Evaluating the risk of aseptic preparations
Pascal Bonnabry, Universities of Geneva and Lausanne
The preparation of injectable drugs is a high-risk process with possible negative consequences for patients in case of wrong drug/dose errors (chemical risk) or failures in the aseptic barrier (microbiological risk). Several simulation and real-life studies have helped understand and quantify the risks and imagine different re-engineering solutions that could improve the safety of the process. The improvement strategies must have the objective to reduce the frequency of errors and to increase the reliability of control steps.
The hospital pharmacy can propose innovative solutions to increase the safety of aseptic preparations. The main evolution is to industrialise the preparation of high-alert injectables, by implementing a five-step strategy:
- Standardisation of dilutions and labelling: this first step is essential both to reduce the risk during the preparation process in wards, as well as to open the door to ready-to-use standardised injectables, prepared by batches at the pharmacy.
- Production under GMP conditions: a strict respect of these rules provides guarantees of a high-quality final product. The quality management system must be applied to the structure (production facilities, equipments), the process (qualification, documentation) and the operators (training and qualification).
- Preparation by batches when possible: the safety of batch productions is higher than individualised preparations. Indeed, the production process can be standardised and qualified and the product can be quality controlled before to be released for patients. Batch production is more and more applied for parenteral nutrition (that is, standards formulations for the first days of life), cytostatics (dose banding) and other standardised dilutions of injectables (CIVAS).
- Quality control even for individualised preparations: quality control is mandatory for batches but it can also be developed to improve the safety of individual preparations, like cytostatics and parenteral nutritions. To reach this objective, a well-equipped quality control laboratory is necessary and the process should be appropriately organised to allow a sufficient period of time for analysis between the compounding and the administration.
- IT and automation: finally, different systems involving IT can help for the documentation and the traceability (computer-assisted production management), for the detection of errors in the process (gravimetric in-process control) or for avoiding the occurrence of errors (automated systems). Automated systems exist to improve the preparation safety and efficiency of parenteral nutrition, cytostatics and ready-to-use CIVAS syringes.
In conclusion, the preparation of injectables in wards is not reliable, essentially due to human factors.
Producing aseptic injectables at the pharmacy improves the chemical and the microbiological safety, by changing the paradigm, but a high-quality assurance system and automation are key factors to optimising safety.
Smart pumps and drug libraries: the way forward
Kath Phillips, Southport and Ormskirk NHS Trust
Smart pumps can prevent medication errors by alerting users to a pump setting which does not match agreed drug administration guidelines. Drug administration guidelines are entered onto a database and form the ‘drug library’. This ‘drug library’ converts a conventional pump into a smart pump.
Smart pumps have been widely available in the UK since 2005 but most hospitals do not use the available technology. The ‘Sign up to Safety’ Campaign will reward Trusts that demonstrate they are taking action to prevent medication errors by reducing their NHSLA premiums. Implementing smart pumps would play a major part in this.
The MHRA brought out a patient safety alert in March 2014 entitled ‘Improving medication incident error reporting and learning’. The smart pump’s software can log and track all alerts, recording the time, date, drug, concentration, and infusion rate, as well as whether an alert has been overwritten or the pump has been reprogrammed with different settings. This information can be used to shape current practice guidelines or highlight areas where further clinician education is needed.
Unfortunately, errors can still occur as staff could choose not to use the drug library or could over-ride ‘soft limits’ when setting up an infusion. A drug could be administered to the wrong patient or the infusion is within agreed limits but is incorrect for the patient (for example, renal failure), or the wrong drug could be chosen from the drug library. A bar-code medication administration (BCMA) would ensure the right patient gets the correct drug, dose, route at the right time.
The presentation described the experience of introducing smart pump technology to Southport and Ormskirk NHS Trust. A ‘drug library’ can consist of a number of care units which can be tailored to specific specialities within a trust. Southport’s ‘drug library’ consists of three care units: ICU; general wards; and obstetrics and gynaecology. Pumps were tested and checked to verify that they alarmed appropriately and the drug library data was correct before it was released to the wards. A number of critical lessons were learned in the process: it was more time-consuming than had been anticipated, the IT department should be involved from the outset, and more discussion was needed with nursing staff and the medical equipment library.
