Objective: To compare the accuracy of prescribing analgesic and anti-emetic drugs to children when using either the 2006 BNF for children (BNFC) or a new device, the Paediatric Analgesia Wheel.
Study design: A simulated prescription chart was created requiring doctors to prescribe seven drugs to each of two children (14 prescriptions in total). 52 doctors from anaesthesia, paediatrics and emergency medicine were recruited and randomly assigned to complete the chart using either the BNFC or a new device, the Paediatric Analgesia Wheel. The mean number of correct answers and mean time taken to complete the chart were compared.
Results: Prescribing using the Paediatric Analgesia Wheel resulted in a significantly greater number of correct answers (mean 10.9) compared with using the BNFC (mean 3.5). In addition, the mean time taken to complete the prescription chart was significantly shorter when using the Paediatric Analgesia Wheel (5.8 min) compared with the BNFC (12.4 min).
Conclusions: The Paediatric Analgesia Wheel provides a time-efficient method of prescribing commonly used analgesic and anti-emetic drugs to children and results in improved accuracy when compared with using the BNFC.
Statistics from Altmetric.com
Prescribing medicines to children can be challenging, especially for those who do not routinely work within children’s services. The lack of paediatric formulations frequently necessitates the use of adult products and requires complex calculations and accurate measurement of small liquid volumes.1 2 Evidence suggests that medication safety needs to be improved, particularly in babies and young children.1 In the hospital setting, medication errors in children occur at similar rates to adults but have three times the potential to cause harm.3
It was estimated that in 1997 medical errors ranked as the fifth leading cause of death in the United States, accounting for up to 98 000 deaths per year.4 In the United Kingdom, almost 60 000 medication incidents were reported to the National Patient Safety Agency between January 2005 and June 2006. Over 80% of these errors originated in hospitals, with children aged under 4 years more likely to be involved in a reported incident compared with their hospital activity. It has been estimated that medication related harm costs the National Health Service more than £750 million each year in England alone.5
The continued need to improve the management of acute pain in children with the effective and safe use of medicines has been highlighted.1 The Health Care Commission uses analgesia provided for children with elbow or wrist fractures as an indicator of the quality of clinical care provided by emergency departments.6 Although the BNFC has provided the medical and nursing establishments with a definitive formulary to guide prescribing, there is still a need to ensure the calculated dose can be accurately administered using available formulations and syringes.
Analysis of local practice revealed that the barrier to accurate prescribing was, at the point of actually writing the prescription, failure to consult the available information or to correctly apply it. Little consideration was given to drug formulations and the ability of nursing staff to administer an accurate dose. The Paediatric Analgesia Wheel (the “Wheel”) provides pre-calculated doses of commonly used analgesic and anti-emetic drugs rounded to a volume that can be accurately administered. It is proposed that use of the Wheel may improve prescribing accuracy when compared with using the BNFC.
Practicing hospital doctors from Darlington Memorial Hospital were recruited during November and December 2007. Volunteers underwent simple randomisation to complete a simulated paediatric prescription chart using either the BNFC or the Wheel. Specialist trainees in year 3 and above, staff and associate specialists (SAS) and consultants from anaesthesia, paediatrics and emergency medicine were recruited.
Both groups were required to read the prepared information and instruction sheets that described a normally fit and well child presenting in acute pain and without contra-indication to receiving the drugs to be prescribed. The method of completing the prescription chart was then explained. Group 1 had access to the 2006 BNFC and a calculator and were instructed to prescribe within the drugs’ marketing authorisation. Group 2 had access to the Wheel only. All participants completed the prescription chart in a quiet area free from distractions.
Seven drugs were to be prescribed to each of two children, a 10 kg 1-year-old and a 23 kg 7-year-old. The prescriptions were for three “as required” (intravenous ondansetron, intravenous morphine and oral oramorph) and four regularly administered (oral and intravenous paracetamol, rectal diclofenac and oral ibuprofen) drugs. In addition to dose, “as required” drugs required a minimum dosage interval (MDI) and maximum dosage frequency (MDF) to be prescribed. The regularly administered drugs required administration times to be entered and from these, MDI and MDF were extrapolated. For drugs with a dose range, participants were instructed to prescribe the licensed maximum. Each doctor was required to make a total of 14 prescriptions and the time taken to complete the exercise was recorded. For those who completed the prescription chart using the Wheel, there was an area for comments plus three statements:
I think the Paediatric Analgesia Wheel is a good idea.
If available, I would carry the Paediatric Analgesia Wheel.
I felt more comfortable using the Paediatric Analgesia Wheel than when prescribing these drugs as part of my normal day-to-day practice.
The participants were asked to what extent they agreed with these statements using the following scale: strongly disagree, disagree, no strong feelings, agree or strongly agree.
The study protocol was approved by the local research and ethics committee and written informed consent was obtained from volunteers.
