Article Text
Abstract
Background Paracetamol poisoning remains a major cause of morbidity and mortality. Clinical care of paracetamol poisoning depends on a range of patient variables and typically involves both medical and nursing care. An integrated care pathway (ICP) is a multidisciplinary management plan that incorporates guidelines and best practice to enhance care and documentation for a specific patient group. Paracetamol overdose is thus amenable to an ICP.
Aim To evaluate the introduction of an ICP on process of care of the paracetamol poisoned patient.
Methods A retrospective case note review of consecutive patients admitted to the Royal Infirmary of Edinburgh following a paracetamol overdose was conducted. Data were collected for a 3-month period before and after introduction of the ICP to the emergency department and toxicology inpatient unit.
Results The ICP was used in 77% of cases in the time period studied and was associated with improvements in initial documentation of patient assessment (pre-ICP vs post-ICP: 87/161 (54%) vs 101/113 (89%), p<0.0001) and appropriateness of blood sampling (146/161 (91%) vs 111/113 (98%), p=0.01), but no change in timely blood sampling (pre 124/161 (77%) vs post 93/113 (82%)). All aspects of intravenous acetylcysteine administration also significantly improved: administration of acetylcysteine if indicated (pre-ICP vs post-ICP: 57/71 (80%) vs 71/71 (100%), p<0.0001); acetylcysteine commenced in a timely fashion (33/71 (46%) vs 55/71 (77%), p=0.0002); and acetylcysteine correctly prescribed (44/58 (76%) vs 71/71 (100%), p<0.0001).
Conclusions Implementation of an ICP for paracetamol poisoning significantly improved patient management and helped to standardise inter-professional decision making in this challenging patient group. This is likely to improve patient outcome.
- Paracetamol
- overdose
- integrated care pathway
- acetylcysteine
- mental health
- overdose
- poisoning
- toxicology
Statistics from Altmetric.com
Introduction
Paracetamol overdose is the leading cause of acute liver failure in the UK1 and paracetamol is the most common product ingested in overdose,2 accounting for over 40% of all poisoning episodes presenting to hospital.3 Management of paracetamol toxicity is complex and requires careful assessment. An accurate history with respect to time of ingestion, amount ingested and identification of risk factors which may increase risk of hepatotoxicity are required before commencing treatment with the antidote acetylcysteine. Assessment may be further complicated by the presence of co-ingested drugs or alcohol, or evidence of a psychiatric disorder. Guidelines on the management of paracetamol poisoning are widely available through TOXBASE4 and the British National Formulary (BNF),5 but concern remains among health professionals, particularly over accurate risk assessment and commencement of antidotal therapy in late and staggered (repeated) paracetamol overdoses.6 This can lead to crucial delays in acetylcysteine administration that may result in hepatotoxicity or even death.7
Management is conventionally determined by the dose ingested or plasma paracetamol concentration and time from ingestion, together with presence or absence of risk factors. For patients presenting less than 8 h after ingestion, a plasma paracetamol concentration is obtained at least 4 h post-ingestion to allow the risk of liver damage to be determined using the paracetamol treatment graph.5 A paracetamol concentration plotted above the ‘normal’ or ‘high risk’ treatment lines, as appropriate, determines the need for antidotal therapy and acetylcysteine should be commenced within 8 h post-ingestion if maximum protection is to be obtained.8 For patients presenting more than 8 h post-ingestion, treatment should be commenced without delay if they are considered at risk of hepatotoxicity. This should be implemented if more than 150 mg/kg of paracetamol (75 mg/kg in the presence of risk factors) have been ingested. When available, the plasma paracetamol concentration can then guide the need for continuation of treatment. Risk assessment for those patients who present following staggered ingestions of paracetamol is particularly complex and the paracetamol concentration cannot be relied on to determine treatment. This may lead to further delays in decision making and administration of treatment.7 Furthermore, calculation of dosage regimens is complicated for both prescriber and those preparing and administrating the infusions. Acetylcysteine doses are based on the patient's body weight and require administration of three infusions. Each infusion contains a different dose and a different volume of the diluent glucose 5%, and is administered at a different rate. Perhaps, unsurprisingly, this can lead to errors in acetylcysteine prescribing and administration.7 9–11 As a result, management of this patient group can be challenging, particularly to inexperienced junior medical and nursing staff working in busy clinical environments, frequently rotating through different hospital departments. It is therefore potentially amenable to an integrated care pathway (ICP).
