outline some new approaches to the preven-
`tion of these errors.
`
`In doing this, we place emphasis on understanding
`.
`the nature of error. We introduce the concept of error
`management, of which the prevention of error is only
`one part. We include a brief discussion of blamewor-
`thiness in relation to medication errors because adop-
`tion of the principle of a just culture is an important
`advance in the promotion of patient safety. Errors
`may be blameless, but violations often play a part in
`their genesis, and violations may reflect carelessness. In
`the pursuit of patient safety, it is essential to focus on
`the system, but the mind-set of individual practitioners
`is also of central importance.
`
`Pediatric Anesthesh 21 (2011) 743—763 © 2011 Blackwell PLbiishing Ltd
`
`743
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLc - Exhibit 1042 — Page 743
`
`
`
`

`

`Medication errors
`
`A.F. Merry and B.J. Anderson
`
`Errors, violations and other definitions
`
`Box 1 A medication error with a good outcome
`
`An error may be defined as ‘the unintentional use of a
`wrong plan to achieve an aim or failure to carry out a
`planned action as intended’ (5). Stated simply, this can
`be expressed as follows: ‘an error is when someone is
`trying to do the right thing but actually does the
`wrong thing’
`(8). A violation, on the other hand,
`involves the ‘deliberate—but not necessarily reprehen-
`sible—deviation from those practices appreciated by
`the individual as being required by regulation or neces-
`sary or advisable to achieve an appropriate objective,
`while maintaining safety and the ongoing operation of
`a device or system’ (5). The essential difference is that
`violations involve an element of choice, while errors
`do not. It follows that errors are not, in themselves,
`blameworthy (9). However, violations often predispose
`to error, and a predisposing violation may be blame-
`worthy even if an associated error is not. The blame-
`worthiness of a violation depends on context and
`degree.
`Wheeler provides the following definition of medica-
`tion error: ‘A medication error is an ‘‘error in the pre-
`scription,
`dispensing,
`or
`administration
`of
`a
`medication with the result that the patient fails to
`receive the correct drug or the indicated proper drug
`dosage’’’ (10). This definition includes outcome. There
`is a natural tendency to judge events on their outcome,
`but it is both more logical and more conducive to
`patient safety, to evaluate errors by the cognitive pro-
`cesses involved in their generation. It is not necessary
`to have an adverse outcome for an action or decision
`to be an error (see Box 1). It is also sensible to refer-
`ence types of error to a single definition of the word.
`Therefore, we define a medication error as ‘any error
`involving the prescribing, ordering,
`selection, or
`administration of a medication.’
`Medication errors do not necessarily involve culpa-
`ble or blameworthy elements, but some might. The
`choice to practice according to the best available evi-
`dence is a key element of safe medication management.
`This includes evidence about the processes of medica-
`tion administration as well as evidence about
`the
`choices of medications for any given clinical setting. In
`this context, making every reasonable effort to practice
`safely does not guarantee an absence of error, and the
`occurrence of an error does not of itself provide any
`reason to suspect negligence or other forms of blame-
`worthy behavior. However, choosing to ignore simple
`steps
`(such as
`routinely labeling syringes) widely
`accepted as important for safety may be construed as
`a violation and might reasonably be considered blame-
`worthy (9). In recent years, the concept of a ‘no-blame’
`
`An anesthetist decided to administer a bolus of ephedrine to
`treat the unexpected sudden onset of profound hypotension in
`a simulated patient. He accidentally administered 100 lg of
`epinephrine (the two drugs were presented in similar formats).
`It then became apparent that the cause of the hypotension was
`a (simulated) accidental total spinal anesthetic and that the indi-
`rectly acting ephedrine may well have been ineffective. Thus,
`epinephrine, which is directly acting on receptors, was arguably
`a better choice, and this error might well have produced a
`better outcome than the intended action. Nevertheless,
`it
`would still have been a failure to carry out a plan as intended.
`In fact, the plan itself would have been imperfect – so this is
`an illustration of one error canceling another.
`
`approach to error in healthcare has given way to that
`of a ‘just culture’. In a just culture, the accountability
`of individuals is evaluated in light of an informed view
`of human cognition and a recognition that many, but
`not all, of the things that go wrong in health care have
`no element of blameworthiness (8,9).
`An adverse drug event (ADE) is ‘any injury related
`to the use of a drug’ (11). This should include any
`injury related to the omission of an indicated medica-
`tion. Some, but not all, ADEs are preventable. Some
`preventable ADEs are caused by errors alone, but in
`others, violations may also play a part.
`
`The frequency of medication errors in pediatric
`anesthesia
`
`There have been many estimates of the frequency of
`medication error in anesthesia in adults (1) and a
`number in pediatrics (12,13). Reports of medication
`errors in children typically come from subsets of adult
`orientated studies and tend to lack detail. Many
`reports lack denominator data and/or are based on
`incident reports in which it is difficult to know how
`many actual events were reported. It is also likely
`that at least some medication errors are made without
`the practitioner being aware of having made them.
`Facilitated incident reporting of errors in adult anes-
`thesia practice suggests a rate of approximately one
`error per 135 anesthetics (1). In patients in the neona-
`tal intensive care unit and postnatal wards of a uni-
`versity-affiliated urban general hospital
`in New
`Zealand, ADEs occurred at the rate of 2.1 per 100
`prescription episodes, 12.9 per 100 admissions and
`22.1 per 1000 patient days (14). About half were clas-
`sified as preventable, and the estimated annualized
`cost of
`these was approximately NZ$87 000. The
`greatest rate per 1000 patient days was seen on the
`surgical pediatric ward. Nearly half of the ADEs were
`
`744
`
`Pediatric Anesthesia 21 (2011) 743–753 ª 2011 Blackwell Publishing Ltd
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1042 – Page 744
`
`

