Work Schedules: Shift Work and Long Work Hours

Extended Abstracts from Conference:
Long Working Hours, Safety, And Health: Toward A National Research Agenda
University of Maryland, Baltimore, Maryland
April 29 - 30, 2004
Using a multi-disciplinary approach, this conference explored the sociological, economic, and health dimensions of long work hours.

Todd Dawson, Anneke Heitmann, Alex Kerin
Circadian Technologies, Lexington , MA

Introduction

Employees in the U.S. work the highest number of hours per year compared to the rest of the world – about 70 more hours per year than workers in Japan , and 350 hours more than in Europe . This discrepancy is largely due to longer work weeks and fewer weeks of vacation in the U.S. In addition, overtime and second jobs add to the regular work week. This presentation focuses particularly on the trends, risks and costs, and the management of long working hours in extended hours operations, businesses that operate outside the 7 a.m. to 7 p.m. timeframe.

Method

The data presented here are based on Circadian’s research on extended hours operations. The main data sources include Circadian’s growing employee survey database and on Circadian’s annual Shiftwork Practices Survey. The employee database includes data from over 60 companies in diverse industries including responses from over 10,000 employees. The annual Shiftwork Practices Surveys collected data from managers in extended hours facilities in the U.S. and Canada covering all industry sectors. Here we present data from the 2004 and 2002 Shiftwork Practices Surveys (Circadian Technologies, 2002; Kerin, 2004) reflecting responses from 623 and 550 facilities, respectively, and representing approximately 120,000 and 160,000 shiftworkers, respectively. In addition, case studies of work schedule interventions at selected client sites are reported.

Trends of Long Working Hours

While working hours are federally regulated for some occupations (e.g., transportation workers), most workers do not have federal requirements for their working hours. In the transportation industry, rules limit the maximum number of consecutive work hours, maximum number of weekly work hours, and minimum duration of rest periods. For example, the recently updated regulations for truck drivers limit the maximum duration of work periods to 14 hours (including up to 11 hours of driving) after which a minimum rest period of 10 hours is required. In addition to these federal regulations, some states do mandate maximum work day and week hours, after which non-exempt employees could choose to go home. Regulations prohibiting mandatory overtime in health care are also becoming common in some states. While these federal and state rules give some protection against excessive working hours, they cannot fully prevent work schedule related health and safety risks as they do not sufficiently take into account circadian aspects, and do not prevent irregular and unpredictable work patterns.

For most non-regulated industries, work shifts vary usually between 8 and 12 hours. According to Circadian's most recent Shiftwork Practices Survey, 43% of the extended hours facilities surveyed used 12 hour shift schedules, compared to 38% who used 8-hour shifts (Kerin, 2004). However, overtime can dramatically extend the work day. Almost 60% of the responders to the Shiftwork Practices Survey stated that the maximum hours that employees are allowed to work per day are between 12 and 16 hours. Almost 23% of facilities had either no limit on maximum hours or a limit in excess of 16 hours. Over 40% of facilities had no limit on the maximum number of consecutive days worked. Operations using 12-hour schedules were more likely to impose a limit.

Overtime in extended hours operations was on average 12.6% in 2003, slightly higher than in 2002 (11.9%). Overtime varied among industries, with utilities, process manufacturing, and services having the highest levels (Figure 1) (Kerin, 2004). Within each industry, overtime levels varied between facilities. According to Circadian's Shiftwork Practices Survey (Circadian Technologies, 2002), most extended hours facilities (74%) had an average individual overtime level of under 300 hours per year (14.4% overtime). However, about 10 % of the surveyed facilities had an average individual overtime level of 500+ hours per year (24.0 + % overtime). Higher overtime levels are associated with higher fatigue. Facilities reporting severe fatigue problems averaged 15.4% overtime compared to 8.7% for facilities who reported no fatigue problem (Figure 2). Within each facility, overtime is rarely distributed evenly, that is, a minority of the employees works the majority of the overtime (Figure 3), some of them reaching extreme levels of over 1000 hours per year.

