Creating a Measurement Plan

• Creating a measurement plan
• Mini-surveys
• Third next available appointment
• Defect check sheets
• Patient flow sheets
• Sampling plan

Every QI team needs to be able to measure quality throughout the project in order to monitor progress towards a target.  To do so, your team needs to generate a measurement plan with the following elements:

Selection of Measures

• Core measures:  outcome indicators, process and balancing
• Cause of poor quality measures

Data Collection Plan

• Data source
• Sampling plan
  

Selection of Measures

Your QI project should aim to track a mix of outcome, process and balancing measures, throughout the course of the project.  

Outcome Indicators

Outcome measures are the “voice of the patient”.  They represent the bottom line to the patient:  am I happier, feeling better, or likelier to live longer as a result of the care being provided?  Examples include the following:
                                                           

Dimension of Quality	Outcome
Patient-centredness	Satisfaction rate – overall Satisfaction rate – with navigation through system, getting information, available options for care, etc.
Effectiveness	Clinical measures that should be within a certain threshold (blood pressure <130/80, LDL cholesterol < 2.5, HbA1C < 7.0, Body Mass Index (BMI) < 25, Peak Expiratory Flow (PEF)>80% best score, etc.) Rate of adoption of healthy behaviours (e.g. stop smoking, exercise) General quality of life (e.g. SF-6 or SF-36 score, WOMAC score for arthritis, pain score, depression score, asthma symptom index)       Mortality rate Complication rate
Safety	Mortality rate Complication rate (e.g. surgical site infection) Adverse event rate (e.g. drug errors)
Access	Wait times (at median or 90th percentile) 3rd next available appointment time

Process Measures

Process measures check whether or not some process or activity which has been shown to have a positive impact on outcomes is actually being done.  This is also known as the “voice of the system”.  Health care providers are particularly interested in processes because they answer the question: “am I doing all the things I’m supposed to be doing to improve health for my patients?”  Examples include the following:

                                                           

Dimension of Quality	Outcome
Patient-centredness	Satisfaction rate – overall Satisfaction rate – with navigation through system, getting information, available options for care, etc.
Effectiveness	Clinical measures that should be within a certain threshold (blood pressure <130/80, LDL cholesterol < 2.5, HbA1C < 7.0, Body Mass Index (BMI) < 25, Peak Expiratory Flow (PEF)>80% best score, etc.)      Rate of adoption of healthy behaviours (e.g. stop smoking, exercise) General quality of life (e.g. SF-6 or SF-36 score, WOMAC score for arthritis, pain score, depression score, asthma symptom index) Mortality rate Complication rate
Safety	Mortality rate Complication rate (e.g. surgical site infection) Adverse event rate (e.g. drug errors)
Access	Wait times (at median or 90th percentile) 3rd next available appointment time

Balancing Measures

These measure unintended or undesirable consequences of a QI initiative.  Here are some general principles to help you brainstorm:

• if you are targeting improvement of quality for one group of patients served, make sure that quality doesn’t suffer for other patients
• if you trying to reduce waste / inefficiency, make sure the patient experience doesn’t suffer, or that costs in another part of the system haven’t increased
• if you are trying to improve patient experience, check for changes in cost
• if you are improving access, check for any increase in inappropriate use (decreased effectiveness)
• if you are reducing a bottleneck in one part of the health care system, check to see if you are inadvertently worsening a bottleneck downstream in the system
  

Causes of Poor Quality

This is not a quality indicator per se.  However, your QI teams need to know what are the main causes of why poor quality is occurring and focus their improvement efforts there.  QI teams should go through the process of brainstorming about causes of poor quality, using tools such as the Ishikawa (Cause and Effect diagrams).  If the team is certain about which are the most important causes to be addressed, then data may not be necessary.  However, if there is major disagreement or uncertainty within the team, then this information should be collected, using tools such as a defect check sheet. 

Data Collection Plan

The ideal situation is to obtain your core measures using pre-existing automated data sources.  These may include the IT system of a hospital or the electronic medical record of a clinic.  If not available, then consider either adapting your IT system to collect the new data, create a new automated system, or use manual data collection. 

For manual collection, keep the amount of data collected brief and focused. 

Here are examples of data collection instruments:

Data tool	Useful for measuring:
Defect check sheet	Safety, causes of poor quality
Mini-surveys	Any dimension of quality
Patient flow sheets	Effectiveness
Third next available appointment worksheet	Access

Gathering Baseline Data

Collect some baseline data about quality, if possible.  As a rule of thumb, try to have measurements of quality at ten points in time prior to the start of testing any improvements.  For example, if you plan on tracking quality every month, then look at data for the preceding ten months.  This step helps you identify when you’ve made a significant change in improvement later.  However, if it is impossible or impractical to get previous data, then don’t delay the start of the QI project just to get baseline data.


Measurement Plan Template

QI Project:  

Time Frame for project:

What are your Core Measures?
         

Core Measures:		Data Source
Outcome
Process
Balancing

How do your outcome and process measures link to your organization`s corporate dashboard?


Will you measure causes of poor quality?  yes/no
    If so, for which measures of quality? 

Will you collect baseline data?   yes/no
    If yes, time frame for baseline:

How often will you report on each measure of quality?
Will you use sampling for any manual data collection methods?   yes/no

If yes, for each data collection method, indicate sampling method (block or systematic) and protocol. 

To download a copy of this template, click here.