`

`A.F. Merry and B.J. Anderson
`
`Medication errors
`
`classified as serious, and 15% resulted in persistent
`disability or were considered life-threatening. In addi-
`tion, many potential ADEs were identified. Medica-
`tion errors in children appear to be more common in
`intensive care settings than other clinical areas, and
`approximately 12% appear to result in harm; children
`are at higher risk in the perioperative setting than
`adults for harm from this cause (2–4). Thus, there is
`every reason to believe that these errors represent an
`important risk to pediatric patients, before, during,
`and after anesthesia.
`
`The nature of medication error in children
`
`Children undergoing anesthesia are subject to the same
`errors as adults, but they may be more vulnerable than
`adults because of immaturity of physiological systems,
`and several factors increase the likelihood of medica-
`tion errors in children. These are discussed latter. Simi-
`larly, medication errors during anesthesia may be seen
`as simply a subset of the overall problem of medica-
`tion errors in hospital, but there are some differentia-
`ting features of this setting. Anesthesia is the only area
`where medications are typically prescribed, prepared,
`administered, and recorded by a single individual (an
`anesthesiologist or anesthetist) without any other
`health professional to check or monitor the process.
`This fact increases the risk to patients of an error, but
`it also creates an unparalleled opportunity for the mis-
`use of controlled drugs. In addition, it may create for
`practitioners some risk of being wrongfully suspected
`of misusing drugs.
`Medication errors (in any age group) may occur
`through commission or omission. The former involves
`the wrong drug, the right drug inadvertently repeated
`(so-called insertion errors),
`the wrong dose,
`the
`wrong route, or the wrong time. In errors of commis-
`sion, harm may occur through unintended effects of
`incorrect actions
`[e.g.,
`inadvertently administering
`dopamine instead of doxapram has caused cardiac
`arrest and death (9)]. In errors of omission harm
`may occur through the absence of intended effects
`(e.g., nosocomial infection after the omission of pro-
`phylactic antibiotics, or awareness during inadequate
`anesthesia).
`Medication errors dominated critical incidents rele-
`vant to pediatric anesthesia reported recently to the
`UK National Reporting and Learning System: unin-
`tentional additional medication doses were the most
`prevalent, but wrong drug, wrong dose, and wrong
`route errors were also common; analgesics and anti-
`biotics were the commonest medications involved in
`these errors (6). In reports from intensive care or high
`
`dependency units, 61% of medication incidents were
`associated with drug administration and 26% with pre-
`scription (7).
`Dosage errors of one sort or another are particularly
`common in children (15,16). Growth, maturation, and
`size are critical determinants of dose. Clearance, the
`pharmacokinetic
`parameter
`dictating maintenance
`dose, is immature at birth and matures over the first
`few years of life. Bupivacaine toxicity in infants receiv-
`ing
`continuous
`regional neuronal blockade has
`occurred through failure to appreciate immature clear-
`ance (17). Clearance has a nonlinear relationship to
`weight (18): when clearance is expressed using a linear
`)1Ækg
`)1), it is highest in the 1 to 2-
`function (e.g., lÆh
`year-old age band, decreasing throughout childhood
`until adult rates are achieved in late adolescence. Drug
`)1) will typ-
`dose scaled directly from adult dose (mgÆkg
`ically be inadequate. Failure to recognize this point
`has resulted in inadvertent systematic underdosing of
`HIV-infected children in the United Kingdom and Ire-
`land with antiretrovirals (19). The use of remifentanil
`parameters derived from adult studies for infusions in
`children results in lower concentrations than antici-
`pated because clearance expressed per kilogram is
`higher in children (20). This may actually be advanta-
`geous, because hypotensive effects will be correspond-
`ingly less pronounced (21).
`Pharmacodynamics has been inadequately studied in
`children and especially in infants. This is partially
`attributable to a lack of effect measures (e.g., spectral
`edge frequency, bispectral index, entropy, and cerebral
`state index) in young people undergoing anesthesia.
`The paucity of
`integrated pharmacokinetic–pharma-
`codynamic (PKPD) studies involving total intravenous
`anesthetics (22) predisposes to increased awareness in
`children (23,24). There is not yet an adequate means
`of monitoring the effect of anesthesia in infants, and
`this also contributes to anesthetic vapor administration
`errors, a major cause of morbidity and mortality with
`halothane (25).
`Infants are unable to swallow pills, but pediatric
`oral formulations are not available for the majority of
`commercially available medications. When no liquid
`oral formulation is available, intravenous preparations
`are often administered orally (e.g., midazolam, pro-
`pranolol), without adequate information about their
`hepatic extraction ratio or the effect of the diluent
`used to improve palatability: this may lead to inappro-
`priate dose (26).
`Children generally require smaller doses than adults.
`Because medications are packaged for adult use, dilu-
`tion is commonly required in anesthesia practice. This
`
`Pediatric Anesthesia 21 (2011) 743–753 ª 2011 Blackwell Publishing Ltd
`
`745
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1042 – Page 745
`
`

