Occupational Health and Safety in Hong Kong's Catering Industry

Introduction

Hong Kong is known for its international cuisine and, characteristically, a high number of restaurants — many offering catering services — exist in the city. Yet the polished environments of world-class restaurant dining areas contrast sharply with kitchen and backroom workspaces. The catering industry in Hong Kong attains extraordinarily high rates of occupational accidents. According to the 2010 Occupational Safety and Health Statistics Bulletin of the Labour Department of Hong Kong Special Administrative Region (HKSAR) of the People's Republic of China, the catering sector has the highest rate of occupational accidents of any industry in the territory — exceeding even construction, manufacturing, mining and quarrying, transport and servicing, and utilities. Numerous catering companies in Hong Kong have attempted to implement safety measures to reduce workplace accidents, and considerable resources have been directed toward these efforts (Kuen, 2005). Nevertheless, corrective measures have not made a substantive impact on worker safety in the catering industry (Kuen, 2005).

The continued growth of global tourism further underscores the urgency of this problem. With approximately 10,000 eating establishments accounting for roughly 170,000 jobs, it is important that this mainstay service industry maintain a robust operational capacity capable of sustaining high levels of employment (Kuen, 2005).

Most studies on accident rates in catering companies have focused on the causes and costs of unsafe workplaces. Researchers have examined the problem from the perspectives of personnel injury, direct and indirect company costs, company reputation, job retention, and employee morale (Kuen, 2005). Direct costs addressed in the literature include workers' compensation expenses, sick leave taken by injured workers, occupier's liability premiums, and legal representation costs (Kuen, 2005). Indirect costs can be difficult to quantify or estimate (Kuen, 2005). While a wide array of worker-related statistics is available through government agencies such as the Census Statistics Department and the Labour Department, the thrust of information collection in government and academia has been the aggregation of data. There is a conspicuous dearth of prescriptive information about how to mitigate the situations that lead to high worker accident rates. Businesses have long relied on case studies to provide models for transformative practice, and this research is grounded in the belief that catering companies in Hong Kong can benefit from reviewing case studies that illustrate effective safety programs.

This study is designed to identify causal factors in occupational accidents occurring in the workplaces of catering companies, and to identify approaches and methods used by companies overseas where catering businesses have successfully lowered their accident rates.

A partnership program of the Occupational Safety and Health Council (OSHC), the Green Cross Group offers research outcomes, benchmark studies, and training programs in the area of safety and health management ("OSHC," 2012). The OSHC is a statutory body for promoting workplace health and safety and for safeguarding and sustaining the Hong Kong workforce ("OSHC," 2012). OSHC facilitates networking for the purpose of benchmarking the safety and health performance of different organizations against the best performers in their industries ("OSHC," 2012). A primary purpose of the benchmarking initiative is the identification of variables associated with best practices in order to create models and make recommendations for peak performance in safety and health management ("OSHC," 2012). OSHC also provides a lending library, creates local networking events, and convenes meetings of participating organizations to form common interest groups by organization or functional process ("OSHC," 2012). Safety training is made available to organizations through the OSHC in order to explore the application of benchmarking and quality improvement initiatives ("OSHC," 2012). This effort is strengthened by local and international study missions, workshops, and symposia, with the shared expectation that these experiences will yield deeper insights into best-practices implementation ("OSHC," 2012).

Literature Review: Occupational Safety Data and International Comparisons

Green Cross research has established that workplace injuries occurring in the catering business in Hong Kong tend, on the whole, to be less severe than injuries in the construction industry ("OSHC," 2012). However, on an absolute basis, the number of injuries in the catering sector exceeds the number in construction ("OSHC," 2012). In 2004, Green Cross conducted a survey-based investigation into the number and root causes of industrial accidents ("OSHC," 2012). Of the 5,827 accidents that occurred in the catering business, they accounted for roughly half of all industrial accidents. The number of accidents in the construction trade fell steadily from approximately 14,000 in 1999 to about 4,000 in 2003 — roughly one-third of the 1999 rate ("OSHC," 2012). In contrast, accidents in the catering industry fell from approximately 12,500 in 1999 to 5,827 in 2003 — just under half the 1999 rate ("OSHC," 2012). Moreover, the number of catering-industry accidents was consistently higher than the number of construction-industry accidents, and the gap between the two industries was substantial in 2001, 2002, and 2003 ("OSHC," 2012). In summary, each major high-risk industry experienced a substantive drop in accident rates over this period, yet the catering industry continued to record the highest rates ("OSHC," 2012).