Currently only the general ward drug library is in use because the lack of wireless connectivity means all pumps have to pass through the medical equipment library for the drug library to be uploaded. However, 15 drug errors have been prevented, three of which involved insulin. Wireless connectivity is due to be introduced soon and will greatly improve the speed with which the drug libraries can be uploaded and updated.
Electronic reminders in type 2 diabetes patients
Marcia Vervloet, Liset van Dijk et al, Netherlands Institute for Health Services Research
Many patients experience difficulties in taking their medication. As a result, patients do not optimally benefit from their treatment. In addition, it increases healthcare costs. Poor adherence can have different causes. Patients might forget their medication, but they might also intentionally skip a dose. We studied a technological solution for diabetes patients who forget to take their medication.
With real time medication monitoring (RTMM), patients are reminded of their medication intake. This system uses an electronic medication dispenser that registers each opening. If the dispenser is not opened within an agreed time period, an SMS is automatically sent to the patient to remind them of the medication intake. Patients only receive an SMS when they forget to take a dose. We studied the effect of RTMM among 161 patients from 40 pharmacies. Patients selected for the study used oral anti-diabetics for at least a year and had collected less than 80% of their diabetes medication at the pharmacy in the year preceding the study. They were divided into three groups: two groups received the medication dispenser, one of which additionally received SMS reminders and the third group – the control group – did not receive the dispenser or reminders.
Patients who received the SMS reminders revealed higher adherence levels than the control group. Also in the long term (1.5 years after the SMS reminders had stopped) these higher levels remained. Furthermore, compared with the patients who used the medication dispenser but were not reminded, they took more doses within the agreed time period. Previous studies only investigated the short-term effects of electronic reminders on adherence. Moreover, in these studies, a reminder was sent regardless of whether the patient had taken the medication or not. This can lead to habituation and, as a consequence, loss of effectiveness. Our study is the first to show long-term effectiveness of electronic reminders on adherence with the smart medication dispenser in diabetes patients. Another finding of our study was that patients especially forget their medication when their daily routine is interrupted, for example during weekends and holidays. These moments require more attention.
A smart medication dispenser can be an effective tool for healthcare professionals to support patients who tend to forget their medication or are inaccurate with their timing. But it can also strengthen self-management – not only for patients with diabetes, but also for patients with other chronic diseases.
- Real time medication monitoring (RTMM) with SMS improves long-term adherence
- RTMM with SMS improves regularity of medication intake
- But you need to be aware of interruptions in daily routine!
- RTMM is a tool for healthcare providers to support patients in their medication use.
Improving patient safety with prefilled syringes
David Whitaker, Manchester Royal Infirmary, UK
Millions of prefilled syringes (PFS) are currently in use. Twenty-four of the 48 leading blockbuster injectables are currently available in PFS. Subcutaneous heparin and vaccines are also widely available in PFS because they are easy-to-use and there is less waste and hazard compared with multidose ampoules. These particular injectables are often used as the sole treatment at the time and so less prone to error, but all the generic low-cost IV drugs used for the polypharmacy involved in routine anaesthesia, intensive care (ICU) and pain medicine should now be provided in prefilled, prelabelled syringes to improve patient safety. Currently very few are.
In 2004, in Seattle, a 69-year-old woman, Mary McClinton, died after being injected with an unlabelled syringe containing 10ml chlorhexidine cleaning solution instead of IV contrast by mistake. In 2010, at Great Ormond Street Hospital, a young girl, Maisha Najeeb, was brain-damaged and made blind in one eye when two unlabelled syringes, one containing IV contrast and the other containing glue, were mixed up. In January 2014, this resulted in the highest ever NHS compensation pay-out (£24 million).
Studies have shown the incidence medication errors in anaesthesia to be as high as 1 in 133. The quarterly reports from the UK Safe Anaesthesia Liaison Group (SALG) always report medication errors as the group with the second highest incidence (901 April to June 2014). 50ml syringes prepared by nurses on the ICU show wide variations in concentration; for example, up to 400% variation in the case of magnesium and some contain no active drug at all. Simulation studies have shown the time to set up a noradrenaline infusion is reduced by 106 seconds when a PFS is used. In 2010, the Anaesthetic Patient Safety Foundation (apsf.org) held a consensus conference with pharmacists, clinicians and manufacturers, which made strong recommendations for standardisation of drug concentrations and equipment and the use of PFS whenever possible. There are many advantages of PFS, for example, by ensuring correct filling and correct labelling they remove two human factor errors immediately and provide a sterile product. A total of 6% of syringes prepared in theatres have been shown to be contaminated, 16% on the wards but 0% in a clean room. PFS are also instantly available, cause less distractions and fewer needlestick injuries. They improve efficiency with less preparation time and less overall storage, and have a reduced environmental impact with less packaging, less transport and less waste disposal.
