REDUCING CONSTRUCTION-RELATED INJURIES AND THEIR COSTS

INTRODUCTION

It can be argued that construction companies have a moral obligation as well as a legal responsibility to provide a safe working environment for their employees. However, even if these issues are not considered, simple economics dictate that careful and continuous consideration be given to construction safety.

Accidents cost the construction industry over 9 billon dollars every year [1]. Workmen's compensation insurance rates have increased dramatically in the past few years, and evidence indicates that this trend toward increasing rates will continue and very likely accelerate [2].

There is a direct and substantial monetary gain that can be realized by reducing the number and severity of construction accidents. Ideally, corrective actions should be directed so that they yield the highest benefit to cost ratio. Unfortunately, many contractors are not aware of the specific areas where their safety performance could and/or should be improved. The focus of this study was to analyze existing accident data, identify specific problem areas, and make recommendations that will likely yield a high return on investment.

ECONOMIC IMPACT OF ACCIDENTS

The economic impact of accidents can best be discussed by way of a short example. Suppose a contractor has an annual volume of 100 million dollars, which would have placed the contractor 220th on the 1988 Enqineerinq News Record top 400 list. Suppose further that the annual payroll is 10% of this figure, and that the average base rate for workmen's compensation is also 10% of annual payroll. The annual cost of workmen's compensation would then be calculated as noted.

Hypothetical Construction Company
Annual Volume=$100,000,000
Payroll=10% of Annual Volume
Workmen's Compensation Rate=10% of Payroll

Annual Cost of Workmen's Compensation
($100,000,000) (10%) (10%)  $1,000,000

The base rate for workmen's compensation insurance is modified by what is called an experience modification rating (EMR). This rating is based on company safety performance over the past three years. EMR's typically range from 0.5 to 2.0. For the hypothetical case discussed previously, variation in the EMR from 0.5 to 2.0 would result in annual premiums of either $500,000 or $2,000,000 as noted.

In other words, the difference in insurance premium cost for a very poor safety record versus an excellent one for this hypothetical company is $1,500,000. The cost of accidents is not just a fixed cost of doing business. It is a cost over which management has a great deal of control.

INDIRECT COSTS OF ACCIDENTS

In fact, the costs described above are just a percentage of the true cost of accidents. There are indirect costs associated with every accident that occurs on a construction site. A few examples of these indirect costs are listed below.

These items represent just a few of the many factors that are seldom accounted for when the cost of an accident is determined. They are often subjective in nature and difficult to calculate but are nonetheless very real costs that must be paid by the contractor. Because these expenses are usually hidden in total labor costs, management does not get an accurate picture of the true economic damage caused by job-site accidents. Safety experts conclude that it is conservative to estimate the indirect costs of an accident to be 4 times the insurance company direct costs. They may be as high as 17 times the insurance costs [3].

EFFECT OF ACCIDENT FREQUENCY ON INSURANCE PREMIUMS

Two primary factors should be considered when looking at the history of a company's accidents. These factors are frequency and severity. There is a tendency to incorrectly assume that accidents which incur only minor costs are unimportant.

When experience modification ratings are calculated, the frequency of accident occurrence is much more important than the severity of the accident [4]. Minor accidents have a major impact on the amount of money a contractor must pay for workmen's compensation insurance.

RESEARCH PROCEDURE

Four large general contractors agreed to supply all relevant safety records for the 1987 calendar

year. They also agreed to provide corresponding insurance costs wherever possible. Each of the four companies has an annual volume that places them in the Engineering News Record top 400 list, and is engaged primarily in the construction of buildings.

Each accident report form was analysed in detail and broken down into 17 different categories. These categories included such items as age of worker, type of craft, day of week, cause of accident, body part injured, and insurance cost (see appendix charts Al and A2). A total of 594 accident report forms were catalogued in this fashion.

After the forms were analyzed, the more than 10,000 individual data were entered into a customized computerized database that was developed specifically for this project.

The data entry phase was followed by a period of verifying and checking to ensure that no input errors had been made.

Once the data were entered and verified, an . extensive and comprehensive analysis was performed. The data were collated and categorized in a myriad of different ways. Results of this work will be explained in the section that follows.

DISCUSSION ON FINDINGS

Average Insurance Cost

Average insurance cost was determined considering only those cases for which insurance costs were available. Most cases were closed. If they were not, the insurance reserve was added to the cost paid to date.

Analysis of the research data indicated that the average direct cost paid by the insurance companies for each injured employee was $1,541. If the employee's injury resulted in lost or restricted workdays, the average direct cost was $4,347 (Chart 1). Considering that approximately 35 percent of the insurance company's premium revenue goes toward their profit and general overhead costs, the construction contractor's workman's compensation insurance premium cost is about 1.54 times the direct cost paid by the insurance company for injuries. This does not consider the impact of the EMR. As stated previously, the construction contractor may also have indirect costs for accidents that are 4 to 17 times the insurance company direct costs [3].