The Labour Department has established a comprehensive database on work-related issues, including accidents in the catering industry ("OSHC," 2012). Approximately two-thirds of worker injuries in catering occupations are caused by the use of hand tools, contact with hot substances or surfaces, and some type of fall, including slipping and tripping ("OSHC," 2012). Additional common causes include striking a fixed or stationary object, lifting and carrying injuries, and contact with moving machinery or objects ("OSHC," 2012). Catering workers also sustain injuries from being trapped between objects, from falling objects, and from exposure to harmful substances or animal contact in the course of their work ("OSHC," 2012).

Labour Department statistics show that the overall number of accidents in 2001 was 10,149, with an accident rate of 54.7 per 1,000 employees ("OSHC," 2012). Of the 49,649 employee compensation cases reported in 2002, 21% — or 10,484 cases — originated in the catering industry ("OSHC," 2012). In 2003, 47,376 injury cases resulted in employees taking sick leave for three days or more; 10,358 (22%) of these were from the catering industry ("OSHC," 2012). Injured employees and their families received approximately $61 million in compensation, representing 7.7% of the total $804 million paid in 2003 for all employee-related accidents ("OSHC," 2012). Approximately 160,000 workdays were lost due to workplace injuries in 2003 at a total cost exceeding $1.2 million ("OSHC," 2012). These figures underscore the strong financial motivation for improved workplace safety at the corporate level ("OSHC," 2012).

In 2001, the OSHC conducted a survey on the conditions of kitchen workers in Chinese restaurants (Kuen, 2005). A total of 471 kitchen workers — including chefs, cooks, assistant cooks, grill cooks, dim sum cooks, and material preparation cooks — from 159 different Chinese restaurants participated (Kuen, 2005). Of those surveyed, 80% reported being injured on the job, and more than 60% took sick leave due to work-related accidents (Kuen, 2005). Injuries included cuts, burns, musculoskeletal problems, and stab wounds from shrimp or fish (Kuen, 2005). The predominant health problems reported by kitchen workers were gastrointestinal disorders and headaches (Kuen, 2005).

The OSHC conducted a parallel survey in 2001, with 388 participants who were exclusively floor staff — those not working directly in kitchen areas — from 69 Chinese restaurants (Kuen, 2005). Floor staff included receptionists, waiters and waitresses, head waiters and waitresses, food servers, dim sum servers, stewards, managers, supervisors, and table-clearing staff (Kuen, 2005). The OSHC was particularly interested in how the hectic restaurant environment might be affecting stress levels among floor workers, and accordingly investigated both physical injuries and mental health problems that could leave workers vulnerable to further injury (Kuen, 2005).

In the United States, the North American Industry Classification System (NAICS) code 445 represents Food and Beverage Stores of the Retail Trade Sector (44–45). Industries in this subsector typically retail food and beverage merchandise from fixed point-of-sale locations, using specialized equipment such as freezers and refrigerated display cases, and employ staff trained in food processing to ensure proper storage and sanitary conditions required by regulatory authorities (NAICS, 2012).

The Bureau of Labor Statistics (BLS) annually compiles, analyzes, and reports the number and frequency of work-related fatal injuries and nonfatal injuries and illnesses ("Bureau of Labor Statistics," 2012). The Census of Fatal Occupational Injuries (CFOI) and the Survey of Occupational Injuries and Illnesses (SOII) provide the underlying data ("Bureau of Labor Statistics," 2012). The CFOI uses diverse sources — including federal and state agency records, workers' compensation reports, death certificates, and news accounts — to identify, verify, and profile fatal work injuries across the United States ("Bureau of Labor Statistics," 2012).

The 2011 BLS data for food and beverage stores shows: approximately 2,859,700 persons employed; 4.7 incidents per 100 full-time workers; and 2.7 cases per 100 workers involving lost work time, restricted duties, or job transfer due to injury ("Bureau of Labor Statistics," 2012). These figures assume greater significance in light of projections that food and beverage industry employment will continue to grow. In 2008 alone, approximately 6,299,930 food and beverage serving workers were employed in the United States; if the growth rate at that time continued, this number was projected to increase by roughly 10% by 2018 (Martinez, 2011). The breakdown of food and beverage server positions in 2008 was approximately as follows: 2,371,750 waiters and waitresses; 2,708,840 combined serving and food-preparation workers; 527,530 counter attendants; 503,420 bartenders; and 188,390 non-restaurant food servers (Martinez, 2011).