`

`Medication errors
`
`A.F. Merry and B.J. Anderson
`
`predisposes to dosage errors (15,16), often in the form
`of 10-fold overdoses (27).
`Technique is particularly important in the adminis-
`tration of medications to small children and babies.
`Part of
`the volume of medication may easily be
`retained in the dead space of any part of an intrave-
`nous administration set or syringe, with the result that
`the desired effect may not be obtained. Alternatively,
`additional doses of medication may be given inadver-
`tently on account of this dead space medication, and
`the effect may then be excessive and potentially lethal
`(28). Apnea, bradycardia, hypotension, and hypotonia
`have been reported in a premature neonate weighing
`1.6 kg after an overdose of morphine, arising from the
`additional medication in the ‘dead space’ of the syringe
`(29).
`Dilution may also be necessary to maintain osmolal-
`ity low enough to prevent phlebitis
`(i.e., below
`)1). The amount of diluent required for
`600 mosmolÆl
`drugs such as pentamidine and ganciclovir increases
`the risk of heart failure in critically ill children (30,31).
`Although medications are usually prescribed on a
`)1), children are not often
`weight basis (e.g., in mgÆkg
`weighed. A survey of 100 children’s notes in a busy
`emergency department revealed that only 2% were
`weighed prior to prescribing (32). Twenty-nine per cent
`of physicians’ estimates, 40% of nurses’ estimates, and
`16% of parents’ estimates differed from actual weight
`by more than 15% (33). The accuracy of methods to
`estimate weight also varies (34,35).
`
`Understanding medication errors
`
`Error
`We think it reasonable to assume that essentially all
`anesthesia providers are motivated to administer medi-
`cations correctly. Thus, ADEs seldom occur because
`of any intention to harm. Rather, they occur because
`of
`the inherent human propensity for error and
`because the system by which medications are adminis-
`tered during anesthesia (see Table 2) is complex and
`tightly coupled (36,37), with numerous latent factors
`predisposing to failure (38).
`Reason has described a General Error Model (39)
`that is based on an attempt to understand the cogni-
`tive processes at play when error occurs. There is
`much theoretical and empirical research to inform
`such an understanding.
`In his model, errors are
`divided into action failures (slips and lapses) and
`decision (or planning) failures (rule-based or knowl-
`edge-based errors)
`(8,39): decision failures are also
`known as mistakes. More recently, Thaler and Sun-
`stein (40) have shifted the emphasis from the distinc-
`
`they
`Instead,
`tion between actions and thoughts.
`explain that
`there are two systems with which we
`both think and act. The automatic system is uncon-
`trolled, fast, effortless, associative, unconscious, and
`skilled. Slips and lapses (39) occur in this cognitive
`mode, but some rule-based decisions may also fall
`into this category. Human cognition functions pre-
`dominantly through pattern recognition, and many
`skilled activities depend on recognizing complex situa-
`tions as a whole (i.e., as patterns) and responding by
`reference to prestored schemata (which are, in effect,
`also patterns), within the subconscious memory (39).
`This is highly efficient, but may lead to error for
`many reasons,
`including distraction,
`incorrect inter-
`pretation of situations, and poor training (leading to
`the learning of unsafe automatic responses to certain
`situations). The reflective system, by contrast, is con-
`trolled,
`effortful,
`slow, deductive, and conscious.
`Rule-based decisions may be conscious or uncon-
`scious (i.e., they may fall
`into either system), but a
`key feature of these decisions is that they are induc-
`tive and feed-forward (39). Reflective decisions [also
`called knowledge-based (39) or deliberative (9) deci-
`sions] are deductive, and progress is made through
`evaluating feedback and through an iterative process
`of trial and error. The term knowledge-based error
`places emphasis on the incomplete picture one often
`has of
`the world. A practitioner’s own knowledge
`(acquired from training and experience)
`is only a
`small part of this. There are many other sources of
`important information, such as patients’ notes or the
`knowledge of other people, but
`these may not be
`accessible to (or accessed by) the person making a
`decision about which medication to administer (see
`Box 2).
`A key point emerges clearly from the empirical and
`theoretical research into error: simply trying harder to
`avoid error, on its own, is unlikely to be successful.
`
`Box 2 A lapse leading to a knowledge-based (or deliberative)
`error
`
`A medication allergy was known to a patient (who had a compli-
`cated history with many comorbidities) and documented in the
`notes. A resident anesthesiologist read the notes, but in an
`otherwise comprehensive hand over, forgot to communicate
`the allergy to an attending anesthesiologist who took over the
`case halfway through the surgery. This was a lapse (39).
`In
`consequence, a decision by the attending to administer the
`medication in question was based on inadequate knowledge:
`this was a knowledge-based error (39), also known as an error
`of deliberation or a deliberative error (8). The example illus-
`trates how one error can lead to another. Failures of this sort
`have been identified as particularly difficult to address with
`technology (9).
`
`746
`
`Pediatric Anesthesia 21 (2011) 743–753 ª 2011 Blackwell Publishing Ltd
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1042 – Page 746
`
`