Figure 1: Overtime levels in extended hours operations, by industry (Kerin, 2004)

Figure 2: The effect of overtime on employee fatigue in extended hours facilities (Kerin, 2003)

Figure 3: Overtime distribution at an extended hours facility (Kerin, 2004). In this example, 20% of the employees are working 60% of the overtime

In a tight economy, overtime is not only increasing, but also becomes less optional. Circadian’s Shiftwork Practices Survey showed that 41.5% of the extended hours facilities required at least some mandatory overtime to provide the coverage they need, and in 9% it was entirely mandatory (Circadian Technologies, 2002). The method of assigning overtime (e.g., hold-over or on-call) makes an impact on workers performance, morale, fatigue and safety. About three-quarters (73%) of the extended hours facilities used hold-overs, where the normal shift length is extended, and 40% used hold-overs exclusively. This may give employees limited warning of their overtime assignment, and often results in double shifts without sufficient preparation. About one-quarter (27%) relied exclusively on call-ins, which reduce the number of days off, but do not usually increase the length of the work day.

The issue of long working hours becomes more complex and severe if one also considers second jobs. According to the Bureau of Labor Statistics, 5.4% of U.S. workers work two and more jobs. Circadian data collected from shiftworkers demonstrate an even higher percentage with 28% of full-time extended hours employees working a second job. Over half of these people work more than 10 hours per week at a second job (Kerin and Carbone, 2003).

Public awareness of long working hours and overtime issues is increasing, stimulated, for example, by revised federal regulations for truck drivers, the introduction of limits on medical residents, strikes against mandatory overtime, new overtime payment rules, overtime law suits and driving accident litigation related to work hours.

Costs of Long Working Hours

Implementing research on the health and safety effects of long working hours into the real world can be accomplished most effectively if supported by economical data as the financial bottom line is one of the main operational drivers. As overtime of non-exempt employees is usually paid at a rate of one and a half time their hourly wage, businesses often justify overtime based on the costs of training and benefits for new hires and weigh them against the hourly wages. For example, the average costs to employers for recruiting and training an extended hours worker is $25,550, but varies widely depending on the industry (Kerin, 2004).

While some amount of overtime is often operationally needed and desired by some employees, excessive overtime hours can compromise safety, health and productivity – adding to the true costs of long working hours of exempt and non-exempt workers. These indirect costs relate, for example, to an increased risk of heart attacks; diabetes; high blood pressure and mental illness; a greater risk of retirement disability; increased safety risks due to human error; lowered productivity and presenteeism; increased chance of turnover and absenteeism; and costs of potential liability issues and law suits.

Few studies have directly investigated the financial consequences of long working hours. For example, in a study on white-collar jobs, performance decreased by as much as 20% when 60 or more weekly hours are worked (Nevison, 1992). Data from 18 manufacturing industries in the U.S. show that a 10% increase in overtime resulted, on average, in a 2.4% decrease in productivity measured by hourly output ( Shepard and Clifton, 2000). High overtime levels can cause poor employee morale, which can affect productivity and absenteeism. For example, Circadian showed that 31% of extended hours operations that have extremely high overtime hours (25% or greater) also had poor morale, compared to only 13% of companies with low or normal overtime (Kerin, 2003). Long working hours and overtime contribute to increased worker fatigue and safety problems. For example, the average cost of workers compensation claimed per individual at extended hours facilities that reported severe fatigue problems was considerably higher ($4,037) than at facilities that report moderate ($2,240), minor ($981) or no ($276) fatigue problems (Figure 4) (Kerin, 2004).

Figure 4: Average cost of workers compensation per worker for facilities with different levels of fatigue problems (Kerin, 2004).

Management of Long Working Hours

There are various reasons for overtime such as: a temporary or permanent increase in demand for production or service; changes in the pattern of demand; non-sufficient productivity of existing equipment or employees; scheduled or unscheduled employee absences; lack of policies and procedures to control overtime; maintenance schedules or unexpected work/emergency situations; or company downsizing strategies. Approaches for reducing overtime depend on the root causes of overtime.

