Accident Data Analysis

1
Accident Data Analysis

• Many times, safety professionals conduct data
  analyses involving accident data as a means to
  identify prevention activities
• Common examples of performance indicators
  include OSHA rates and workers compensation data
• While these measures can be useful, they can also
  create some problems.
• Small data sample biases
• Severity measure biases

2
Safety Performance Indicators

• Performance indicators are activities that
  employees and management can use to continually
  guide their improvement.
• Performance indicators address problem areas,
  provide some measure of accountability for
  employees, and corroborate the fact that the
  organizations goals are (or are not) being met.

3
Categories of Performance Indicators

• Three broad classes of these measures are
  trailing indicators, current indicators and
  leading indicators.
• A good combination of trailing, current and
  leading indicators, based on the company's own
  needs, are always best.
• Performance indicators can be further classified
  as activity based or outcomes based

4
Trailing Indicators

• Trailing indicators are the traditional metrics
  that measure past safety efforts.
• When using trailing indicators, data is collected
  after the fact (after a number of accidents or
  illnesses, after two years of workers'
  compensation, etc.)
• Trailing indicators provide an organization with
  feedback in terms of how well the performance has
  been over a period of time
• Examples of trailing indicators include accident
  records, loss reports, injury and illness
  statistics, injury costs, and workers'
  compensation costs

5
Pros and Cons of trailing Indicators

• Some of the pros for using trailing indicators
  include the availability of the data and the ease
  with which the data can be compiled and presented
• To be effective, a cause and effect relationship
  must be identified between the trailing
  indicators and the program procedures
• The safety program procedures must be related to
  that which is being measured by the trailing
  indicators

6
Current Indicators

• Current indicators tell an organization how well
  the management systems are working now, as
  opposed to in the past.
• In addition, they provide a measure of potential
  losses over the short term.
• Current indicators can provide the safety manager
  with immediate feedback on areas that require
  improvement.

7
Leading Indicators

• Leading indicators are those measures that can be
  effective in predicting future safety
  performance.
• Leading indicators can be considered
  before-the-fact" measures.
• These measures assess outcome actions taken
  before accidents occur and are measures of
  proactive efforts designed to minimize losses and
  prevent accidents.

8
Leading Indicators

• Assess the quality of incident investigations and
  arrange for improvement so that investigations
  address true causal factors and so that the
  reports can be a source for selecting leading
  indicators for serious injury prevention.

9
Benefits of Safety Performance Indicators

• Safety performance indicators have certain
  distinct advantages.
• These indicators are flexible and allow for
  individual supervisory styles.
• The same indicator need not be used for all
  supervisors.
• They are excellent for use in objective-setting
  approaches.
• Feedback is swift, since most of these techniques
  require supervisors to report their level of
  performance to upper management.
• They measure the presence, rather than the
  absence, of safety.
• They are usually simple and, thus,
  administratively feasible.

10
Shortcomings of Safety Performance Indicators

• While the benefits of establishing and using
  measurements are numerous, there are also some
  shortcomings that can be experienced if the
  measurements are not properly defined and used.
• Examples of these shortcomings can include 
• A cause and effect relationship between the
  measurement and the activity does not exist.
• There are delays between the activity and the
  measurement.
• Reliability in the measurement process may be
  lacking.
• Outside factors may be influencing results, thus
  affecting the validity of the measurement.