`

`A.F. Merry and B.J. Anderson
`
`Medication errors
`
`Indeed, it may even have the ironic effect of increasing
`the risk of certain errors (41,42). Conscientious deter-
`mination to practice safely does have a role in the
`avoidance of violations, but it will not, of itself, guar-
`antee the avoidance of pure error: if medication error
`is to be reduced, at least some redesign of the methods
`used to administer medications will also be necessary.
`
`The objectives of medication administration – the
`traditional view
`
`The objectives of medication administration are often
`summarized in the ‘five rights’: the right patient, dose,
`medication, time and route of administration (43). We
`suggest a sixth ‘right’: a right (or accurate and compre-
`hensive) record of the medications administered and of
`any medications wasted (notably unused portions of
`ampoules of controlled drugs). It is, perhaps, under-
`standable that less attention is paid to errors in record-
`ing than to errors in prescribing or administering
`drugs, but an accurate and complete medication record
`may be important clinically, for audit, for inventory
`management or medico-legally. There are many rea-
`sons for errors in recording (44). One that is particu-
`larly relevant to the sixth ‘right’ is that people tend to
`record the medications they believe they have adminis-
`tered. In fact, many medication errors pass unnoticed.
`It is also true that many medication errors do not, in
`the end, cause harm. But these two points are not syn-
`onymous. For example, one anesthesia provider might
`administer a dose of gentamicin, fail to record this,
`and then hand over the anesthetic to another, who
`gives what he believes to be the sole dose of the drug.
`This error might never be identified and will not be
`reflected in the record, but otological or renal harm
`may well result. Another relevant point is that anesthe-
`sia providers who wish to misuse controlled medica-
`tions may find it easy to falsify their records to
`conceal their misuse. Thus, the accurate tracking of
`medications administered during anesthesia is part of
`careful medication practice.
`It is noteworthy that, in the USA, the Joint Com-
`mission has issued new standards that place greater
`emphasis than before on safety in medication manage-
`ment (45) and that at least some recent surveys have
`targeted medication management during anesthesia
`(46).
`
`The objectives of medication administration – a
`modified view
`
`The emphasis on the rights of medication administra-
`tion is useful, but an excessive focus on avoidance of
`
`error may prove counterproductive. The goal is patient
`safety. Measures to mitigate the consequences of error
`may be more effective than undue efforts to avoid all
`errors: the obvious analogy here is provided by airbags
`in automobiles – but error reducing measures such as
`speed limits also matter. There are several ways to
`reduce the consequences of error.
`Not all errors are equal in their potential to cause
`harm, and it seems sensible to stratify the relative risk
`of different types of error and consider exchanging
`dangerous errors for less dangerous ones. Using color
`coding by class of drug may not reduce the number of
`errors, but within-class errors may be less likely to
`cause serious harm than between-class errors.
`Removing concentrated potassium chloride from the
`operating room reduces the potential for one particu-
`larly dangerous type of medication error.
`Mitigation of medication error is more likely to be
`possible if one knows that an error has been made.
`Given the relatively high risk of awareness arising
`from medication errors in anesthesia, monitoring con-
`sciousness is an obvious element of an overall error
`management strategy. Measures to objectively track
`medications
`that have been administered are also
`important. One such method involves the retention of
`all used ampoules and vials in an orderly fashion (9),
`beginning afresh for each anesthetic. In this way, any
`doubt about what might or might not have been given
`can readily be resolved. One can also use prompts.
`These may be simple (e.g., placing an ampoule of anti-
`biotic on the working surface ready to be administered
`at the appropriate time) or may be provided through
`technology (e.g.,
`computer generated prompts
`to
`administer antibiotics).
`The use of bar codes at the time of medication
`administration (9) seems likely to result in more accu-
`rate record keeping than manual records or computer-
`ized systems that depend on retrospective entry of
`medication names and times of administration.
`
`A systems view of medication administration
`error
`
`The process of medication administration involves
`many steps (47), beginning at the manufacturing stage,
`and ending with safe disposal of empty ampoules and
`incompletely used medications, and an accurate record
`(Table 2). Attention to the overall system of medica-
`tion administration is important in preventing medica-
`tion errors; many of the issues have been discussed in
`greater detail elsewhere (1,9,48). A systematic review
`of the literature identified 98 relevant references (14
`with experimental designs or incident reports and 19
`
`Pediatric Anesthesia 21 (2011) 743–753 ª 2011 Blackwell Publishing Ltd
`
`747
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1042 – Page 747
`
`