The cost of the 1.2m adverse drug events with injectables in the US is $5 billion per year. Who should fill the world’s syringes in 2014: nurses and doctors drawing them up at the bedside or sterile filling machines in a clean environment?
- Very few routine generic anaesthetic, ICU or pain medicine drugs are in prefilled syringes (PFS).
- One medication error occurs for every 133 anaesthetics; 1 in 250 of these is fatal.
- High-profile unlabelled syringe mix ups have occurred; for example, Mary McClinton, Maisha Najeeb.
- With the correct drug and correct label, PFS remove two human error steps immediately.
- In 2014, nurses and doctors prone to unavoidable human error should no longer have to fill syringes.
Rational selection of drugs using matrix models
Rob Janknegt, Orbis Medisch Centre
Biosimilars have a molecular structure, which is similar, but not identical, to the reference product. For biological medicines, no generics will be introduced. There is no clinical need for biosimilars, but they may contribute to cost containment. The cost of developing a biosimilar is considerably higher than for a generic drug.
It is expected that the prices of biosimilars will be 30–35% lower than those of the reference products. It is presently unlikely that the prices of reference products will be lowered to the same extent.
Several biosimilars are already available in the EU, such as erythropoietin, filgrastim and somatropin. Two biosimilars of infliximab are already approved by the EMA and will be introduced early in 2015. Other introductions, such as rituximab, trastuzumab and etanercept are expected in the next couple of years, although none of these have yet been filed.
The are many uncertainties around biosimilars for physicians, stimulated by the innovative industry, based on the complex nature of a ‘biological’ and production process. Items such as glycolysation and immunogenicity have been discussed extensively. Another problem is that the aim of ‘similarity’ trials is not to prove ‘benefit’ but confirm ‘similarity’. The chemical and biological characterisation of a biosimilar is of greater importance than clinical trials, which is not easily accepted by physicians. Extrapolation from one indication to another without supporting clinical studies is also problematic for most physicians. In this light, it was remarkable to witness an easy switch from intravenous to subcutaneous Orencia and Herceptin. Immunogenicity and lack of trials appeared to be much less relevant in this case….
Therefore, there is a clear need for a rational selection of biosimilars, based on clinically relevant selection criteria. A European panel of hospital pharmacists has developed a set of no less than 31 selection criteria that could be applied to biosimilars. Per criterium, it was judged whether this was adequately dealt with by the EMA.1 Acquisition cost was not taken into account to allow a preselection on quality aspects only. In the final round only cost is taken into consideration.
The Final set of criteria were as follows:
- Is the manufacturer of the API and the medicinal product experienced in the production of biopharmaceuticals?
- How long has the biopharmaceutical been on the market?
- How extensive is the clinical experience with biosimilar? Expressed as the number of patient-days worldwide.
- Are there any differences in drug formulation and administration in comparison to the reference product or other biosimilars?
- What is the number of registered indications for the biopharmaceutical/biosimilar?
- Are there different results in comparison to the reference product?
- Which (serious and mild) adverse events and in which frequency were reported in clinical trials with the biopharmaceutical?
- Are there any contraindications, precautions or warnings which are different compared to the reference product?
- Is immunogenicity, as far as known, caused by a homogeneous type of antibody or is there a high intra- or inter-individual variability? Is there a difference between biosimilar products regarding drug antibody homogenicity?
- Are there differences in the incidence and severity of drug interactions?
Knockout criteria included a guaranteed supply over a long time period and adequate personnel to process adverse events reports.
A project has been initiated to make this set of criteria available in an interactive format in a (large) number of European hospitals.
Each hospital (or hospital pharmacist, or clinician) can assign his own weight to the selection criteria, thereby making a personal choice between the biosimilars (and reference product). The reference product will almost always proceed to the final (because of the better documentation). Each hospital determines whether/how many biosimilars proceed to the final, based on personal weightings and acceptable scores. In the next round, only cost is applicable (procurement). Because of the preselection on quality aspects, only high quality biosimilars will be used and much less time is needed to make the selection, because of the matrix filled by European experts. This improves quality and reduces cost.
