Interface Development Pharmacist
CPPE Local Tutor
For most pharmacists the concept of risk and risk management is relatively new. We are used to emphasising accuracy and the use of quality assurance, but what is risk?
The dictionary tells us that risk is “a chance or possibility of danger, loss, injury or other adverse consequences”. A risk can also be “a person or thing causing a risk”, and “to risk” is to carry out an activity that involves a risk. All pharmacy activities carry risks. Pharmacists are not experienced in risk assessment and ensuring that risks are either eliminated or at best managed to acceptable levels. Where such risk assessments have been carried out, they have tended to be task- rather than patient-focused. Dispensing follows standard procedures to manage out the risk of error, but risks to the patient arising from the dispensed products are not assessed or managed in the same way.
Other professions carry out patient-focused risk assessments, particularly with regard to circumstances such as discharge from hospital. Pharmacists must take more responsibility for managing risks beyond prescribing, dispensing and administration, by carrying out such risk assessments. In the UK, this is now required within the National Service Framework for Older People as part of the single assessment process.(1)
The Bandolier website carries a range of articles, which evaluate both risk and benefit using the NNH (number-needed-to-harm) or NNT (number-needed-to- treat).(2) It is imperative that when discussing risk a system that enables easy comparison is used. Bandolier advises us of two rules for comparing risk:(3)
- Is the risk stated in a straightforward language that I can understand, such as 1 in 1,000? If not, why not?
- Is the risk stated per year, per month, per day or per some period of time?
The concept of a risk ladder is discussed, with degrees of risk of an incident occurring being graded in levels between 1 (high risk) and 8 (low risk). The risk of death from peptic ulcer is graded at level 5 and expressed as being between 1 in 10,000 and 1 in 99,999 per year. The risk of being struck by lightning is level 8 (between 1 in 10,000,000 and 1 in 99,999,999). Comparable statements allow us to evaluate the real possibility of an adverse event happening and to then evaluate the impact on reducing that risk after actions have been introduced. Can this approach be applied to our role as pharmacists?
The Spoonful of Sugar report states that risks associated with medicines management are a serious problem.(4) As hospitals get busier and drugs used get more potent, the incidence of medication errors is on the increase, as are deaths associated with such errors.
To reduce this trend it is necessary to look at these risks, assess the probability of their occurrence and introduce systems that move the risk to a more favourable position on the risk ladder. Eliminating risk completely is impossible, as human error is a consistent factor. We also need to evaluate the impact of an event on the patient, as no medication is risk-free even when perfectly prescribed, dispensed and administered. Effort needs to be focused on the risks with the worst outcome, hence the effort by the UK National Patient Safety Agency to reduce incorrect intrathecal administration of drugs, as error is usually fatal.(5) Medicines management carries risks at every stage, from the traditional act of dispensing through to the final outcomes for the patient.
Double-checking means that relatively few errors leave the pharmacy.(6) In many hospitals, technicians are being trained to higher levels with final check responsibilities and ward-based clinical roles. Recent research on prescribing errors in London hospitals has looked at the factors that cause errors, with workload and poor and busy environments coming high on the list.(7) Technicians are trained to focus on dispensing using rigid procedures, so they tend to dispense and check to a higher standard.(8) Pharmacists are better utilised for checking the clinical aspects of the prescription to minimise the risks from prescribing anomalies, interactions, and so on. The most common errors include wrong drugs, strengths and formulations, plus a whole set of labelling discrepancies. For many the ideal solution is to employ robots for routine dispensing and eliminate the human risk.
Less well understood is the frequency of errors in relation to the actual number of dispensing activities and the relative impact on the patients. Where on the risk ladder does dispensing come? If the level of undetected errors listed in the article by Barker(8) is correct (ie, 16–18 items per 100,000), this is still at level 4 or 5, so still halfway up the ladder!
Financial risk is an interesting area because clinical risk may be ignored to ensure best pricing, including switching to a generic or parallel import. For example, Losec (omeprazole; AstraZeneca) is in a different dispersible formulation to the new generic, which is not suitable for dispersal in water. Parallel imports and switching contracts resulted in the accumulation by one patient of six different packs of amlodipine, as shown in Figure 1. The sixth pack was being used by the patient as her own drugs for her inpatient stay. This admission was largely caused by the confusion associated with the various packs supplied. In this case the dispensing was within the contracted terms for the community pharmacist but the risk to the patient had not been considered. No error occurred, but this is still high-risk dispensing and could occur in hospitals.
The range of medicines available for use in a patient and the associated storage all carry risks as well. On admission, most patients should have a comprehensive review of their own medication. The prescribing error rate on admission for the patient’s regular medication, before new drugs are prescribed for the acute event, is as high as 1.5 errors per drug prescribed, putting prescribing at the top of the risk ladder. Concentrating pharmacist effort at admission can identify anomalies in community prescribing and misinterpretation of some standard directions, as well as eliminate this prescribing risk.
Mr A was admitted for benign prostatic hyperplasia examination. His medication on admission included lisinopril 10mg daily and captopril 25mg three times daily. He was also prescribed glyceryl trinitrate patches and oral isosorbide mononitrate. His blood pressure was low, and he had experienced dry cough and headaches. This combination had been both prescribed and dispensed without question for two years. The doctor had included all medication on the chart after admission and was reluctant to correct the prescription despite being reminded that the clinical responsibility now lay with him. An important factor in this scenario was that the notes from the previous admissions were clear and provided evidence to support the pharmacist’s intervention. Doctors’ failure to respond to pharmacists’ advice increases risk to patients if not based on sound reasoning.
Pharmacists are particularly poor at recording their interventions and advice. This puts us at risk when things go wrong because we do not have the written evidence that we gave advice or what that advice was. Why is this? Where should we record our notes? Do we go back and check that our recommendations have been acted upon? Verbal messages and post-it notes left on wards are unacceptable practice. Pharmacists must be taught how to make good records and how to communicate directly with doctors to resolve queries quickly. It is unacceptable in a court to rely on memory when all others have written records and notes.
Patients will be discharged on a range of medications, some old, some new, some long term, some short term, some regular and some for symptom relief. This is a high-risk period for patients. Much work has been done to support discharges but there is still a major lack of validated risk assessment tools for pharmaceutical risk. This will be covered in the next issue.
- Department of Health. National service framework for older people. London: DoH; 2001. Available from URL: http://www.doh.gov. uk/nsf/olderpeople/index.htm
- Available from URL: http://www.jr2.ox.ac.uk/bandolier
- Available from URL: http://www. jr2.ox.ac.uk/bandolier/band28/ b28.html
- Audit Commission. Spoonful of sugar. Medicines management in UK hospitals. London: Audit Commission; 2001.
- Available from URL: http://www. npsa.org.uk/alerts/allalerts.asp
- Cox A, Marriott J. Dealing with dispensing errors. Pharm J 2000;264:724.
- Dean B, et al. Causes of prescribing errors in hospital patients: a prospective study. Lancet2002;359:1373-8.
- Barker S. Dispensing errors: recording, analysis & human error.Pharm Manage 2003;19(1):11-6.