Quality Improvement Tools and Procedures

Statistics

It is the mathematical science of collecting,

analyzing, interpreting, and presenting data.

Data

Discrete entities are described objectively without interpretation.

Quantitative data

measurable data that focus on numbers and frequencies, such as repeat analysis data or patient waiting time.

Qualitative data

provide information about descriptive characteristics and are based on observation rather than measurable data

Population

It comprises the entire set or group of items being measured

Sample

It is the number of items actually measured from a population.

Dataset

It is the information or measurements acquired by evaluating the particular sample

Frequency

It is the number of times a particular value of a variable occurs or the number

of observations of an event.

Dependent variables

Observed in statistical studies to change in response to independent variables

and are not controlled during the study.

Dependent variables

It can also be referred to as response variables, measured variables, or

output variables.

- Response variables

- Measured variables

- Output variables

Dependent variables can also be referred to as ..., ..., or ...

Independent variables

Deliberately manipulated to change the dependent variables.

predictor variables or input variables

Independent variables may also be referred to as ... or ...

Continuous variables

Being studied that have an infinite range of possible mathematical values.

Dichotomous variable

Being studied that have only two opposing choices, such as on or off, black or

white, etc.

Central Tendency

The central position of a sample frequency

Mean

Is the most reliable of the three measures of central tendency.

Mean

The calculated average of a set of observations

µ, x ̄ , or M

Mean is denoted by ...

Median

A point in a data distribution above which exactly half of the values lie and below which the other half lie.

Median

it is the numeric middle or halfway point in the

observed dataset

Mode

The one value that occurs with the greatest frequency in the dataset.

mode

The ... provides information about the most typical occurrence, but this is usually just a rough estimate of the central value.

Reliability

The consistency of repeated measurements of the same thing, or the reproducibility or variation of a result

Precision

Reliability is sometimes known as ...

Accuracy

The ability to measure what is purported to be measured

Validity

Accuracy aka ...

Bias

A systematic or nonrandom difference between the true value of a property

and individual measurements of that property, or the presence of a systematic

error.

prejudice

Sometimes bias in a statistical context is thought of as being synonymous with the term ...

Measurement error

refers to the difference between the measured value and

the true value of the variable being measured.

Systematic errors

result from factors such as malfunctioning equipment, not correcting all outside influences, and poor design of the data collection process.

Determinate errors

Systematic errors aka ...

Random errors

are caused by statistical fluctuations or uncertainties such as quantum mottle.

Indeterminate errors

Random errors aka ...

Range

The difference between the highest and lowest values or the width of the distribution of values and is a measure of the dispersion of the data distribution.

Standard deviation

The range of variation or dispersion of a set of values surrounding the

mean, or the spread or distribution of a dataset

SD or sigma (σ)

Standard deviation can be symbolized

by the capital letters ... or the small Greek letter ...

Variance

The square of the standard deviation in Poisson statistics, and is used to determine if the separate means of several different groups differ significantly from each other (e.g., between male and female patients, different age groups).

Gaussian Distribution

it is also known as a normal distribution and creates a bell-shaped curve that is

continuous, with both tails extending to infinity.

Carl Friedrich Gauss

Gaussian Distribution was named after the German mathematician, ...

Poisson Distribution

is a discrete probability

distribution that is used to determine whether events occur randomly or not

(such as the number of particles emitted by a radioactive source in a given

period of time).

Simeon Poisson

Poisson distribution is named after a French mathematician, ...

Variation

Refers to anything that would cause a process to deviate from acceptable

standards (or the norm)

Accuracy

Validity sometimes referred to as ...

Construct validity

the extent to which a measure would agree with other survey instruments that have been used to measure the same parameters

and have proven accuracy.

Content validity

the extent to which a survey will cover all of the content area.

Criterion validity

compares the results obtained in a survey to an

established criterion measure or benchmark.

- flowcharts

- cause-and-effect diagrams

- histograms

- Pareto charts

- scatterplots

- trend charts

- control charts.

Seven basic statistical tools:

Flowchart

A pictorial representation of the individual steps that can be contained in a process.

Flowchart

It is designed to present the sequence of events in the process from its beginning point to its end point.

Terminal symbol

is a rounded rectangle or oval that identifies

the beginning or end of a process.

Activity symbol

The ... symbol in a flowchart is a rectangle.

Decision symbol

The ... symbol is a diamond.

Cause-and-Effect diagram

A causal analysis tool

Fishbone chart or Ishikawa diagram

Cause-and-Effect diagram aka ... or ...

Kaoru Ishikawa, University of Tokyo, 1943

Cause-and-Effect diagram was developed by ... of the ... in ...

