Medication related litigation in Ireland: A 6‐year review

Abstract

Aims

The primary aims of the study were to identify those medications most frequently associated with clinical litigation in Ireland and to quantify the cost of such litigation. Secondary aims were to identify where in the medication‐use process claims were most likely to arise, the medication incident types involved and the primary injury alleged.

Methods

The National Incident Management System (NIMS) for incident and claims management was searched to identify all medication‐related claims finalised from 2011 to 2016 (inclusive). The physical case files were obtained and additional data not available on NIMS was extracted in order to build a detailed picture of the incident and subsequent claim.

Results

The search identified 79 relevant claims, of which 48 closed with a payment to the plaintiff. These 48 claims involved 54 medications. Medication groups identified included general anaesthetics (n = 7), opioids (n = 6), penicillins, antithrombotics and local anaesthetics (all n = 5). The errors alleged occurred exclusively at the administration (58%) and prescribing (42%) stages of the medication‐use process. Medication incident types included wrong dose/strength (n = 17), wrong drug (n = 7) and adverse drug reaction (n = 6). The most commonly pleaded primary injuries were allergic reaction (n = 9), deterioration in clinical status (n = 9) and post‐traumatic stress disorder (n = 8). The median total cost of these claims was €60 991, including median damages of €33 858.

Conclusions

This study links data on medication incidents, actual harm to patients and litigation costs. Thus, it presents a comprehensive picture of the consequences of medication error.

What is already known about this subject

• Medication errors are common and costly.

• The quality of data in medication incident reports is often poor.

• Analysis of medication‐related closed claims offers the opportunity to learn from medication error and to determine the financial impact of such claims.

What this study adds

• The medication groups most commonly involved in clinical claims include general anaesthetics, opioids, penicillin antibiotics, antithrombotic agents and local anaesthetics.

• The most frequently pleaded injuries in medication‐related litigation include allergic reaction (including anaphylaxis), deterioration in clinical status and post‐traumatic stress disorder.

• The median total cost, in purely financial terms, of a medication‐related claim that closed with a payment to the plaintiff was €60 991, including median damages of €33 858.

1. INTRODUCTION

An adverse drug event (ADE) is defined as any injury resulting from drug‐related administration.1 ADEs include adverse drug reactions (ADRs), many of which are neither predictable nor preventable, and medication errors (MEs) which are preventable. MEs are defined as any preventable event that may cause or lead to inappropriate medication use or patient harm while the medication is in the control of the healthcare professional, patient or consumer.2 The terms medication error and medication incident are similar.2 High Alert medications are defined as medications that, when given in error, have the highest likelihood of causing harm.3 The medication‐use process has been described as a 5‐stage process encompassing medication prescribing, transcribing, dispensing, administration and monitoring.1

The estimated rate of MEs in UK hospitals is between 1 and 10%,4 while it has been suggested that 20% of doses administered in US hospitals are associated with an error.3 However, error rates in high dependency settings and in certain high risk processes can be significantly higher. The error rate in paediatric emergency departments has been reported to be 39%,1 while the rate in the administration of intravenous medications has been reported to be as high as 49%.4

A recent study estimated that there were 237 million MEs at the various stages of the medication‐use process in England annually, of which 25.8% had the potential to cause moderate harm and 2.0% had the potential to cause serious harm.5 In Ireland, medication incidents accounted for 14.7% of the 10 most common clinical incidents reported by publicly funded healthcare organisations over a 5‐year period.6

The cost of MEs has been estimated at $3.5 billion annually in the USA7 and $42 billion annually worldwide.8 The annual incremental cost of inpatient preventable ADEs associated with injectable medications in the USA has been estimated to be between $2.7 billion and $5.1 billion.9 Avoidable ADRs cost the National Health Service (NHS) in England £98.5 million (€114.4 million approximately) annually, accounting for 181 626 bed‐days and causing and contributing to 712 and 1708 deaths respectively.5 A systematic review of the economic impact of ME found the cost per error, adjusted to euro 2015 values, ranged from €2.58 to €111 727.10 Closed claims analysis has been employed to determine litigation costs associated with MEs. A study of litigation resulting from drug errors in anaesthesia in the NHS identified 93 claims with a total cost of £4.9 million (€5.7 million approximately).11

Although previous studies have provided data on the prevalence of MEs in Irish hospitals,6, 12 there is little data on the medications, or groups of medications, involved in clinical claims in Ireland. Analysis of claims data provides a valuable opportunity to learn from adverse events in healthcare and to determine the financial impact of litigation related to such claims.11