For example, the continued push toward lean manufacturing, with short lead times and low inventory, and the greater demand for night and evening customer service has resulted in larger fluctuations in production and service levels by hour of day, day of week, week of month, month of year. If staffing levels are not matched to meet the varying demand, there will be situations where the operation is understaffed and overtime needs to be added, and other times when employees sit idle. This can be addressed by flexible and proportional staffing. Circadian’s Shiftwork Practices Survey (Kerin, 2004) showed that 42% of facilities experience fluctuations throughout the day, yet only 17% of them attempted to match staffing level to these changes. Facilities in which demand varied by time of day tended to have higher overtime rates than those with no fluctuations (14% versus 11.6%), and facilities that matched staffing level to daily demand fluctuations tended to have lower overtime than those that did not (12.1% versus 13.4%).

Managing overtime through proportional staffing is demonstrated in the following case study of a police department (Dawson and Trutschel, 2003). To ensure that officers’ time was used most effectively, the volume of priority calls that were received each hour and day over the average week was analyzed. Times when understaffing and overstaffing occurred were identified. A number of potential shift schedules were designed that matched demand better, and the officers chose the option that worked best for them. As a result, the coverage expected and desired by the city and the department was provided and overtime levels were lowered. The efficiency of the new schedule (a measure of the utilization of the officers’ time) rose to 80%, compared to 66% previously.

The management of overtime should be taken into account in company policies. While overtime policies can help limit excessive overtime, other human resource policies can also contribute to discouraging overtime as the following case study illustrates (Kerin, 2003). After a serious fatigue-related workplace accident, a company investigated overtime logs and found that the most senior staff, who represented 10% of the workforce and included the injured worker, were working 40% of the total available overtime hours. As the employees’ pensions were based on the average annual compensation of the final five employment years, older employees were requesting and receiving a disproportionate share of the overtime. The pension policy was changed, removing the incentive for excessive overtime which resulted in a better distribution of overtime across all employees.

The number of working hours (e.g., overtime) is only one dimension of work schedules. Companies and employees can also benefit from interventions related to other characteristics such as start/end times of work periods, regularity of work patterns, speed and direction of shift rotation, etc. For example, a study with trucking companies showed that driver fatigue levels as well as number and severity of accidents could be reduced by changes to the daily and weekly work and rest patterns, including adjustments to the start and end times of work shifts, reduction of the number of consecutive work shifts, and provision of rest breaks which allowed two consecutive nights of sleep (Moore-Ede et al. 2004). The number of truck accidents dropped 23%, and the average cost per accident dropped 65.8%. The number and average cost of severity accidents (over $20,000) dropped 65% and 66.7%, respectively.

References

Circadian Technologies (2002). Shiftwork practices 2002. Circadian Technologies, Lexington , MA

Dawson, T. & Trutschel, U. (2003). Flexible workforce management. Circadian Technologies, Lexington , MA .

Kerin, A. (2003). Overtime in extended hours operations: Benefits, costs, risks, and liabilities. Circadian Technologies, Lexington , MA .

Kerin, A. & Carbone, J. (2003). Financial opportunities in extended hours operations: Managing costs, risks, and liabilities. Circadian Technologies, Lexington , MA .

Kerin, A. (2004). Shiftwork practices 2004. Circadian Technologies, Lexington , MA .

Moore-Ede, M., Heitmann, A., Guttkuhn, R., Trutschel, U., Aguirre, A. & Croke, D. (2004). Circadian alertness simulator for fatigue risk assessment in transportation: Application to reduce frequency and severity of truck accidents. Aviation, Space and Environmental Medicine 75 (3), A107-A118.

Nevison, J. (1992). White collar project management questionnaire report. Oak Associates, Concord , MA ..

Shepard, E. & Clifton, T. (2000). Are long hours reducing productivity in manufacturing. International Journal of Manpower 21(7), 540-552.