POWER February 2015 - 50

SAFETY
Best Practices for Aligning
Safety Metrics, Incentives, and
Performance
The Occupational Safety and Health Administration requires certain incidents
to be recorded and reported, which generates a set of statistics that many
companies use to gauge safety performance. However, other metrics may
be better predictors.
Aaron Larson
T
he Occupational Safety and Health Administration
(OSHA) requires all employers
with more than 10 employees,
and whose establishments are not classified
as a partially exempt industry, to report workrelated
injuries and illnesses. Companies
must post statistics annually, such as the total
number of deaths, cases with days away from
work, cases with job transfers or restrictions,
and total other recordable cases.
The numbers are eye-opening-on average,
more than 12 people die on the job every
single day in the U.S., and more than one
million cases each year require days away
from work to recuperate. But these statistics
don't tell the whole story. Each number represents
people who did not expect to be killed
or injured on the job when they went to work.
Companies want those numbers to be zero,
but how can they get there?
Assessing the Safety Culture
It's important to understand that OSHA
metrics are trailing indicators of safety performance.
Once a person is injured and the
incident recorded, the only thing a company
can do is try not to let something similar
happen again. So while OSHA data offers
a useful snapshot of historical performance,
better, more predictive metrics are needed to
gauge safety trends within a company and
determine what areas must be improved to
prevent future incidents.
Conducting regular safety audits is one
option. These often include a standard
checklist of safety inspection items, such
as: Are workers wearing their personal protective
equipment? Have they locked out
the system they're working on? Has confined
space monitoring been completed, if
necessary? While audits provide a standard
method for monitoring safety, some safety
experts argue that they rarely do more than
50
verify whether or not workers are meeting
the minimum requirements.
According to Gary Higbee, president of
Higbee & Associates Inc.-a full-service
safety training and consulting firm-and
co-author of Inside Out: Rethinking Traditional
Safety Management Paradigms,
safety culture is not driven by compliance.
He says compliance is extremely important-a
legal obligation-but employees
need training on individual skills that they
can use everywhere.
" Do they look before they move? Are they
working at a reasonable speed? Are they in
a good body position related to the work? "
asked Higbee. " Those are behaviors that employees
control. "
Behavior Observations
Higbee recommends conducting observations
that evaluate employee behaviors. He says
that there are four " states " -or conditions-
that lead to errors: rushing, frustration, fatigue,
and complacency. The significance of
an error is a function of luck and the amount
of hazardous energy that is present in the vicinity
of an incident, but it's the behaviors
that cause it.
Some estimates suggest that up to 90% of
workplace accidents are caused by human error.
An error may or may not lead to an injury,
but if a company can modify employee
behaviors to prevent errors from occurring,
injury rates are bound to improve.
For example, a worker could lose his grip
on a 10-pound sledgehammer. In one instance,
the hammer could fall harmlessly to
the deck grating directly below, causing no
real harm. In another case, the hammer could
fall through an opening in the deck grating
and strike someone on the ground 50 feet below.
The error-dropping the hammer-was
the same in either case, but the result of the
www.powermag.com
incident would be considerably different.
The question a company must ask is: Why
was the hammer dropped? This is where
many people may not consider all of the factors
that led to the error. It is easy to say such
things as: " It was an accident. " " It slipped. "
" It was an act of God. " or " Stuff happens. "
But is it really that simple? If we try to determine
which of Higbee's four states led to
the error, we may find that the accident was
predictable. Consider the states and some
questions that could be asked:
■ Rushing. Did the incident occur near the end
of the day as the worker was trying to finish
a job so he could go home? Was there time
pressure placed on the worker to get a piece
of equipment returned to service?
■ Frustration. Had the individual been
struggling for an extended period of time
on a difficult task with little success? Did
he feel that he was not getting the help that
he needed?
■ Fatigue. Was the person working an abnormal
shift or more hours than normal?
Does the individual have a young child
who may have interrupted sleep the night
before?
■ Complacency. Had the person done the
job numerous times before without incident?
Was the work considered a simple
task that did not require complete focus?
Companies try to compensate for errors in
many ways. Personal protective equipment,
such as a hard hat protecting an employee
from being injured by falling objects, or a
lanyard keeping a tool from falling beyond a
certain point, can help, but in these examples
the errors still happen. One key to preventing
injuries is to reduce the number of errors that
occur in the first place.
To get a snapshot of a work group's safety
POWER | February 2015
http://www.powermag.com
POWER February 2015

Table of Contents for the Digital Edition of POWER February 2015