1.1. Background

The State Claims Agency (SCA) provides indemnity cover for publicly funded healthcare organisations in the Republic of Ireland and manages clinical negligence claims taken against these organisations and their staff. The SCA hosts the National Incident Management System (NIMS), a web‐based incident and claims management solution (see Box 1). Indemnified healthcare organisations have a statutory obligation (under the National Treasury Management Agency [Amendment] Act 2000) to report adverse incidents on NIMS. NIMS is thus the principle source of national data on incident and claims activity for the Irish health service.6

Box 1. The National Incident Management System (NIMS) explained.

What is the NIMS? NIMS is an end to end risk management web‐based system. NIMS facilitates the reporting of incidents (which include claims), to the State Claims Agency. It is hosted by the State Claims Agency for the Health Service Executive, other healthcare enterprises and State Authorities.6

Each year, a small number of claims related to medications are initiated against publicly funded healthcare organisations in Ireland and are managed on behalf of these organisations by the SCA. Claims may proceed to settlement with a payment to the plaintiff. However, many claims do not proceed for various reasons including that the claim is statute barred or that the claim is intimated but not pursued. In addition, claims may be dismissed in court. A finalised claim is one in which all matters associated with the claim, including costs, have been agreed but some associated payments and reimbursements may still be outstanding.6 A closed claim, by comparison, is one in which litigation has concluded and all associated financial matters have been paid. Claims are a rich source of information on ME as they contain copies of the original medical notes and prescriptions, along with witness statements and expert reports. This study provides an analysis of medication‐related claims managed by the SCA which were finalised between 2011 and 2016 (inclusive).

1.2. Aims

The primary aims of this study were to identify the medications, or groups of medications, most frequently associated with clinical litigation in Ireland and to quantify the cost of such litigation. Secondary aims were to identify the stage or stages of the medication‐use process where claims are most likely to arise, the medication incident types involved, the primary injury alleged and the incident severity.

2. METHODS

The NIMS database utilises the World Health Organisation's Anatomic Therapeutic Chemical (ATC) classification system for drugs. This classes drugs according to the organ or system on which they act and their therapeutic, pharmacological and chemical properties. At level 1, drugs are divided into 14 main anatomical groups, at level 2 into therapeutic subgroups and at level 3 into pharmacological subgroups.13 NIMS facilitates the recording of the medications involved in these claims and classifies them according to ATC levels 1–3.

A report was generated on NIMS where the Sub Hazard type was medication and the claim finalised date was between 1 January 2011 and 31 December 2016 (inclusive). Relevant data, including file numbers, were extracted from NIMS and exported to an Excel spreadsheet. The physical case files associated with the claims were obtained and underwent detailed manual analysis. Additional data, not available on NIMS, were extracted from the case files using a data collection form (Appendix A1). These data were added to the Excel spreadsheet to build a detailed picture of the circumstances surrounding the incident and subsequent claim.

The claims data were analysed to determine the medications involved, the stage in the medication‐use process where the incident occurred, the medication incident type, the primary injury pleaded as a result of the medication incident and the severity rating of the injury. Severity rating is assigned by NIMS based on the outcome at the time of reporting.6 The outcome categories (with severity rating) are as follows: injury requiring medical treatment (moderate); long‐term disability/incapacity including psychosocial (major); permanent incapacity including psychosocial or death (extreme).

The financial data associated with these claims also underwent detailed analysis. To compare the cost of medication‐related claims to the cost of all clinical claims, the financial data for all claims initially identified (Figure 1) were compared with the financial data for all clinical claims over this 6‐year period, regardless of the outcome of the claim. This allowed the figures to be compared on a like‐for‐like basis. Separately, financial data for the medication‐related finalised claims that concluded with a payment to the plaintiff were analysed to allow a determination of median damages and median total costs for this cohort of claims. All financial figures quoted were correct as of 31 January 2018.

Figure 1.

Case outcome of medication‐related finalised claims 2011–2016

3. RESULTS

The report generated on NIMS identified 79 relevant claims over the 6‐year study period, including 1 linked claim—i.e. a claim linked to another claim—and these were counted as 1 claim for the purposes of the analysis. Many of these claims did not proceed for various reasons or were dismissed in court (Figure 1). There were 48 claims in which a payment was made to the plaintiff and the results presented here are based on the findings in relation to these claims unless otherwise stated.