In the next phase, a registry will be developed documenting efficacy and safety of expensive biological agents.
Hospital pharmacists who would like to participate, please contact the author:
- Boone N et al. How to select a biosimilar. Eur J Hosp Pharm 2013; doi:10.1136/ejhpharm-2013-000370.
Optimising input to medicines reconciliation
Linda Dodds, Medway School of Pharmacy, UK
Prescribing errors can occur in up to two-thirds of hospital admissions and may have a clinically important impact on up to 60% of affected patients. Medicines reconciliation (MR) at transfer of care has been demonstrated to reduce these admission prescribing errors by up to 75%, and research has indicated that pharmacists are most effective at delivering this service. However, many pharmacy services struggle to deliver MR to all admitted or transferred patients. Optimisation of the MR service is thus a key priority for those charged with delivering it. A large service evaluation and audit carried out to guide optimisation in East and South East England demonstrated that1:
- Prioritising a MR service by care area the patient is admitted to is not a safe option
- Prioritising a MR service by number of drugs prescribed is not a safe option
- Resources available with the patient, such as medicines and lists of medicines are insufficiently recognised by admitting clinicians as a valuable adjunct to medicines reconciliation
- Planned (elective) admissions could be better managed to ensure that prescribing errors at admission are reduced or eliminated
- Pharmacist time might be better spent on supporting, through education and training, admitting clinicians to undertake MR effectively at admission, with referral of complex patients only to pharmacy services.
Questions that need to be addressed by pharmacy managers include: whether the correct skill mix is being used for MR; whether the staff involved are trained to deliver optimal outcomes; whether the outcomes of MR are being acted upon in an appropriate timeframe (ideally within 24h of admission); whether MR identifies admissions due to adherence issues or those which are drug-related, and addresses them accordingly; whether ward staff and managers within the organisation are aware of the importance to patient safety of getting MR right.
Answers to these questions should help ensure services are developed which meet not only the need of the patients admitted but also use the strengths associated with pharmacy involvement to best effect
- Dodds L. Optimising pharmacy input to medicines reconciliation at admission to hospital: Lessons from a collaborative audit and service evaluation of pharmacy-led MR services in 30 acute hospitals in England. Eur J Hosp Pharm 2014;21:95–101.
Do electronic ward cabinets improve medicines management?
William (Bill) Pulman, Guy’s & St Thomas’ Foundation Trust Hospital, UK
The Omnicell system for medication and supplies is an automated electronic system that was designed by a man in the US having seen first hand the system that was in place during the time his wife was in hospital. The timescale of how long it took nurses to obtain and administer pain-relieving medication created in him a desire to challenge and change the way things were done.
It has been designed to be quite intuitive and easy to use given a short period of directive training. It is possible to store all of the different types of medication within the open, sensor and security drawers as well as the multi-adjustable shelving. Omnicell program coding allows it to interface with every type of pharmacy program available in the US, Europe and the UK.
At Guy’s and St Thomas’ Hospital we have installed the system in over 40 ward areas including ICU, NICU, A&E, elderly care and both surgical and medical wards during a six-month period from September 2009 to March 2010. We are in the process of expansion of the pharmacy service and hope by the end of March 2015 to have installed it in another 50 wards and areas over all parts of the Trust (Guy’s Hospital at London Bridge and both St Thomas’ and Evelina London Children’s Hospital at Westminster Bridge). This will include various re-built areas and some new areas currently under construction (Cancer Centre, Dermatology, and an enlarged A&E).
Our trained staff numbers are expected to be around 7000, which include nearly 30% of staff that are in the Trust for short periods due to training courses, placements and leaving the Trust for pastures new. The system has expanded the quantity and quality of reporting information available to all staff and explains removal/return habits, stock reductions, savings and helps plan changes to the stocklists enabling fine-tuning of the service required for patients.
The installation of a system requires some pre-planning, including stocklist review, user requirements, organisation of robust training, site management for installing power sockets on their own circuit (in case of mains failure) and data points (unless the in-house WiFi is suitable and can be trusted to not ‘drop out’ unexpectedly). All these aspects will create an efficient and cost-effective system providing worthwhile information and service for many years to follow.
- Think of future expansion/future proofing
- Robust training
- High quality installation team and support
- Account for all items removed/ returned to system
- If all else fails, contact us and maybe we can solve your problem.