The average cost of injuries caused by employee acts was found to be 30 percent higher than

for injuries caused by the employee's environment. Considering also that 62 percent of the injuries were caused by employee acts, it is this area that should receive emphasis for corrective action.

The average cost of injuries to carpenters was $2,327, which was more than twice the $1,022 average cost for laborers. This difference nay be caused in part by the fact that carpenters normally earn higher salaries than laborers. The average cost for other crafts was not deter­mined because of insufficient sample size.

Married employees had an average cost per injury of $1,950, which was more than 2.5 times higher than the $741 average cost for single employee. This somewhat surprising statistic may be partially the result of single employees being generally younger and less skilled, hence having a lower wage rate.

Time of Day

When all accidents were considered, the data indicated that most accidents occurred at 10

M, perhaps because the job's production intensity is greatest at mid-morning. The data also showed that there were about 60 percent more accidents in the morning than in the afternoon (Chart 2).

Analysis of accidents which resulted in lost or restricted workdays showed the same trend (Chart 3). It is interesting to note that the data does not support the common notion that accidents occur at the start and ending of the workday. The percentage of accidents caused by an employee's act increased slightly at mid-morning and again late in the day (Chart 4). The average insurance cost per injury was 50 percent higher in the morning than the afternoon, which suggests that the more serious accidents occur in the morning (Chart 5).

Day of Week

Accidents occurred with approximately the same frequency for each day of the week (Chart 6),

but there was a trend for more lost or restricted workday accidents to occur on Friday (Chart 7). Approximately 60 percent of the accidents were caused by an employee's act regardless,of the day of week (Chart 8). However, employee's acts caused over 70 percent of the lost-day accidents, while again showing no significant difference between the days of the week (Chart 9). The average insurance cost paid per injury was highest on Monday at $1,865, decreased as the week progressed, and then rose on Friday to $1,393 (Chart 10).

Employee Age

The number of accidents was determined for various age groups, but without knowing the age mix in the workforce, the frequency or propensity for accidents in any group could not be determined (Chart 11). Research by others has shown that worker's age has no relation to incidence of injuries [5].

The severity of accidents relative to age group, when measured by average insurance cost paid per injury, increased with age to age-group 30 to 34 (Chart 12). The increase in cost is influenced by the likely increase in workmen's compensation benefits generally paid to older workers due to their higher skill and wage rates. The relatively small sample sizes below age 20 and after age 34 precluded reliable results for those age groups.

The cause of accidents, act versus environment, showed no significant difference between age groups. The total sample showed that 62 percent of the accidents were caused by employee acts (Chart 13).

Time Employed and Time on Job

The number of accidents decreased as time on the job and time employed by the contractor increased (Charts 14 and 15). Although it may be that workers new to the job or new to the employer are more likely to have accidents, the data could not be used to support that conclusion because the mix of the workforce relative to time employed and time on the job was not known. Other research has shown that frequency of job-related injuries is directly related to employee turnover [6].

The data did not show a relationship between average cost of accidents and time on the job or time employed, possibly because of inadequate sample size. This was caused in part by these data not being provided in some of the accident reports or by having to discard time on the job for some cases where it was obviously entered erroneously in the accident report. The data did show that there was no relationship between cause of accident, act versus environment, and time on the job or time employed by the contractor.

Marital Status

The marital status of the injured employees was recorded, but the propensity for accidents for married versus single could not be determined without the mix of married and single in the workforce. However, other research has shown that frequency of injuries is not related to marital status [5]. It was found that the cause of accidents was about the same for married and single employees, 60 percent acts and 40 percent environment (Chart 16).

Most Frequent Accidents

The most frequently occurring accidents, body­parts injured, and type injuries are shown in charts 17 to 19. Frequency of accidents and injuries, regardless of average cost, is an important consideration when determining prior­ities for corrective action to lower insurance cost because of its impact on premium rates. As stated previously, the experience rating plan for workman's compensation insurance pre­miums gives greater weight to accident frequency than to accident severity.

After examining the more frequent accidents, it became apparent that some of the categories, such as accidents where an employee was struck by an object or where injuries resulted in cuts, puncture, sting or bite, were too broad to focus on a particular corrective action without further category breakdown. However, the category "foreign body in the eye" has a relatively high frequency, 89 cases out of the 594 total cases, and is sufficiently specific to allow focus on a correc­tive action. Other categories of this type are strains due to lifting, 48 cases; cut fingers and hands, 49 cases; and foot punctures, 29 cases.