Jobs across the food and beverage industry tend to attract young workers — many still teenagers or in their twenties — who have little or no prior work experience (Martinez, 2011). The nature of food and beverage industry work exposes employees to hazards not shared by workers in most other sectors. Employees often work long shifts spanning lunch and dinner service and spend most of their time on their feet, conditions that contribute to fatigue and the corresponding mental impairment and physical slowness that predispose workers to accidents (Martinez, 2011). Add to this the routine handling of heavy trays and containers of hot food, breakable dishware, sharp utensils, and slippery floor surfaces, and the elevated accident rate becomes unsurprising (Martinez, 2011).

Over the six years from 2003 to 2008, the number of fatalities among food and beverage workers reached 141 (Martinez, 2011). In 2008 alone, there were 26,870 nonfatal injuries and illnesses causing days away from work among food and beverage servers in the private sector — down 29% from the 37,860 nonfatal injuries recorded in 2003 (Martinez, 2011). By comparison, all other occupations combined recorded 1,078,140 nonfatal injuries and illnesses in 2008, a figure that had decreased 18% from 1,315,920 in 2003 (Martinez, 2011). For nonfatal occupational injuries and illnesses requiring time off, food and beverage servers recorded approximately 70.6 cases per 10,000 full-time workers in 2008, compared with 113.3 per 10,000 full-time workers across all other occupations combined (Martinez, 2011).

An important distinction must be made between injuries resulting from accidents and those resulting from workplace violence (Martinez, 2011). Most fatal injuries to food and beverage workers result from homicides associated with crimes, intended crimes, or interpersonal disputes in the workplace (Martinez, 2011). By contrast, most nonfatal injuries and illnesses in food and beverage workers result from accidental events during normal job performance — such as contact with dangerous equipment or substances, slipping or falling, and overexertion resulting in strains (Martinez, 2011).

Disaggregation analysis allows researchers to examine relationships between the parts of the body injured, the source of injury, and the nature of the accidental event (Martinez, 2011). Using 2008 as the unit of analysis, the two most prevalent injury types for food and beverage workers were sprains and strains, and cuts and lacerations (Martinez, 2011). In 2008, 1,910 food and beverage servers experienced sprain or strain injuries from overexertion, with the trunk as the injured body part and containers as the source; of these, 1,440 resulted specifically from overexertion in lifting (Martinez, 2011). Additionally, 1,090 workers incurred cuts or lacerations to the finger or fingernail when a knife or the item being cut slipped during use (Martinez, 2011). These injury types are common to food and beverage workers and comparatively rare in other occupations (Martinez, 2011).

The Institution of Occupational Safety and Health (IOSH) is the chartered body for health and safety professionals, with approximately 40,000 members across 85 countries, making it the world's largest professional health and safety organization ("IOSH," 2012). IOSH pursues standard-setting, member support, resource provision, training, guidance, and lobbying on issues affecting millions of workers, and holds registered-charity and international-NGO status ("IOSH," 2012).

Beginning in September 2012, IOSH launched an annual survey in the UK designed to examine accidents in the food and drink industry closely ("IOSH," 2012). Businesses were asked to respond anonymously in order to provide a more candid picture of health and safety conditions ("IOSH," 2012). The first results, drawn from 45 companies employing nearly 70,000 persons — approximately 14% of the total food and beverage workforce — reported two deaths, 137 major injuries from accidents, and 808 injuries requiring more than three days of lost work ("IOSH," 2012).

The project is a joint initiative of IOSH's Food and Drink Group, the Food and Drink Federation, and Dairy UK. According to the IOSH Food and Drink Group:

"The true picture of work-related injuries and ill-health across UK industry is hard to capture because reporting can be quite inaccurate. We only have very limited information on accidents in food and drink. We wanted to develop a reporting survey that gives us a fuller picture over the years, including a clearer view of ill-health — of which there are thousands of incidents costing millions of pounds annually." ("IOSH," 2012)

The challenge of collecting accurate, meaningfully aggregated data is common across all countries examined in this study. Even when businesses provide data to official agencies, there are strong disincentives against full disclosure. The research strongly recommends that companies be permitted to report anonymously when surveys will aggregate data and the purpose of collection is not remediation or penalty enforcement.