`

`Medication errors
`
`A.F. Merry and B.J. Anderson
`
`Table 1 Some of the strategies for reducing the risk of medication error in anesthesia identified through a systematic review of the litera-
`ture, modified from Jensen et al. (49); strategies 1–6 were strongly supported. See also the guideline promulgated by the Australian and
`New Zealand College of Anaesthetists (90) and the National Patient Safety Agency (91)
`
`1. Systematic countermeasures should be used to decrease the number of drug administration errors in anesthesia.
`2. The label on any drug ampoule or syringe should be read carefully before a drug is drawn up or injected.
`3. The legibility and contents of labels on ampoules and syringes should be optimized according to agreed standards (92,93).
`4. Syringes should always be labeled (or almost always: if, during the process of drawing up and administering a single medication, the
`syringe never leaves the practitioner’s hands, a case can be made that a syringe label is not necessary, but it is probably safer simply
`to label all syringes).
`5. Medication drawers and workspace should be formally organized, and potentially hazardous medications (e.g., epinephrine, halothane,
`bupivacaine) not used during routine and uneventful anesthetics should be separated from those that are (in another drawer, or
`outside the OR).
`6. Labels should be checked with a second person or by means of a device (such as a bar code reader linked to a computer) before any
`medication is drawn up or administered.
`7. Errors in intravenous drug administration during anesthesia should be reported and regularly reviewed.
`8.
`Inventory management should focus on minimizing the risk of drug error: there is a strong case for designating a pharmacist to the
`operating theaters (53), and any changes in presentation should be notified ahead of time.
`9. Similar packaging and presentation of medications should be avoided where possible.
`
`with reports of cases or case series). A number of evi-
`dence-based recommendations
`for
`safer medication
`practice were formulated and tested against a database
`of
`incident
`reports
`involving medication
`error
`(Table 1) (49). Many of these addressed the wider sys-
`tem but some, notably the requirement to read the
`label of every ampoule and syringe, were directed to
`individual practitioners. This emphasizes the point that
`there is currently no obvious way to make medication
`errors inherently impossible simply by reengineering
`the system. Improvements can certainly be made, but
`few of the actions that can be taken in this context are
`effective in the absence of engaged care by clinicians.
`Pin indexing (e.g., of gas cylinders and vaporizer fill-
`ers) is an example of a forcing function that has little
`dependency on users, but initiatives such as improving
`the design of syringe labels will only increase safety if
`the labels are applied and read. Within a ‘just culture’,
`a deliberate and sustained refusal to do this may not
`be considered entirely blameless.
`
`The importance of measuring and reporting
`
`Measurement is fundamental to quality improvement
`(50). Large pediatric anesthesia outcome studies have
`identified factors associated with increased risk, such
`as age <1 year (25). Improved training and facilities
`for those working with this age group has lead to a
`decrease in mortality (51,52). The use of trigger tools
`has recently gained credibility in the measurement of
