Has that dose been given?
An audit of missing administration signatures on medication charts in a psychiatric hospital
Sam Carter, Pharmacy Student, Aston University, Caroline Cahill, Senior Clinical Pharmacist, St Andrew’s Healthcare and Camilla Haw, Consultant Psychiatrist, St Andrew’s Healthcare
In studies of medication errors in general hospitals, omission of a dose of medication without a valid reason is one of the most common types of medication error (Barker, et al 2002, Tissot, et al, 2003). The same holds true for medication errors in psychiatric units (Grasso, et al, 2003; Haw, et al 2007). When reviewing medication charts retrospectively, it is impossible to know if an empty administration signature box indicates that the dose has not been missed or if it simply means the administering nurse has forgotten to sign the chart. This ‘missing signature’ type of error, where a scheduled dose is not documented as administered, is potentially serious when it involves medication given for physical illnesses, such as insulin or cardiac drugs, and psychiatric medication, such as depot antipsychotics, since the patient is at risk of either not receiving the medication at all or could receive a double dose. The National Patient Safety Agency (NPSA) has recently issued a Rapid Response Report highlighting the harm from omitted and delayed medicines to hospital inpatients (NPSA, 2010).
There appears to be little published research about ‘missing signatures’, their frequency and the possible reasons for their occurrence. We therefore set out to survey the nature, timing and frequency of ‘missing signatures’ on the medication charts at our large psychiatric hospital. We also inquired of nursing staff their opinion as to the likely factors leading to ‘missing signatures’.
St Andrew’s Hospital, Northampton, is a 600-bedded tertiary referral centre and charitable hospital taking patients of all ages with a wide variety of mental health needs. Almost all patients are funded by the NHS and the majority are detained under the Mental Health Act, 1983. Prescriptions are written on a standard hospital paper medication chart and there are corresponding boxes for the administering nurse to sign after each authorised dose has been given to the patient.
Data collection and analysis
In May 2009, two patients’ medication charts from each of the hospital’s 40 wards were photocopied. Included charts had a minimum of five regular prescription items each and where possible fourteen days worth of administered medication. Regular prescriptions only were studied (as required or prn medication was excluded). For each chart, the total number of doses that should have been given in the last fourteen days was calculated, as was the number of scheduled doses not signed for as administered. The types of drugs given, the route of administration and the time of day they were scheduled to be administered were also studied, both for signed for doses and for those that had not been signed for. Percentages of missing signatures were calculated for type and route of drug and for the time of day the drug was due to be administered. The Chi-square test was used to make comparisons between groups. We asked nursing staff on several wards for possible explanations as to why missing signatures arise.
Seventy-nine patients’ charts were studied. A total of 381 (3.2%) scheduled doses had not been signed for as administered out of the 12,052 doses studied. Regarding the time of day the drug was scheduled to be administered, of the four regular drug rounds the teatime one had the highest proportion (5.0%) of missing signatures and breakfast the lowest (2.2%) (see Table 1). Topical preparations had by far the highest proportion of missing signatures (16.0%) (see Table 2). Among orally administered medicines, drugs for physical illnesses had a significantly higher proportion of missing signatures than those for psychiatric illness (66/4854, 1.4% vs 39/5056, 0.8%, χ2 = 8.18, p<0.005). The percentage of antibiotic doses with missing signatures was 2.3%. In addition, there were five doses of insulin for which signatures were missing, representing 2.6% of all insulin doses.
‘Missing signature’ doses by time of medication round
Time of medication round
|Total no. of doses that should have been signed for||‘Missing signature’ doses|
‘Missing signature’ doses by type of drug and drug route
|Type and route of drug||Total no. of doses that should have been signed for||‘Missing signature’ doses|
Oral physical health medicines
Nursing staff considered anything that disrupted the routine of medicines administration to be likely to lead to missing signatures. Distraction, such as noise and interruptions during medicines administration was thought to be a major factor leading to failure to sign the medication chart after administering a drug. Sometimes patients will refuse medication but nursing staff will then decide to approach the patient later to see if they have changed their mind but in the event staff are distracted by something else and the omission is forgotten about. Topical preparations are frequently given to patients to self-administer in private. The patient later returns the preparation, often to another nurse and hence the drug may not be signed for as having been given. Sometimes topical preparations are only required for a short period. The patient ceases to require them and the prescriber has not been notified to cancel the prescription.
Missing signatures were present for 3.2% of doses. Although this represents a small proportion of administered drugs, sometimes the missing signatures involved key drugs such as insulin where omission of a dose or equally receiving a duplicate dose could have serious consequences. Topical preparations were by far the most likely drugs to have missing signatures at 16.0% of doses. Psychiatric drugs had a lower missing signature rate than those drugs for physical ailments, a not surprising finding for a psychiatric hospital where greater emphasis might be expected on drugs to treat mental illness.
This study was a retrospective review of medication charts. It was thus not possible to determine if the drug had actually been administered but not signed for or if the dose had simply not been administered and this is a study limitation of this type of study design. In an observational study of medicines administration carried out at the authors’ hospital, 27% of administration errors were caused by staff omitting a drug without a valid clinical reason and 24% of errors were due to staff not signing for an administered medication (Haw, et al, 2007). The error rate was much higher than in the current study – omission without a valid clinical reason occurred for 7% of all doses and failing to sign the chart for 6% of doses. Direct observation of drug administration is known to be a more sensitive method for detecting medication administration errors than retrospective chart review (Haw, et al, 2007). This audit was conducted in a large independent sector psychiatric hospital and thus the findings cannot be generalised to NHS units. However, the audit could readily be repeated in other units and the findings compared to ours.
Nursing staff accounted for the missing signatures by anything that disrupted the orderliness of the drug round and concentration of nursing staff administering medication. Thus, in order to minimise the number of doses with missing signatures, nursing staff need to administer medication in a quiet and methodical way, avoiding if possible distractions and interruptions. Regular audits by pharmacists of missing signatures highlight the problem and encourage nursing staff to reduce such errors by being vigilant and methodical when administering medication. In the future, electronic prescribing, with electronic medicines administration should reduce the problem of missing signature errors, since such systems do not permit blank spaces to be left in the medicines administration record (Sakowski, et al, 2005).
Barker, KN, Flynn, EA, Pepper, GA, Bates, DW, Mikeal, RL. (2002) Medication errors observed in 36 health care facilities. Archives of Internal Medicine.162,1897-1903.
Grasso, BC, Genest, R, Jordan, CW, Bates, DW. (2003) Use of chart and record reviews to detect medication errors in a state psychiatric hospital. Psychiatric Services. 54, 677-681.
Haw, CM, Stubbs, J, Dickens, G. (2007) An observational study of medication administration errors in old age psychiatric inpatients. International Journal of Quality and Safety in Healthcare.19, 210-219.
National Patient Safety Agency Rapid Response Report. Reducing harm from omitted and delayed medicines in hospital. NPSA/2010/RRR009. Available at: http://www.nrls.npsa.nhs.uk.resources/?EntryId45=66720.
Sakowski, J, Leonard, T, Colburn, S, Michaelsen, B, Schiro, T, Schneider, J, Newman, JM (2005) Using a bar-coded medication administration system to prevent medication errors. American Journal of Health-System Pharmacy. 62, 2619-2625.
Tissot, E, Cornette, C, Limat, S, Mourand, JL, Becker, M, Etievent, JP, Dupond, JL, Jacquet, M, Woronoff-Lemsi, MC. (2003) Observational study of potential risk factors of medication administration errors. Pharmacy World Science. 25, 264-268.
Copyright Priory Lodge Education Ltd 2010
First Published April 2010