Increased emphasis on proper lifting technique would certainly lower the back-strain cases which are 70 percent of the strains due to lifting. The wearing of proper eye protection, gloves, and steel-soled shoes would considerably lower injuries to the eyes, hands, and feet, possibly by as much as 50 percent. When examining lacerated hands alone, it was estimated that 70 percent of those injuries would have been reduced or eliminated by wearing gloves. All of the foot punctures would probably have been eliminated by the wearing of steel-soled shoes.

Most Costly Accidents

The most costly accidents, body-parts injured, and type injuries, as determined by multiplying number of cases by average direct insurance cost per case in each category, are shown in charts 20 to 22. Since the ranking by total insurance cost considers both frequency and severity, it too is a good indicator for deter­mining priorities for corrective action.

Falls from height were found to be the most costly type accident. Thirty-one percent were from scaffolds, often improperly secured. Wet or muddy conditions caused 25 percent of the falls at the same elevation. The diversity of causes suggest only that workers be trained to be more aware of their environment, and that foremen pay more attention to providing a safe working environment.

STANFORD ACCIDENT COST ACCOUNTING SYSTEM

Each accident was analyzed to determine an estimated cost using the Stanford Cost Accounting System [7] and entered into the computer for comparison with insurance company direct costs. It was hoped that a correlation would be found which would allow modification of the Stanford Cost Schedule to account for inflation and allow a recommendation for its use as a predictor of accident costs. However, a correlation was not found, probably because the data base was too small. Nevertheless, construction companies should look for ways in which the cost of accidents might be considered in determining the profitability of individual jobs and in rating the performance of job superintendents.

INCIDENCE RATES

National and state averages are published which list the frequency of OSHA reportable injuries and illnesses and on the frequency of those cases with lost or restricted workdays. The incidence rate for OSHA reportable cases in non-residential building construction in Alabama for 1985 was 21.0 cases per 100 full-time employees [8]. The incidence rate for lost and restricted workdays (LWD/RWD) in Alabama that year for the same workforce was 7.6 cases. Contractors should compare their company's incidence rate with the state and national rates, but they should also recognize that different contractors interpret what is an OSHA reportable case in different ways. This will influence the company's

incidence rate. The incidence rates calculated for the four construction companies using the data provided were as follows:

Are these incidence rates good for comparison of the company's accident experience with state or national rates? Probably not. Contractor A's LWD/RWD incidence rate indicates a high accident rate, but the recordable case incidence rate shows an excellent record. It is possible that contractor A did not interpret many of the minor recordable cases as being recordable. Contractors B and C have good lost workday rates but poor recordable case incidence rates. It is likely that they recorded cases that need not have been recorded. Contractor D has such excellent incidence rates for both categories, that they are questionable. It could be that the reported manhours were more than the work activity for which accidents were considered.

COMPANY SAFETY PROGRAMS

Since formal safety programs have been credited with helping lower injury rates in construction [9], the written safety programs for each of the four companies submitting accident data were requested for review. Only one of the four submittals was considered a comprehensive company safety program. One of the companies submitted a job safety plan similar to that required on Corps of Engineers contracts, one submitted an excellent safety booklet provided to their craftsmen, and one submitted the AGC manuals on accident prevention and OSHA compliance. Although the latter three submittals represented excellent aspects of a company safety program, they were not in themselves sufficient in providing a comprehensive company safety program.

The most important action that a construction company can take to reduce accidents is to establish a policy that makes the safety of its employees as important, if not more important, than production. This policy should be in writing, signed by the CEO, and communicated to all personnel with the understanding that compliance is a condition of employment.

For a large company, a comprehensive safety program should include written descriptions of the responsi­bilities of all personnel as they relate to safety, a project safety requirements planning checklist, employee safety procedures and requirements, safety inspection and accident/incident investigation check­lists and procedures, accident emergency proce­dures, accident/illness reporting and recordkeeping procedures, employee craft skills and safety training, safety motivation program, and subcon­tractor safety requirements. A study in 1980 indicated that the cost of administering a construction safety program amounts to about 2.5% of direct labor cost. The ratio of savings to the costs of administering the program was found to be 3.2 to 1 [3].

A vital aspect of a construction company's safety program is the analysis of safety hazards found on inspections and investigations, and follow-up to insure that corrective action has been taken. A systematic approach is

necessary to be effective and will show management's true commitment to safety. The National Safety Council has published an excellent booklet describing a relatively new and detailed approach to correction of hazards [10].