An ICP is a locally produced multidisciplinary management plan that incorporates guidelines based on evidence-based practice for a specific group of patients to enhance care and documentation.12 ICPs have been shown to be effective in improving care and clinical management in many surgical specialties13 14 and terminal care,15 but there seems less experience in acute medicine.16 17 The purpose of this study was to assess whether the introduction of an ICP for paracetamol poisoning would improve the management of this challenging patient group.
Methods
This study was in two parts. An audit was conducted retrospectively of patients admitted to the Royal Infirmary of Edinburgh before work began on the ICP. An ICP for the management of paracetamol poisoning was then developed by specialist nurses and clinical toxicologists over a period of 18 months. This document was based on the paracetamol poisoning entry on TOXBASE and the BNF, to ensure that it reflected best practice. A pilot study demonstrated that a single ICP for all presentations was too long, overly complex and difficult to use. Therefore, four separated time-interval ICPs were produced to reflect times from ingestion at presentation: 0–8 h, 8–24 h, >24 h, and staggered ingestions. Key components of the ICP addressed five domains: summary of overdose presentation; immediate assessment; initiation of antidote; end of antidote management; and subsequent care and discharge (box 1). An example ICP 0–8 h is presented in the supplementary file online.
Key components of the integrated care pathway for paracetamol poisoning
Summary of overdose
Date and time of ingestion
Products ingested
Amount of paracetamol ingested
Calculation mg/kg paracetamol ingested
Staff signature log
Stage 1. Immediate assessment
Treatment plan with time frames
Treatment line graph
Risk factor list
Initial blood results
Stage 2. Initiation of treatment with acetylcysteine
Dosage regimen for standard and extended treatment
Formula for calculation of volume of antidote stock solution
Volume of diluent (5% glucose)
Rates of infusion
Stage 3. End of treatment with acetylcysteine
End of treatment blood results
Guidance for discontinuing antidote
Stage 4. Subsequent management and discharge
Discharge checklist
Record of variances in clinical management
The ICP documents were introduced to the emergency department and the toxicology unit of the Royal Infirmary of Edinburgh in October 2008. Although not mandatory, medical and nursing staff were encouraged to use the ICPs when caring for paracetamol poisoned patients and advice and support were given by specialist nurses. Four weeks after implementation, a questionnaire was distributed to all grades of medical and nursing staff to review opinions and concerns. All comments were considered, and where necessary, changes were made to the ICPs before final implementation in January 2009. A second audit (post-ICP) was conducted between April and June 2009.
Case notes of all patients admitted to hospital following a paracetamol overdose (ingestion of more than the recommended daily dose of 4 g) in a 3-month period prior to and following the introduction of the ICP were examined. Data collected included patient demographics, quantity of paracetamol ingested, recorded date and time of ingestion, time of presentation, time of blood sampling, specific blood samples obtained, completion of paracetamol treatment graph and risk factor list, time of acetylcysteine initiation and accuracy of acetylcysteine prescription.
Outcome measures audited were: completeness of patient assessment documentation; appropriateness of blood sampling; timely blood sampling; acetylcysteine administered if indicated; acetylcysteine initiated in a timely fashion; and correct dosage of acetylcysteine prescribed. The standards used to audit outcome measures are presented in table 1 and were set against the current UK National Poisons Information Service as carried on TOXBASE at the times of the audits (2006 and 2009). Statistical analysis was undertaken using the two-tailed Fisher exact test where appropriate.
Results
In the pre-ICP audit period, 177 patients were admitted to hospital following a paracetamol overdose. Of these, 16 patients were excluded for the following reasons: management commenced in another hospital (n=2); the patient provided an inconsistent history (n=8); and case notes unavailable (n=6). This left a pre-ICP study group of 161 patients.
In the post-ICP audit period, 177 patients were also admitted with a paracetamol overdose. An ICP was used in 77% of cases; however, where the ICP was not initiated (n=40) the patient was excluded. Other reasons for exclusion were the following: management was commenced in another hospital (n=2); the patient provided an inconsistent history (n=10); patient non-compliance with management (n=8); time of blood sampling or initiation of acetylcysteine not documented (n=3); and case notes unavailable (n=1). Therefore, a total of 64 patients were excluded, leaving a study group of 113 patients.
Patient demographics of age and gender across the pre- and post-ICP groups were similar (table 2). Despite differences in cohort size (pre-ICP 161 vs post-ICP 113), treatment was indicated for 71 patients in both groups.