3.1. Medications identified

A total of 54 medications were identified in the 48 claims, as a number involved >1 medication. The medications involved in these claims were classified at ATC level 1. Nervous system drugs (n = 24) were the largest group identified at this level, associated with 3 times more claims than the next most frequently encountered groups, which were anti‐infectives for systemic use (n = 8) and drugs acting on blood and blood forming organs (n = 7). The next most frequently encountered groups at this level were musculo‐skeletal system drugs (n = 4), alimentary tract and metabolism drugs (n = 3), and antineoplastic and immunomodulating agents (n = 3). Smaller numbers of drugs fell into other ATC level 1 categories.

At ATC level 3, general anaesthetics (n = 7) were the most frequently cited group followed by opioids (n = 6). Penicillins, antithrombotic agents and local anaesthetics were each cited in 5 claims. Antipsychotics were associated with 4 claims, with other therapeutic groups associated with smaller numbers of claims (see Figure 2).

Figure 2.

Most common medication groups classified at Anatomic Therapeutic Chemical level 3

Table 1 identifies the medications by generic name most commonly involved in these claims i.e. cited in 2 or more claims. The opioid anaesthetic, fentanyl, the opioid analgesic, morphine and the penicillin antibiotic, amoxicillin (+/− clavulanic acid) appear jointly at the top of this list (all n = 4). The local anaesthetic, bupivacaine (n = 3) is also notable. Other drugs associated with 2 claims each included methotrexate, suxamethonium chloride, sevoflurane, clozapine, diclofenac and insulin.

Table 1.

Medications involved in 2 or more claims

Generic name	Claim count
Fentanyl	4
Morphine sulfate/tartrate	4
Amoxicillin +/− clavulanic acid	4
Bupivacaine	3
Methotrexate	2
Suxamethonium chloride	2
Sevoflurane	2
Clozapine	2
Diclofenac	2
Insulin human	2

3.2. Stage of medication‐use process

Administration errors accounted for 58% (n = 28) of the alleged errors, with prescribing errors accounting for 42% (n = 20).

3.3. Medication error by type

Figure 3 shows the ME alleged in these claims by type. The largest error type was wrong dose/strength (n = 17). Wrong drug was the error type identified in 7 claims, adverse drug reaction and drug contraindicated were each associated with 6 claims, while omitted/delayed dose was the error type identified in 4 claims.

Figure 3.

Types of medication error identified in these claims

3.4. Injury pleaded as a result of the medication error

Figure 4 illustrates the most frequently pleaded primary injuries as a result of the ME. Allergic reaction (including anaphylaxis) and deterioration (both n = 9) were the most frequently pleaded injuries. Further analysis of allergic reaction claims revealed that 6‐were associated with β‐lactam antibacterials and the patient's allergy status was previously known in 5 of these cases. Deterioration was selected when there had been a deterioration in the patient's clinical status but is uninformative.

Figure 4.

Most frequently pleaded injuries in these claims

Eight claims pleaded post‐traumatic stress disorder (PTSD), half of which related to either anaesthetic awareness or inadequate epidural blocks. Likewise, analysis of 5 cases pleading anxiety/trauma revealed a further case of anaesthetic awareness and 2 further cases of inadequate epidural block. Claims relating to inadequate epidural block featured local anaesthetics in combination with another agent which, in 3 of 4 cases, was fentanyl. There were 4 cases where the injury was categorised as other as it did not fall into 1 of the offered categories. There were 2 claims where the primary injury pleaded was brain injury/concussion, 1 of which was associated with an anticoagulant, and 2 claims where the claimed injury was excessive bleeding associated with either an anticoagulant or antithrombotic.

3.5. Severity rating

Half of the alleged incidents had a moderate severity rating i.e. injury requiring medical treatment (n = 24; 50%). The remainder had a severity rating of major i.e. long‐term disability/incapacity (n = 14; 29.2%) or extreme i.e. permanent incapacity/death (n = 10; 20.8%).

3.6. Financial data

Utilising the data for all 79 claims initially identified, Table 2 shows that medication‐related claims accounted for 3.1% of all clinical claims by number over the study period but only 2.0% by cost.

Table 2.

Financial data for claims regardless of case outcome

	Medication‐related finalised claims	All clinical finalised claims	Medication related as a percentage (%)
Number of claims	79	2577	3.1
Paid total (€)	7 709 364	382 115 677	2.0

For the 48 claims that closed with a payment to the plaintiff, median (interquartile range [IQR]) damages were €33 858 (€45 388) and the median (IQR) total cost per claim was €60 991 (€135 253). Total cost of these claims over the 6‐year period studied was €7 291 968 (Table 3).