RECOMMENDATIONS

Wear Eye Protection

Many companies involved in manufacturing require that safety glasses be worn by employees when they are in or near the process areas, essentially eliminating eye injuries. Yet, for some reason, few workers in the construction industry are required to wear eye protection.

Eye injuries comprised 19% of all the accidents that were logged. Most of these accidents were avoidable. Although the average cost for these injuries was relatively low, a company's experience modification rating (EMR) is more closely tied to the frequency of accidents rather than severity. Therefore, the true cost of these accidents is considerably higher than the average cost of $173 per case.

Require Proper Hand and Foot Protection

We recorded a large number of lacerations and puncture wounds to hands and feet. Many of these accidents could have been avoided by wearing gloves and/or safety boots with steel toes and steel soles. Although some may argue that gloves may be cumbersome when working at some tasks, their use would greatly decrease injuries to the hand and should be considered. There is no justification for not wearing safety shoes.

Use Proper Lifting Procedures

Among the most costly accidents are those involving back injuries when lifting. All superintendents should be aware of proper lifting techniques and should educate workers on proper lifting techniques through toolbox meetings or other training methods. While these accidents can never be eliminated, their frequency can be reduced with a proper educational program.

Be Aware of Environment

A suprisingly large number of accidents occurred because a worker slipped on a slick surface. Muddy, oily, or slick surfaces should serve as a red flag to superintendents, and workers should be cautioned accordingly.

Establish a Comprehensive Safety Program

Beyond the moral and legal responsibilities that a construction company has for the safety of its employees, economic interests dictate the desirability of a comprehensive safety program. All large construction companies should establish a written comprehensive company safety program to include:

Keep Better Records

While gathering data from the accident report forms, we were hampered by the inconsistancy between forms from different states, the sometimes inadequate description of the cause of the accident and the extent of the injury, and occasionally the incomplete or erroneous entries in the forms. If company executives are to achieve an effective safety program, they must be able to monitor the type and frequency of accidents that are occurring. Project superintendents should be required to fill out a detailed form in addition to the required OSHA or insurance form. Upon reaching the main office, the data from these forms should be stored in a computerized database, so that they might be organized and categorized for further analysis.

Include Safety Performance in Evaluatinq Superintendents and Job Profitability

Since the safety record of a company affects its profitability, and since personnel are often provided bonuses based on their contribution to profitability, it is recommended that some measure of the cost of accidents be considered in evaluated performance. One such method might be to use the direct cost paid by the insurance company plus any reserve set aside for accidents occurring on a project. The true cost of the accidents may be considerably higher than this cost. Some of the cost is already charged to the job in lost productivity, rework, and overhead charges, and some of the cost will show up on future jobs in the form of higher insurance premium cost. However, much may be gained by drawing attention to the economic aspect of safety.

ACKNOWLEDGEMENTS

The authors wish to extend sincere gratitude to Mr. Steve Belflower, student research assis­tant, for the excellent job he did in assisting with the gathering and analysis of data for this report. In addition, we wish to thank industry members Jim Rein, Robin Savage, Jim Gorrie, and Ben Nevins for providing their time and the mass of information upon which this report is based.

Lastly, a very special thank you to Mrs. Jane Parrotte for her endless patience and her excellent advice regarding the preparation of this document. She is a unique and invaluable asset to the Department of Building Science.

REFERENCES

1. The Business Roundtable, Improving Construction Safety Performance, 200 Park Avenue, New York, N.Y., January 1982.

2. "Premiums Pulse Upward for Workers' Compensa­tion," ENR, September 15, 1988, p. 58.

3. "Improving Safety Helps Improve the Bottom Line," Civil Engineering, ASCE, October, 1982, p. 48.

4. National Council on Compensation Insurance, An In-depth View of Experience Rating, One Penn Plaza, New York, N.Y., 1982.

5. Hinze, J., "Human Aspects of Construction Safety," Journal of the Construction Division, ASCE, Vol. 107, No. COI, March, 1981, p. 63.

6. Hinze, J., "Turnover, New Workers, and Safety," Journal of the Construction Division, ASCE, Vol. 104, No. C04, December, 1978, p. 412.

7. Levitt, Raymond E. and Samelson, Nancy M., Construction Safety Management, McGraw-Hill, New York, N.Y., 1987, p. 30.

8. Alabama Department of Labor, Alabama Occupa­tional Injuries and Illnesses Survey 1985, 64 North Union Street, Montgomery, Alabama.

9. Hinze, J. and Harrison, C., "Safety Programs in Large Construction Firms," Journal of the Construction Division, ASCE, Vol. 107, No. C03, September, 1981, p. 464.

10. National Safety Council, Accident Investiga­tion ...A New Approach, 444 North Michigan Avenue, Chicago, Illinois, 1983.