Significant differences exist in occupational safety and health policies and practices from country to country. For meaningful comparison, it is important to ensure that accident rates are expressed in equivalent units. The Labour Department of HKSAR reports accidents in occurrences per 1,000 workers, whereas the United States reports workplace accidents per 100 workers. Industry reporting definitions also vary: Hong Kong uses "catering industry" as the unit of analysis, while the United States uses "food and beverage industry." These distinctions can complicate direct comparisons; however, the differences in rates are so substantial that further disaggregation would likely sharpen rather than obscure the distinctions.

Data Analysis and Research Findings

Canada and the United States. Although Starbucks' safety practices and policies are not limited to North America, they are well established there. The approach is all-inclusive and entails rigorous training of all staff. Once every shift, a different employee conducts an on-site survey of the store, following an established checklist and submitting a report to the store manager immediately upon completion. This ensures that reporting is always current, that management and staff can take immediate corrective action, and that a running record of problems and resolutions is maintained as a resource for future training and the development of safety standards.

Australia. The National Research Centre for Occupational Health and Safety Regulation (NCOHSR) in Australia is focused on developing excellence in health and safety practices and regulation (Wood, et al., 2010). The Centre conducts empirical and policy-focused research in occupational health and safety regulation, with government agencies charged with reform initiatives as its primary audience (Wood, et al., 2010). The Centre's fundamental goal is to ensure the efficacy of regulation directed at improving health and safety outcomes in Australian workplaces, and it receives funding from WorkCover New South Wales, WorkSafe Victoria, and Workplace Health and Safety Queensland (Wood, et al., 2010).

Australia employs responsive regulation practices based on the belief that regulators should function in a manner that is responsive to the culture, context, and conduct of those they seek to regulate when determining the appropriate level of intervention (Wood, et al., 2010). The tagline for this approach is "soft words before hard words, and carrots before sticks" (Wood, et al., 2010). The model was developed by Braithwaite and Ayres in their book Responsive Regulation: Transcending the Deregulation Debate, and its ongoing development continues to be a collective effort of numerous scholars and institutions (Wood, et al., 2010).

The fundamental principles of responsive regulation are illustrated by a widely recognized figure known as the regulatory pyramid (Wood, et al., 2010). The base layer is labeled advisory and persuasive measures; the next layer up represents mild administrative sanctions; and the top layers represent more punitive sanctions believed to be sufficiently aversive to "halt the behaviors of the most determined offenders" (Wood, et al., 2010). The premise is that regulatory activity should primarily occur at the base of the pyramid and escalate only when less intrusive measures have proved ineffective, with de-escalation always remaining a live option (Wood, et al., 2010). Importantly, the presumptive preference for acting at the base of the pyramid is countered by what is called the "credible enforcement peak" — the requirement that follow-up enforcement actually occur if the model is to be effective (Wood, et al., 2010).

A constant comparison method is used to code and analyze the data in order to identify emerging themes (Dye, et al., 2000). Glaser and Strauss (as cited in Lincoln & Guba, 1985, p. 339) describe the constant comparison method as a four-stage procedure: (1) comparing data applicable to each category as categories emerge; (2) integrating categories and their properties to reduce the data set and data noise; (3) further delimiting the theory based on the reduced data set; and (4) writing the theory. The constant comparison method allows a social phenomenon to be simultaneously recorded, categorized, and compared across several categories (Goetz & LeCompte, 1981). Relationships among the data are identified and, since coding is integral to both data collection and data analysis, themes are continuously built. "As events are constantly compared with previous events, new topological dimensions, as well as new relationships, may be discovered" (Goetz & LeCompte, 1981, p. 58).

By creating categories during analysis, the researcher can cluster data into a taxonomy that reveals the conceptualization necessary for meaningful data analysis (Dey, 1993). Key patterns in data content are identified, categorized, and coded inductively so that both the taxonomy and its meanings "emerge out of the data rather than being imposed on them prior to data collection and analysis" (Patton, 1990, p. 390). The taxonomy is built by moving "back and forth between the logical construction and the actual data in search of meaningful patterns" (Patton, 1990, p. 411). Analysis proceeds in a bottom-up fashion, with codes derived from the data rather than from any pre-existing theory.

Many injuries and illnesses in the food and beverage industry are similar to those in other industries. However, a number of accident patterns are unique to food and beverage work and hold promise for meaningful, positive change. As discussed, the two primary accident types are: (a) sprains and strains, and (b) lacerations and burns. As the literature demonstrates, these accident types do not occur in isolation; they are dynamically and significantly related to fatigue, stress, and a critical disregard for established safety precautions.