Table 3.

Financial data for claims which closed with a payment to the plaintiff (n = 48)

	Damages (€)	SCA legal costs (€)	Plaintiff legal costs (€)	Expert costs (€)	Total (€)
Total	3 906 958	1 052 770	2 187 974	144 266	7 291 968
Median	33 858	13 501	21 398	1130	60 991
IQR	45 388	26 310	53 060	2714	135 253
Mean	81 395	21 933	45 583	3006	151 916
SD	208 564	29 733	73 346	5088	308 249

IQR = interquartile range; SCA = State Claims Agency; SD = standard deviation.

Note that some rounding of figures has occurred.

4. DISCUSSION

This study has identified general anaesthetics followed by opioids and then penicillins, antithrombotics and local anaesthetics as the agents most commonly implicated in medication‐related litigation in Ireland. The errors that led to the claims occurred exclusively at the prescribing and administration stages of the medication‐use process. The most common error category identified was wrong dose/strength followed by wrong drug, adverse drug reaction, drug contraindicated and omitted/delayed dose. The most frequently pleaded primary injuries included allergic reaction including anaphylaxis, deterioration, PTSD, anxiety/trauma and other, where the alleged injury did not fit in 1 of the offered categories. The median total cost of these claims was €60 991, with median damages of €33 858.

The medication groups most commonly identified at ATC level 3 in this set of claims, with the exception of general anaesthetics, feature on a recently published starter formulary based on the top 100 drugs and classes in England suggesting that most of these medications are commonly used.14 Many also regularly appear on lists of High Alert medications. General anaesthetics, antithrombotic agents, opioids and epidural medications appear on the current Institute for Safe Medication Practices list of High Alert medications in acute care settings.15 Warfarin, other anticoagulants (both classed here as antithrombotics), opioids and antibiotics made up 4 of 10 medications or medication classes responsible for 73% of all fatal MEs in a systematic literature review.16 Many of the medication classes most commonly identified in this study also feature prominently in reviews of medication incident reports. Opioids, antibiotics, warfarin and low molecular weight heparins (both classed here as antithrombotics) were the top 4 medications or medication groups associated with fatal or severe harm outcomes in a 6‐year study of NHS medication incident data.4 Antithrombotics, penicillin antibacterials and opioids were the top 3 ATC level 3 classes in medication incidents reported by Irish hospitals in 2016.17

The largest group identified at ATC level 3 were general anaesthetics (n = 7), which included 3 cases of anaesthetic awareness. The incidence of awareness during anaesthesia is estimated at 1–2 cases in every 1000 patients.18 The feeling of paralysis while aware can cause significant psychological injury.19 PTSD was pleaded in 2 of the anaesthetic awareness cases here, while anxiety/trauma was pleaded in a third case.

Opioid analgesics (n = 6) were the next most commonly identified group with half of such claims involving an inadvertent opiate overdose. Five claims involved penicillin antibiotics, all of which resulted in an allergic reaction/anaphylaxis and all but 1 of which were cases in which the patient's allergy status was previously known. Two of these cases had a fatal outcome, once again underlining the potential for harm when penicillins are used inappropriately and the need for vigilance with regards to patients' allergy status.

Antithrombotic agents, which include anticoagulants and heparins, were cited in 5 claims, 3 of which involved excessive dosing. Local anaesthetics were also cited in 5 claims, in all cases in combination with another agent. Three of these cases involved inadequate epidural block during Caesarean section, all of which resulted in anxiety/trauma or psychological injury including PTSD. In 3 of 4 epidural block errors, the combination drug used was fentanyl (classed as an opioid anaesthetic by ATC). Failure of epidural analgesia may be related to factors other than the drug or drug combination employed, such as incorrect catheter placement as a result of the type of surgery (as this determines the region of the spinal column chosen for block placement), patient factors and operator inexperience.20 Antipsychotics featured in 4 claims, 2 of these involved the drug clozapine. Other drug groups were associated with smaller numbers of claims.

