NPC Archive Item: Patient Safety First – The How-to Guide for Measurement for Improvement

NOTE – This is an archive post from the NPC and has not been updated since first publication. Therefore, some hyperlinks may no longer be working.

NPC Logo

Published in March 2009 by Patient Safety First, The How-to Guide for Measurement for Improvement stresses the importance of demonstrating whether changes lead to genuine improvement with particular relevance to safety.  In achieving this there need only be a few specific measures linked to clearly stated objectives to show that change is moving in the right direction.

The Guide uses ‘The Model for Improvement’ as developed by Associates for Process Improvement (USA, available at  The model is based on three key questions used in conjunction with small scale testing, using ‘Plan, Do, Study Act’ (PDSA) cycles as the process for testing change ideas.  The three questions are:

  • What are we trying to achieve?
  • How will we know that a change is an improvement?
  • What changes can we make that will result in an improvement?

An overview of the parts of the model can be found in an accompanying document: The Quick Guide to Implementing Improvement.

The National Prescribing Centre has used the model for improvement extensively within the medicines management improvement programmes it has hosted. Learning from these programmes together with recommended improvement methodology can be found on NPC. The floors found within the Improvement Skills & Tools section will be of most relevance.

The Guide emphasises the importance of testing changes and measures on a small scale as this is less disruptive for patients and staff.  Of course, involving people in developing and testing change ideas often results in less resistance to change.

The three main reasons for measuring are research, judgement and improvement.  The Guide suggests that clinical professionals are more familiar and comfortable with measurement for research on a large scale with a fixed hypothesis to reduce variation.  Health Service managers however, may be more familiar with measurement for judgement as a way of understanding performance whereas measurement for improvement seeks to answer the question ‘‘how do we make it work here?’’.

The Guide defines different types of measures as outcome measures, process measures and balancing measures.  Medicines management examples of these measures could include:

  • Outcome measures (reflecting impact on patients) – e.g.  the number of preventable adverse drug events
  • Process measures (reflecting the way systems and processes work) -e.g. the number of planned hospital admissions that are accompanied by a minimum dataset of information from a patient’s GP
  • Balancing measures (indicating  positive or negative impact resulting from a change made elsewhere in the system); e.g., introducing medicines use reviews (MURs) targeted to COPD patients may lead to a  reduction in requests for GP appointments due to acute exacerbation

Measurements can be expressed using counts or tally charts where the number is not going to change very much;  ratios or rates are useful if the population numbers are variable; and percentages can be used where the focus needs to be more specific. However, a constant theme running through the guide is that you need ‘to seek usefulness not perfection’.

What does this mean to medicines management?
In order to relate this to medicines management, we can consider the seven steps to take when measuring for improvement:

  • Step 1 – Decide your aim
  • Step 2 – Choose your measures
  • Step 3 – Confirm how to collect your data
  • Step 4 – Collect your baseline data
  • Step 5 – Analyse and present your data
  • Step 6 – Meet to decide what it is telling you
  • Step 7 – Repeat steps 4 to 6 each month or more frequently

Steps 4 – 6 consist of three stages referred to as the CAR cycle – Collect, Analyse Review.  Do not let measure issues delay the start of your PDSA cycles as the first few should be so small they will not affect your baseline.

When identifying a topic for measurement, for example the percentage of medicines reconciled on a medical admission unit, it can be useful to use run charts to plot the data over a period of time identifying where interventions have occurred and adding a goal line (target) to see at a glance where the work is in relation to achieving the aim.

When considering the implication of the measurement process for medicines management teams it is important to define where the responsibilities lie.  This includes who will collect the data, who will analyse and present it and who will make decisions about results and responsibility for taking action.  Support with data sharing is available in The How to Guide for Leadership for Safety.

Further information on measuring medicines management improvement can be found on the NPC section  ‘Setting objectives and measuring for improvement’

Please comment on this blog in the NPC discussion rooms, or using our feedback form.