The primary endpoint was the number of correct prescriptions and the time taken to complete the exercise. A correct prescription was accurate for all three prescription elements (dose, MDI and MDF) when compared with the 2006 BNFC. In the absence of an agreed standard for prescribing accuracy, a dose to within ±5% of the exact dose was considered acceptable. The exception was rectal diclofenac where the limited range of suppositories could result in an unavoidable larger deviation from the calculated dose. The researcher responsible for scoring and data entry was blinded to the method of completion.
Incorrect prescriptions were classified as an under-dose, overdose, incomplete or an unlicensed use of the drug. Under-doses and overdoses could result from an error in the absolute dose or an error in MDI/MDF that would ultimately result in a low or high total daily dose. An incomplete prescription was one that lacked an essential MDI/MDF with the potential for error depending on interpretation.
The mean number of correct answers and the mean time taken to complete the exercise were compared between the two groups. The null hypothesis was that there was no difference in prescribing accuracy or time taken to complete the prescription charts. An initial pilot study predicted at least 16 individuals were required in each group to have a 99% and 90% chance of achieving a significant difference in mean values from an unpaired, two-tailed t test with p = 0.05 for prescribing accuracy and time to complete, respectively.
Fifty two doctors were recruited; all completed the study with 25 in group 1 (BNFC) and 27 in group 2 (Wheel). Eleven (44%) participants in group 1 were consultants or SAS doctors compared with 13 (48%) in group 2. The number of correct prescriptions is shown for each drug (table 1) and for the number of participants (fig 1). Table 2 shows a summary of primary endpoint results with analysis. The mean number of correct prescriptions was significantly greater in group 2 when using the Wheel compared with group 1. The time taken to complete the prescription chart was significantly shorter in group 2 compared with group 1.
Table 3 summarises additional data analysis. When considering both weight groups together, errors in dose and MDI/MDF were significantly more likely to occur in group 1 compared with group 2. Dose related errors occurred in 62% of 350 prescriptions in group 1 compared with 4% of 378 prescriptions in group 2. MDI/MDF related errors occurred in 54% of 350 prescriptions in group 1 compared with 19% of 378 prescriptions in group 2. Errors in dose and MDI/MDF were equally likely to occur in group 1, whereas errors in MDI/MDF were significantly more likely to occur than dose errors in group 2. In addition, group 1 made a significantly greater number of correct prescriptions in the 10 kg group compared with the 23 kg group. Trainees made a significantly greater number of correct prescriptions than consultants and SAS doctors when using the Wheel. No significant difference in the number of correct prescriptions was found between grades when using the BNFC.
Twenty two out of 27 participants in group 2 completed the final three questions and 13 provided comments. All 22 first-time users of the Wheel thought it a good idea, 21/22 stated they would carry one if available and 18/22 felt more comfortable using it when prescribing.
There is continuing evidence that children’s pain may be inadequately treated in hospital and after discharge.7–9 Analgesics are among the most commonly used drugs that are prescribed poorly with some hospital staff reluctant to prescribe at all or administering a dose that is too small to address the child’s pain adequately.7–10 Under-treatment of pain in children is unacceptable, has undesirable long-term effects11 12 and exposes the child to the risk of an adverse drug reaction without the benefit of receiving an effective dose.13 The drug prescribing and administration process is complex and there are many points at which errors may occur.3 14 15 Even under optimal conditions, substantial numbers of trainee doctors make unforced errors when prescribing for children.16 Dependence on the BNFC in this study resulted in a low number of correct answers with two participants failing to prescribe a single drug correctly and high numbers failing to register that both intravenous ondansetron and rectal diclofenac were unlicensed for a 1-year-old child.
The Wheel was developed to simplify the prescribing process and, by considering in advance the feasibility of administering a calculated drug volume, facilitate accurate drug delivery. It is pocket sized and double-sided (fig 2) with an inner rotating wheel presenting pre-calculated doses and volumes of commonly used analgesic and anti-emetic drugs for children from birth to 16 years of age (3–58 kg).17 To ensure consistency with the BNFC, the Wheel advises the use of weight to align the Wheel unless a child is overweight when age, equating to ideal body weight, should be used.18 All prescribing information is consistent with the 2006 BNFC and when a range of doses exists, the maximum dose has been clearly quoted to reduce the tendency for minimum dose prescribing and administration.2 10 With an absence of an agreed standard defining prescribing accuracy, we attempted to keep all doses to within ±5% of the exact calculated dose. All recommendations are within the drugs’ marketing authorisation to avoid the increased risk of administration errors and adverse drug reactions associated with off-label and unlicensed prescribing.2 19 20 Additional prescribing information such as MDI/MDF is printed on the outer sleeve. The result is accurate and safe information that aims to encourage doctors and nurses to prescribe with confidence.