The MEs referred to in these claims all originated in either the prescribing (42%) or administration (58%) stages of the medication‐use process. A US study in 2 tertiary care centres found that, of errors which resulted in preventable ADEs, 56% occurred at the ordering stage and 34% at the administration stage, while transcription (6%) and dispensing (4%) accounted for considerably less errors.21 There were no dispensing errors recorded in the claims studied here. This may be as a result of the pharmacy profession's adoption of double checks, standardisation and/or protocols in reducing dispensing error. Wider adoption of these practices by other healthcare professions would be welcome, but would need to be resourced. It is notable that two‐thirds of these cases could have been avoided if staff had adhered to the five rights of medication administration i.e. right patient, right drug, right dose, right route and right time.22

Of 78 medication‐related finalised claims (i.e. excluding the 1 linked claim) identified in the 6‐year period, 48 (61.5%) closed with a payment to the plaintiff. This compares to 41% in a much larger study of US medical insurance data and may be accounted for by differences in litigation practices between the 2 jurisdictions.23

The cost of all medication‐related finalised claims (n = 79) regardless of the case outcome was compared to the cost of all clinical claims over the 6‐year period. Although medication‐related claims made up 3.1% of all clinical claims by number, they only accounted for 2.0% of all litigation related costs, signalling that these are, in the main, low value claims. By comparison, a US study of medical malpractice insurance data found that medication‐related closed claims accounted for 4–6% of all closed claims and 4–6% of total dollars expended.24

The median (IQR) total cost of those medication‐related claims which concluded with a payment to the plaintiff was €60 991 (€135 253), including median (IQR) damages of €33 858 (€45 388). The high IQR figures relative to the respective medians reflect the fact that the data include a small number of high value outliers. The mean total cost per claim of €151 916 is significantly higher than the €111 727 per claim (adjusted to euro 2015 figures) for 62 anaesthetic claims reported in the UK.10 This difference may be reflective of different litigation costs in the 2 jurisdictions.

The total cost of the 48 claims analysed in this study amounted to almost €7.3 million. The total cost of all clinical claims finalised over the study period, regardless of outcome, was €382.1 million. The publicly funded Health Service Executive (HSE) budget for 2016 was €14 577.5 million.25 Per capita spending on health in 2015 was $5276, placing Ireland seventh amongst Organisation for Economic Co‐operation and Development (OECD) countries.26 The cost of medication‐related claims is small by comparison to these figures but a focus on claims is likely to grossly underestimate the cost of medication‐related harm. Therefore, these claims should be seen as representative of a wider preventable burden on the Irish publicly funded healthcare system.

This study has several limitations. It is a relatively small study of 48 finalised claims; however, precise data on all 48 claims were available. As data on the numbers treated with the medications identified in this study are not available, denominator data are lacking and so error prevalence rates cannot be estimated, nor are the findings generalizable.27 Further, due to the significant time‐lag between the actual incident complained of and the settlement of the claim, practices may have altered, reducing the relevance of any findings.27 However, closed claims analysis provides a valuable source of data for studying MEs. This level of detail is not available within incident reports. Closed claims analysis, therefore, continues to have a place in providing lessons for patient safety and has been used for this purpose in other jurisdictions.11

5. CONCLUSIONS

In recent decades, extensive data on medication incidents have been published. However, there remains a dearth of robust data on medication‐related clinical claims. This study links, for the first time in an Irish context, data on medication incidents, actual harm to patients and litigation costs. Thus, it presents a comprehensive picture of the consequences of ME. The findings should assist healthcare organisations in Ireland and beyond to target error prevention strategies at specific medication groups and the more vulnerable stages of the medication‐use process, potentially improving patient safety whilst reducing litigation associated costs.

COMPETING INTERESTS

There are no competing interests to declare.

CONTRIBUTORS

M.M. codesigned the study, collected the data, performed the analysis and wrote the manuscript. D.S. co‐designed the study and reviewed and contributed to the manuscript.

APPENDIX A.

A1. Data collection form

NIMS record number:
Date of incident:
Location:
Medication involved:
Type of medication incident:
Type of injury:

NIMS severity rating:
Primary profession involved:
Case outcome:
Date claim finalised:
Paid damages (€):
Paid total (€):
Case note:
Summary of incident:

McCullagh M, Slattery D. Medication related litigation in Ireland: A 6‐year review. Br J Clin Pharmacol. 2019;85:2155–2162. 10.1111/bcp.14035

DATA AVAILABILITY STATEMENT

The study data were sourced from the NIMS database. These data are owned by the SCA in Ireland and, due to commercial and privacy considerations, are not publicly available. Requests for data should be directed to the SCA.