Table 3 shows the outcome measures assessed in the pre-ICP and post-ICP groups.
Initial assessment
Significant improvements were observed in completeness of assessment documentation after ICP introduction (pre-ICP vs post-ICP: 87/161 (54%) vs 101/113 (89%), p<0.0001). Appropriateness of blood sampling also improved following implementation of the ICP (pre-ICP vs post-ICP: 146/161 (91%) vs 111/113 (98%), p=0.01), but there was no change in timely blood sampling (pre 124/161 (77%) vs post 93/113 (82%)).
Treatment decisions
Based on dose ingested or paracetamol concentration, treatment was indicated in 71 patients in both pre- and post-ICP groups. However, of these, only 57 (80%) patients in the pre-ICP group actually received treatment compared to all 71 patients in the post-ICP group (p<0.0001). The reasons for non-initiation of treatment in the pre-ICP group were the following: late presentation, but waiting for blood results (n=9); risk factor present but ‘normal’ treatment line used (n=1); an undetected or low paracetamol concentration in patients who should have been treated as they had an elevated transaminase (n=2); and treatment graph misinterpretation (n=2). One patient in the pre-ICP group was inappropriately treated early with acetylcysteine before the paracetamol concentration was available. A significantly greater proportion of patients in the post-ICP group received treatment in a timely fashion (pre-ICP vs post-ICP: 33/71 (46%) vs 55/71 (77%), p=0.0002). In contrast to previous audits,18 90 min was used as a measure of timely treatment for patients presenting late (8–24 h) or with a staggered ingestion and considered at risk of hepatotoxicity. Irrespective of the time interval assessed (table 4), treatment initiation was quicker in the post-ICP group.
Administration of treatment
No prescribing errors were found following introduction of the ICP (acetylcysteine correctly prescribed pre-ICP vs post-ICP: 44/58 (76%) vs 71/71 (100%), p<0.0001). The pre-ICP errors noted were incorrect dose prescribed (n=10) and written as mg/kg of patient weight, without calculation of actual dose required (n=4).
Staff feedback
Despite initial concerns regarding increases in documentation, formal feedback by questionnaire from a random selection of medical and nursing staff was positive (91% satisfaction). Staff who had used the ICP felt that it encouraged a detailed assessment of the patient, including appropriate blood sampling and accurate identification of risk factors. Calculation and administration of acetylcysteine was also more straightforward when using the ICP.
Discussion
We have shown that the implementation of an ICP significantly improved clinical management of patients presenting with paracetamol poisoning in those patients in whom it was used. Improved documentation and blood sampling following ICP implementation allowed a more accurate and rapid patient assessment. This, in turn, led to an improvement in treatment decisions and initiation of antidotal therapy as a significantly larger proportion of patients in the post-ICP group received acetylcysteine appropriately, and in a timely fashion. Through providing a formula within the ICP to calculate the volume of antidote stock solution required, prescription errors were observed to fall to zero with further beneficial effects on patient management. As its use was not mandatory and the tool was unfamiliar to some staff, a significant number (23%) of patients were not treated under the ICP protocol after its introduction. However, over the 3-month audit period utilisation of the ICP improved from 73% to 85%, suggesting increasing acceptance of the tool.
To our knowledge this is the first study in the UK to investigate the use of an ICP for the management of paracetamol poisoning. The study was conducted in a large tertiary acute teaching hospital serving a local population of approximately 450 000. All patients presenting to the Royal Infirmary with paracetamol poisoning over a 3-month period before and after ICP introduction were audited in the study. Management was initiated in the emergency department and continued in a unit dedicated to the treatment of poisoned patients. Introducing the ICP within an emergency department and a specialist toxicology unit provided an ideal area in which to audit the effect of its introduction on patient care. It did, however mirror a process of care applicable to many UK hospitals in which management commences in an emergency department.