Cause-and-Effect diagram

It is used to demonstrate the causes and effects of different variables or

conditions on a key quality characteristic graphically and, thereby,

potential areas for improvement.

Root Cause Analysis (RCA)

Cause-and-Effect diagram is useful in identifying multiple causes of problems during ... studies

Histogram

A data display tool in the form of a bar graph that often plots the most frequent occurrence of a quantity in the center.

Value, height

A histogram differs from a bar graph in that it is the area of the bar(s) that

denotes the ..., not the ... of the bars.

scorecard display

Histograms can also be adapted in a "..." in which the bars are color-coded

Green

is used if the goal or benchmark has been met or exceeded,

Yellow

is used if it is within a certain percentage of the goal

Red

indicates that the goal or benchmark has not been met.

Gantt charts

... was devised by Henry Gan in the 1910s and is used to illustrate a project's schedule, including start date, progress status, and

projected completion date(s).

Henry Gan, 1910s

Gantt chart was devised by ... in the ... and is used to illustrate a project's schedule, including start date, progress status, and projected completion date(s).

Pareto chart

A causal analysis tool that is named after Wilfredo Pareto, a seventeenth-

century Italian political economist.

Wilfredo Pareto

Pareto chart is a causal analysis tool that is named after ..., a seventeenth-

century Italian political economist.

80/20 rule (80%, 20%)

The Pareto chart was developed to illustrate the ... Rule (that ... of the problems stem from ... of the causes).

Pareto chart

The ... is useful in identifying the main causes (i.e., those that

occur with the greatest frequency) of problems and in demonstrating the

results of improvement strategies that have been implemented.

frequency, severity

It is important to note that Pareto charts document the ... of causes and not necessarily the ... of a cause

Scatterplot

It is a traditional two-axis graph (x-axis and y-axis), with several data

points that have been plotted throughout.

scatter diagram

Scatterplot aka ...

Scatterplot

It is designed to determine whether a relationship exists between two

different variables in a process.

positive correlation

A ... is indicated when both the x and y values increase in relation to each other.

negative correlation

A ... is demonstrated when there is an increase in the x variable in relation to a decrease in the y variable.

Trend chart

Pictorially demonstrates whether key indicators are moving up or down

over a given period of time on an ongoing basis.

Run chart or run-sequence plot

Trend chart is aka ... or ...

Trending

... refers to the evaluation of data collected over a period of time for the purpose of identifying patterns or changes.

Control chart

A modification of the trend chart in which statistically determined upper and lower control limits are placed with a central line that indicates an accepted norm.

Walter A. Shewhart, Bell Labs, 1920s

The control chart was invented by ... while working at ... in the ...

Shewhart chart or process/behavior

chart.

Control chart is sometimes referred to as ... or ...

Strategic plan

Quality improvement plan is aka ...

Quality improvement plan

It is an organizational plan that describes the goals and proposed activities

of a particular quality improvement program.

Quality improvement plan

It must incorporate current standards and benchmarks, performance appraisal, professional assessment and improvement, and specific program needs.

Quality improvement plan

PURPOSE: To provide a formal ongoing process by which an organization

utilizes objective measures to monitor and evaluate the quality of services,

both clinical and operational, provided to patients.

Performance measures

... are designed to measure systems of care and are derived from clinical or practice guidelines.

Performance measurement

... is a process by which a healthcare organization monitors important aspects of its programs, systems, and processes.

Performance management

... is a forward-looking process that is used to set goals and regularly check progress toward achieving those goals.

1. Performance measures

2. Performance measurement

3. Performance management

Quality improvement plans:

1. Examination appropriateness

2. Procedure protocol or workflow

3. Accuracy of interpretation

4. Communication of imaging results

5. Measuring and monitoring performance improvement in quality, safety, and efficiency

5 Dimensions to quality in diagnostic imaging:

Key Performance Indicators (KPIs)

The critical indicators of progress toward an intended result.

1. Financial

2. Operational

3. Quality

Key Performance Indicators (KPIs):

Financial

a measurable value that indicates how well an organization or department is doing regarding generating revenue and profits.

Operational

measures how busy the department is as well as how

efficient it is in handling the work volume.

Quality

measures the desired outcome of the department.

Dashboard

A concise, context-specific display of KPIs for quick evaluation of multiple subsystems

Brainstorming

is a group process used to develop a large collection of

ideas without regard to their merit or validity.

Focus group

is a small group that focuses on a particular problem and then hopefully derives a solution.

quality improvement team

is a group of individuals who implement the

solutions that were derived by the focus group.