The requirement to respect autonomy - The Royal New Zealand ...

improving performancemany general practices—a system often describedin the literature as ‘ad hoc adjustments’. Resultsfrom the venesection were faxed from the lab tothe practice. The doctor decided if a dose adjustmentwas required and when the next test shouldbe. The practice nurse would then contact thepatient with the new instructions.As the doctor involved in the dose adjustments, Itook into considerations several factors other thanthe patient’s past INR results. These included myimpression of the patient’s ‘reliability in followinginstructions’, transport and mobility issues.In order to improve efficiency, we decided todevolve the clinical decision-making process tothe practice nurses by adopting a standardisedmethod for INR adjustment. Following an inhouseclinical education meeting, we decided toimplement the guideline produced by the BritishColumbia Health Service as this appeared wellvalidated and easily understood. 4 For the firstmonth of the new regime, all INR adjustmentsmade by the nurse following the anticoagulationguidelines were checked by the doctor and noerrors were found.The change from a method that was ‘tailor-made’by the doctor for each patient to a ‘one size fitsall’ nurse-led method had resulted in importantimprovements in efficiency. Gains in practice efficiencyare sometimes achieved at the expense ofpatient care. Our aim was to discover if any alterationhad occurred in the standard of anticoagulationcontrol or frequency of patient venesection.Measurement of problemUsing patients as their own controls, a retrospectiveaudit was performed for a period of sixmonths before (ad hoc arm) and six months afterthe change in practice (standardised arm). Patientswho commenced or discontinued warfarin duringthe study period were excluded. Only thosepatients whose target INR was 2.0–3.0 wereincluded. Only a small number of patients inthe practice had a target range higher than this.They were patients with mechanical heart valvesand, while they were managed using the sameguidelines, their numbers were too small to allowany meaningful analysis.WHAT GAP THIS FILLSWhat we already know: Most monitoring of anticoagulation occurswithin the primary care setting. Some practices use ad hoc methods ofwarfarin dose adjustment which can be time-consuming and of untestedeffectiveness.What this study adds: Switching from ad hoc adjustments to a standardisedprotocol improved our efficiency and maintained effectiveness.The dates of testing and the INR results for the12-month period were extracted from the practicemanagement software and analysed in order toanswer the following three questions:1. Has there been any change in the frequencyof testing of INRs? To establish this, thelength of time between tests was measuredfor each patient for both study periods. Amean was calculated with a standard deviationto establish whether a significant change inthe frequency of testing had occurred foreach individual and for the group as a whole.Ninety-five percent confidence intervals wereused in establishing significance.2. Has there been any change in the amountof time spent by the patients in thetarget range over the two study periods?Establishing length of time within thetherapeutic range is a complex issue. It canonly be known for sure whether someone iswithin or outside the range on a particularday when an INR test has been performed. Iftwo consecutive results are within the targetrange then an assumption is made that thepatient has been in range for the entire intervalbetween the tests. If consecutive results showone result in target and one out of target thena linear estimate is made of when the patiententered or exited the therapeutic range. Usingthese assumptions, the percentage of timewithin the therapeutic range was calculatedfor each patient for both study periods using95% confidence intervals in establishingsignificance. A mean and standard deviationwas calculated for the test group for bothstudy periods to establish whether or not asignificant change had taken place.VOLUME 2 • NUMBER 4 • DECEMBER 2010 J OURNAL OF PRIMARY HEALTH CARE 319