Expert: Mike Brown CSP Retired - 11/15/2001

Question
From which columns in th OSHA Form 200 can get the figures for which to calculate the incidence rate and severity rates?

Answer
Subject: Incidence & severity rates

Question: From which columns in the OSHA Form 200 can you get the figures for which to calculate the incidence rate and severity rates?

Answer:  Welcome to the wonderful world of number games.  There is an old joke about the manager who asks his safety director “And what is our incident rate?”  And the Safety Manager responds “And what would you like it to be?”

If you expect a simple answer to this question you will be disappointed.  Getting dependable, useful information from accident/incident statistics from OSHA is like asking a woman her age (being retired means I don't have to worry about being Politically Correct).  The answer is probably accurate to plus or minus 50 percent.  One of the first things you learn in safety is how to manipulate statistics.  The better you become, the further you will advance.

The rates will not come just from the columns in the OSHA Form 200.  At the end of the year, the page or pages must be totaled at the bottom of the sheet (or last sheet).  To arrive at the total number of “recordable incidents” you will need to add the following totals for the year (or time period involved).

---(1) Injury Related Fatalities
---(2) Injuries With Lost or Restricted Work Days
---(6) Injuries Without Lost or Restricted Work Days
---(8) Illness Related Fatalities
---(9) Illnesses With Lost or Restricted Work Days
---(13) Illnesses Without Lost or Restricted Work Days

These totals will serve as the number of incidents you may use in the calculation of the “incident rate” (if you are calculating what I would call the overall Incident Rate).  You will also need the hours of exposure (hours worked) from a source such as payroll or personnel.

First the (Overall) Incident Rate is calculated by taking the total number of  “recordable incidents (as totaled from above)” times 200,000 and dividing the result by the number of hours of exposure. (Note: 200,000 is an arbitrary number established by OSHA and is supposed to represent the hours worked in a year by 100 employees – 100 employees times 50 weeks per year times 40 hours per week.)

(I am not going attempt to define the term “recordable incident or injury” in this answer as OSHA has an entire book on the subject which causes most lawyers to salivate with the way their answers and explanations can be twisted and distorted.  BS ain't a place in Kansas!  What gets counted depends on whom you listen to.  Management will not want to count anything unless there are color images of the blood and gore on the six o'clock news and labor will want to count everything including hangnails.  All that you can be certain of is that you and the plant next door are using different definitions and a comparison of your safety efforts and results with theirs is like comparing apples to oranges!  All this makes for wonderful accuracy which only a politician or government official will try to defend.)

Example – A small plant with 300 employees works the year of 2001 with 6 “recordable injuries “ of which 4 resulted in days away from work.  The hours worked for the year for the facility are 310,015.  The incident rate for the facility is (6 x 200,000)/ 310,015 = 3.870 recordable injuries per 100 employees.

Do not forget the units on the result.  Many times you will see the incident rate listed as x.xx which has some meaning for those in safety who automatically understand the “recordable injuries per 100 employees”.  But this may mean little if anything to managers, supervisors or others who do not work with the calculations on a regular basis or have not been indoctrinated by the safety department.

Now comes the confusion.  What incident rate do you want?  In the example above, the overall incident rate based on the total number of “recordable injuries” is 3.870 recordable injuries per 100 employees.  There can also be an “incident rate” for the injuries resulting in lost work days (or time-loss injuries) which would be calculated as (4 x 200,000) / 310, 015 = 2.581 lost work day incidents per 100 employees.  Many companies will track several or more “incident rates”.  In a company where both illnesses and injuries occur, a decision may be made to track a “recordable injury incident rate”, a “recordable illness incident rate” and an “overall incident rate”.  When we were focused on repetitive injuries as a part of our ergonomic program, we also kept track of this type of injury through an “repetitive injury incident rate”, to see if we were making progress.  We also kept track of “incident rates” for departments within several of our larger operations.  This was done as part of supervisory accountability efforts.  We also tracked “incident rates” on a monthly basis and even weekly in reports to corporate.

As you can see, there are a number of ways this can be calculated depending on the needs, desires (and often whims) of the company or facility.  Thus stating that XYZ Company has an Incident Rate of 2.35 and thus a wonderful safety record is like stating “the sky is blue”.  In the example above, if the six injuries were very minor scratches where the employee was taken to an emergency room and given a prescription for antibiotics to prevent infection the incidents are “recordable” and the overall incident rate is 3.870 (Case A).  If the six injuries each resulted in lost work time with an average of 100 days lost for each injury the overall incident rate is 3.870 (Case B).  And even if the cases were six fatalities, the overall incident rate will be 3.870 (Case C).  Not quite the same in my book.  Like I said, “And what would you like the number to be?”  This explains in part why I always was left wondering what was really meant when someone told me his or her incident rate was such and such.  The subject of “Incident Rates” is one of the many reasons many safety managers will be heard to mutter that there are those who believe that OSHA means Our Savior Has Arrived.)

Severity is not an “official” measurement according to OSHA but a carry over from the old days of the National Safety Council, which used Frequency and Severity Rates.  Accident Frequency was calculated the same as Incident Rate but used different definitions for accidents and 1,000,000 instead of 200,000.  But then what would the National Safety Council know about Accident Statistics?  If I were to calculate a severity rate that meant something to me, I would use the following:

--- A Recordable Incident Without Lost or Restricted Work Days = 1 point each
--- A Recordable Incident With Lost or Restricted Work Days = 5 points each, times the number of days lost or restricted
--- A Recordable Incident Resulting in a Fatality = 10,000 points each

(Thus in the examples above the Severity Rate would be Case A = The quantity of 6 times 1 times 200,000 divided by 310,015 equals 3.870.  Case B, Severity Rate = The quantity of 6 times 5 times 100 times 200,000 divided by 310,015 equals 1935.39.  Case C, Severity Rate = The quantity of 6 times 10,000 times 200,000 divided by 310,015 equals 38,707.8.  These are rates which I feel could be compared and given some sort of validity when making a judgement on the safety effort or lack of safety effort.)

Another method would be to take the number of lost or restricted workdays (or a part of) and substitute it in the “Incident Rate” formula.  In the example this would give you, Severity Rate = (4 Incidents x 100 Lost or Restricted Days x 200,000) divided by 310,015 = 258.05 lost/restricted work days per 100 employees.  You need to determine what specific approach your company or industry group uses and follow their methods (if you agree).  You will find that even within companies there may be locations where local management has “decreed” what will and will not be counted as they wish to cover their behinds.  (The bigger the money losses the better the safety performance, they need to find some way to get that bonus.)

I hope this begins to answer your question – but as you can see, in the real world, there is not a simple, true way these numbers are calculated or compared.  I always felt that the only accuracy that came in our division was because we forced everyone to use the exact same definitions (mine) and then compared the results only from year to year and not with anyone else.  If you take that approach you can use Incident Rate and Severity Rate as a meaningful tool.  If I can provide any additional information, please feel free to contact me at Mike@kilchis-river.com and will attempt to answer any questions.  (I am willing using that address to give you a phone number if that will assist.)

Michael V. Brown CSP Retired