In this study, the use of these pre-calculated doses and volumes resulted in a significantly higher number of correct prescriptions when using the Wheel compared with the BNFC. Use of the Wheel by trainees resulted in a significantly greater number of correct answers compared with consultants and SAS doctors. It is proposed that trainees analysed and applied the presented information rather than relying on their own prescribing knowledge. The Wheel also reduced the time to complete the simulated prescription chart and resulted in a high level of first-time user satisfaction.
Prescribing errors are often dose related3 14 15 and are most likely to originate at the stage of physician ordering.7 14 Prescriptions leading to the administration of an excessive dose occur even for commonly prescribed drugs such as paracetamol. Despite well documented potential for dose-dependant toxicity when used with therapeutic intent,21–23 prescribing errors continue to be reported with physicians either directly over-prescribing or advising an excessive dose to be taken.24–26 In this study, use of the Wheel led to a significant decrease in the number of dose related errors. When current prescribing information is analysed, it is easy to appreciate how confusion may occur. For example, a child aged between 1 and 5 years old (10–18 kg) may legitimately be prescribed 120–250 mg of paracetamol four-times-a-day. Depending on interpretation, this equates to a daily dose of between 27 and 100 mg/kg. In addition, when following BNFC advice on age-range prescribing, it is not possible to prescribe an oral dose of paracetamol to a child between their 5th and 6th birthdays.18
The need for dose calculation increases the risk of error1 27 28 and group 1 demonstrated a significantly greater number of correct prescriptions in the 10 kg group compared with the 23 kg group when using the BNFC. It is suggested this reflects the more complex calculation required in a child weighing 23 kg and avoidance of the calculation step may improve prescribing accuracy. Administration errors in the presence of a correct prescription may also occur14 15 27 and it has been shown that junior doctors often do not consider the feasibility of administering the final volume when performing drug calculations.29 The Wheel’s pre-calculated doses and realistically administrable volumes ensure calculation and administration errors are minimised.
What is already known on this topic
Prescribing analgesics to children is often associated with errors leading to under-treatment of pain or overdose associated with side effects or even death.
Often little consideration is given to the feasibility of administering a drug volume that results from an exact dose calculation.
What this study adds
The accuracy of prescribing paediatric drugs can be improved by reducing steps in the calculation-to-administration pathway and providing pre-calculated doses and volumes to administer.
By considering the final volume first, the prescribed dose administered using available formulations and syringes can be very close to the exact calculated dose.
Errors may be associated with incorrect dose interval and frequency5 14 and “as required” administration based on MDI without qualification by MDF may result in the total daily dose being exceeded.24 Overall, 54% of prescriptions made by group 1 were associated with errors in MDI/MDF with these errors equally likely to occur as errors in dose when using the BNFC. In group 2, MDI/MDF associated errors occurred significantly more frequently than errors in dose and accounted for the majority of errors occurring with the use of the Wheel. Information related to MDI/MDF is printed on the outer sleeve and it is proposed that prescribers may simply have overlooked this information when completing the prescription chart.
Administration of drugs by the incorrect route has been reported14 15 and can lead to death.30 It has been recommended that enterally administered drugs should always be administered using oral syringes of appropriate sizes when the volume is not a multiple of 5 ml.1 18 27 30 The Wheel was developed with consideration of the administration volume with clear instruction regarding syringe choice. Volumes less than 10 ml have been calculated to one decimal place and can be administered using a single oral or intravenous syringe, whichever is appropriate to the prescribed route. Volumes greater than 10 ml are quoted without decimal places. Trailing zeros and the unnecessary use of decimal points have been avoided to minimise the risk of tenfold errors.14 27 28 Doses for discharge analgesia have been rounded to ensure that administration is possible using a prescribed 5 ml oral syringe with 0.5 ml divisions.18
The Wheel’s main limitation is a result of its pocket-sized design and therefore limited space to provide comprehensive drug information. However, the Wheel was never intended to replace the BNFC or circumvent the need for prescribing education and training. Users are referred to the BNFC for drug cautions, contra-indications, side effects, interactions and regimen adjustment for the child with co-morbidity. The second limitation relates to the difficultly of any prescribing aide or formulary to cater for every age/weight combination (eg, children at a high centile for both height and weight but with normal body mass index). Although clear alignment instructions are given for children whose weight deviates from the ideal, the performance of the Wheel under all potential circumstances has yet to be evaluated and may be the subject of future research and development.
In conclusion, we have demonstrated that the new Paediatric Analgesia Wheel provides a time-efficient, safe and effective method of prescribing commonly used analgesic and anti-emetic drugs to children when compared with the 2006 BNFC.
Funding: The Paediatric Analgesia Wheel’s design, future publication and distribution are funded as a service to medicine through a grant from Bristol-Myers Squibb. The authors and their associated research have not received research funding or gratuity in any form. The Paediatric Analgesia Wheel is a non-commercial development with the final version planned for free distribution within the United Kingdom.
Competing interests: None.
Ethics approval: The study protocol was approved by the local research and ethics committee.
If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.