The ICP was designed by healthcare professionals from a range of nursing and medical backgrounds and covered all aspects of management including patient assessment to administration of treatment. The success of the ICP was assessed not only by the clinical outcome measures described but also by staff satisfaction. Involving users, especially nurses in the development and implementation process, may have led to more ready acceptance among our staff.19
Our study did have some limitations. First, the results are based on the introduction of the ICP within a specialist unit, a relatively unique care provision pattern within the UK. However, we would argue that as care was initially not ideal even in this environment, by showing benefit of use of the ICP, it is likely it will be of benefit in other clinical environments. Second, case-note audits were performed retrospectively, thus delivery of care was not directly observed, with data instead being collected from patient records. However, in both the pre- and post-ICP audits documentation was consistent and patients were excluded from the study if care could not be assessed because of an omission in the records. Third, the cohorts were unequal in size, primarily because an ICP was not initiated for 40 patients in the post-ICP group. In these 40 patients compliance to protocol could not be assessed and they were excluded from the audit. However, despite this, the number of patients requiring antidotal therapy was the same for both pre- and post-ICP cohorts. Fourth, this study did not directly compare management of patients at each time-interval after overdose. This was because some time interval groups were small and there was also a larger proportion who presented after 8 h in the pre-ICP group. To do such a comparison a much larger cohort would need to be evaluated. Finally the pre- and post-ICP audits were unavoidably separated by a long period, due to the necessary processes of ICP development. Thus the same staff cohorts were not involved in both audits. This may have had an effect on management of this patient group, as may have other aspects of the departments working practices, which we could not formally study.
Other studies have reported problems with implementation and use of ICPs.17 Success is dependent on involvement of the multidisciplinary team and a local professional person identified as the driver for the development, implementation and continued use.19 Our experience was that despite initial concerns, junior medical and nursing staff found the documents reassuring as they cared for paracetamol poisoned patients and senior medical staff reported that junior colleagues required less advice when managing this patient group. The introduction of the ICP also established a positive link between the emergency department and the toxicology unit by maintaining consistency of care and eliminating the need for duplication of documentation. Many ICP audits have attempted to demonstrate their effectiveness by reducing the length of stay in hospital. This may not however be an appropriate measurement for acute care. As poisoned patients usually require psychosocial assessment, the documented time of discharge in the case notes did not always reflect time when the patient was actually medically fit for discharge. Nevertheless, it is likely that if paracetamol poisoned patients receive prompt assessment and appropriate antidotal treatment, the length of stay should be optimised, and the likelihood of both unnecessary and extended antidotal treatment and complications of paracetamol ingestion should be reduced.
ICPs, introduced to the UK in the 1990s,20 are an approach to enhanced care for specific patient groups.13–15 Their efficacy in the more dynamic environment of acute medicine is less well tested. Self-harm patients are often seen as a clinical challenge, and this is especially the case in paracetamol overdose where targeted therapy is important, but not always easy to apply in practice. A number of deaths occur every year due to problems in correctly applying therapy guidance.7 Improving clinical care of patients with paracetamol poisoning is therefore desirable. We are aware of patients developing liver injury due to unfamiliarity of busy junior medical and nursing staff with the intricacies of therapy. An ICP has the potential to improve care in such circumstances.
In this study we chose a 90-min standard for initiation of treatment if presentation was after 8 h or there was a staggered ingestion and the patient was considered at risk of hepatotoxicity. Audits previously conducted by the College of Emergency Medicine assessed initiation of treatment at 60 min.18 In our experience a 1-h target is difficult to achieve, but the use of an ICP provides a clear audit tool, which we believe can be used to improve standards of care. Table 4 shows post-ICP time to therapy was very obviously quicker in late (8–24 h) and staggered overdose, arguably the groups most likely to benefit most from prompt therapy. Ongoing education of rotating medical staff and triage nurses in utilisation of the ICP should ensure more patients are treated to protocol and therefore more likely to result in prompter treatment.
In conclusion, our findings show that the use of an ICP is effective in improving clinical care for paracetamol overdose patients. Clinical management improved, the ICP is easy to use, and it empowers junior medical and nursing staff in an environment in which junior staff frequently rotate. We believe this tool also facilitates clinical audit. We cannot determine benefit in survival or hepatic injury in such a small study, but it seems likely that improving care standards will improve actual outcomes if this strategy is widely applied.
Acknowledgments
We wish to thank colleagues in the medical and nursing teams of the Emergency Department and Combined Assessment Area of the Royal Infirmary of Edinburgh; in particular Alan Fisher (Clinical Effectiveness Team, Royal Infirmary of Edinburgh) for support in developing the ICP, and consultant clinical toxicology colleagues Dr Stephen Waring and Professor David Webb.
References
Footnotes
Competing interests None.
Provenance and peer review Not commissioned; externally peer reviewed.
Linked Articles
- Primary survey