`errors
`(53). Critical
`incident
`reporting without a
`denominator may be sufficient to promote improved
`safety. Reviewing errors and ADEs can be highly
`motivating, and simply identifying a problem can lead
`
`to change (54–56). For example, in one study, anesthe-
`sia-related complications from medication errors were
`reduced from 37% in the 1994–1997 period to 22% in
`the 1998–2003 period; awareness of complications
`because of halothane lead to an embracing of the safer
`and more easily titratable sevoflurane, and this alone
`was a major contributor to a reduction in medication
`error reporting (57).
`
`New approaches to prevention of medication
`errors
`
`Medication safety depends on adding as many barriers
`to the accident path as possible – in the Swiss Cheese
`analogy (39), the chance of the holes lining up is
`reduced by increasing the number of slices. Therefore,
`effective reduction in patient harm from medication
`errors requires more than just a commitment to do
`better on the part of each practitioner, although this
`element
`is certainly important. An institution-wide
`commitment and strategy are the basis for a worth-
`while and sustained improvement in medication safety.
`The starting point is to develop a shared state of mind
`in which each practitioner recognizes the possibility,
`and potential importance, of these errors in his or her
`own practice. The whole system (Table 2) should be
`reviewed, and pharmacists, administrators, nursing
`staff, and other relevant people were recruited to the
`cause. Ideally, agreement should be reached on a high
`level of rationalization and standardization for each
`part of this process, at least within any hospital, but
`preferably throughout any region in which trainees or
`other anesthesia providers move from institution to
`institution.
`
`748
`
`Pediatric Anesthesia 21 (2011) 743–753 ª 2011 Blackwell Publishing Ltd
`
`Petition for Inter Partes Review of US 8,338,470
`Amneal Pharmaceuticals LLC – Exhibit 1042 – Page 748
`
`

`

`A.F. Merry and B.J. Anderson
`
`Medication errors
`
`Table 2 Some aspects of the system of medication administration within a hospital
`
`Step
`
`Comments
`
`Manufacture
`
`Regulation
`
`Selection of stock
`
`Inventory management
`
`Presentation to users
`Process tools
`
`Technique of
`administration
`Recording
`
`Disposal
`
`Design and ergonomics of presentation are important: easily stored, easily read, easily opened ampoules or
`files are good, and prefilled, prelabeled syringes would be better.
`Many aspects of medication safety are covered by regulation (e.g., stability of the medication in the ampoule).
`The FDA requirement for bar codes on unit doses is an example of useful measure to reduce medication
`error.
`Institutions may often be able to choose well-presented medications and to avoid look-alike sound-alike
`(LOSA) formulations (94,95). Having anesthesia input into this choice is helpful.
`This is part of efficient and safe medication management; all essential medications obviously need to be kept
`in stock and readily available to users, but so do required user-applied labels and other tools of safe process
`(some institutions use red plungers on syringes to identify neuromuscular blocking agents, for example;
`running out of these probably creates greater risk than not using them at all).
`A consistent, well-organized workspace in which medications are clearly presented to users is important.
`Well-designed user-appl