Article Text

Download PDFPDF

Workplace
A prospective study of floor surface, shoes, floor cleaning and slipping in US limited-service restaurant workers
  1. Santosh K Verma1,2,3,
  2. Wen Ruey Chang4,
  3. Theodore K Courtney1,2,
  4. David A Lombardi1,2,
  5. Yueng-Hsiang Huang5,
  6. Melanye J Brennan1,
  7. Murray A Mittleman2,6,
  8. James H Ware7,
  9. Melissa J Perry2
  1. 1Center for Injury Epidemiology, Liberty Mutual Research Institute for Safety, Hopkinton, Massachusetts, USA
  2. 2Environmental and Occupational Medicine and Epidemiology Program, Department of Environmental Health, Harvard School of Public Health, Boston, Massachusetts, USA
  3. 3Department of Family Medicine and Community Health, University of Massachusetts Medical School, Worcester, Massachusetts, USA
  4. 4Center for Physical Ergonomics, Liberty Mutual Research Institute for Safety, Hopkinton, Massachusetts, USA
  5. 5Center for Behavioral Sciences, Liberty Mutual Research Institute for Safety, Hopkinton, Massachusetts, USA
  6. 6Department of Epidemiology, Harvard School of Public Health, Boston, Massachusetts, USA
  7. 7Department of Biostatistics, Harvard School of Public Health, Boston, Massachusetts, USA
  1. Correspondence to Dr Melissa J Perry, Department of Environmental Health, Harvard School of Public Health, 665 Huntington Avenue, Building 1 Room 1413, Boston, MA 02115, USA; mperry{at}hsph.harvard.edu

Abstract

Objectives Slips and falls are a leading cause of injury at work. Few studies, however, have systematically examined risk factors of slipping outside the laboratory environment. This study examined the association between floor surface characteristics, slip-resistant shoes, floor cleaning frequency and the risk of slipping in limited-service restaurant workers.

Methods 475 workers from 36 limited-service restaurants from three major chains in six states in the USA were recruited to participate in a prospective cohort study of workplace slipping. Kitchen floor surface roughness and coefficient of friction (COF) were measured in eight working areas and then averaged within each restaurant. The use of slip-resistant shoes was determined by examining the participant's shoes and noting the presence of a ‘slip-resistant’ marking on the sole. Restaurant managers reported the frequency of daily kitchen floor cleaning. Participants reported their slip experience and work hours weekly for up to 12 weeks. The survey materials were made available in three languages: English, Spanish and Portuguese. The associations between rate of slipping and risk factors were assessed using a multivariable negative binomial generalised estimating equation model.

Results The mean of individual slipping rate varied among the restaurants from 0.02 to 2.49 slips per 40 work hours. After adjusting for age, gender, BMI, education, primary language, job tenure and restaurant chain, the use of slip-resistant shoes was associated with a 54% reduction in the reported rate of slipping (95% CI 37% to 64%), and the rate of slipping decreased by 21% (95% CI 5% to 34%) for each 0.1 increase in the mean kitchen COF. Increasing floor cleaning frequency was significantly associated with a decreasing rate of slipping when considered in isolation but not after statistical adjustment for other factors.

Conclusion These results provide support for the use of slip-resistant shoes and measures to increase COF as preventive interventions to reduce slips, falls and injuries.

  • Slip
  • falls
  • injury
  • restaurants
  • slip-resistant shoes
  • floor cleaning
  • coefficient of friction
  • epidemiology
  • accidents

Statistics from Altmetric.com

Request Permissions

If you wish to reuse any or all of this article please use the link below which will take you to the Copyright Clearance Center’s RightsLink service. You will be able to get a quick price and instant permission to reuse the content in many different ways.

What this paper adds

  • Same-level falls, the majority of which are caused by slipping, are a leading cause of occupational injury in developed countries.

  • A considerable literature documents the association between coefficient of friction (COF) and slipping in laboratory settings; however, few epidemiological studies have systematically examined modifiable risk factors for slipping in actual work environments.

  • This is the first prospective cohort study to report a relationship between COF and rate of slipping in active work environments.

  • This is also the first field study to document a strong association between the use of slip-resistant shoes and reduced rate of slipping.

  • Workers reported a high rate of slipping at the baseline average COF above the recommended value of 0.5 by The US Occupational Safety and Health Administration.

  • The results provide quantitative evidence that suggests that organisations can materially reduce the frequency of slipping, an important precursor to same-level falls, by increasing COF and having workers wear slip-resistant footwear.

The proportion of same-level slip, trip and fall-related occupational injury ranges from 20 to 25% of all disabling occupational injuries in developed countries.1–4 A recent report from the US Bureau of Labour statistics projected a significant increase from 2004 to 2014 in the number of injuries due to falls and a decrease in the number of injuries from bodily exertion and reaction, and contact with objects, the other two major causes of occupational injuries.5 According to the Liberty Mutual Workplace Safety Index, the direct cost of STF-related injuries in the USA in the year 2007 was $7.7 billion, and the inflation-adjusted cost of injuries at work due to same-level falls increased 37% from 1998 to 2007, the greatest increase among major leading workplace injury causes.6 Older adults are at an increased risk of injuries, particularly fractures, due to falls on the same level.2 7 Since the population is ageing in the USA and other industrialised countries, and the proportion of older adults in the workforce is increasing,8 the incidence of injuries from slips and falls may continue to increase in the future.

Slipping contributes to between 40% and 85% of fall-related occupational injuries, depending on the work population studied.4 9–11 In addition, slipping that does not result in a fall can still result in an injury from striking an object or from a muscular strain.12 13 Hayes-Lundy et al reported that 11% of grease burns in fast-food restaurants could be attributed to slips.14

Slips and falls are the leading cause of injury among restaurant workers.15 16 In the USA, they account for one of every three disabling restaurant injuries.15 Among the top 10 highest-risk industrial classes for same-level falls in the State of Washington, full-service and limited-service restaurants had the highest frequency of same-level falls claims from 1999 to 2003.17 Food-service and drinking establishments are among the largest employers in the USA with about 6.4% of the total US workforce.18 Since they represent a large proportion of the total workforce, restaurants contribute significantly to the overall occupational injury burden.18 Restaurants contribute the third largest number of occupational injuries after specialty contractors and hospitals.19

The role of friction is crucial in understanding the causation of slips.20 When the available friction at the floor surface/shoe sole interface is not able to counteract shear forces generated by a foot during walking, a slip is likely to occur.21 The available friction at the floor surface/shoe sole interface depends on the floor surface characteristics, shoe sole characteristics and any contamination at the interface.

Coefficient of friction (COF) and floor surface roughness are two widely used measures of the slip resistance property of a floor.22 However, few studies have examined their association with slipping outside the laboratory environment, in a real-world setting. Similarly, many shoes are marketed as slip-resistant; however, to date, their effectiveness in reducing slips in actual working populations has not been documented scientifically.

Floor surface contamination is frequently reported when slips occur.15 23 Although regular effective floor cleaning to reduce floor contamination may reduce the incidence of slipping, the process of cleaning itself, such as when the floor is wet during and after mopping, can also present a slipping hazard. Studies of the effect of floor cleaning on the risk of slipping in the workplace are rare.24

Although slipping has been shown to be the major initiating event leading to same-level falls, few studies have systematically examined the risk factors for slipping in the workplace. Well-designed analytical studies are necessary to examine factors that contribute to risk of slipping and quantify their effects. In an effort to identify modifiable risk factors for slipping, a prospective cohort study was conducted in limited-service restaurant (also known as ‘fast-food,’ North American Industry Classification System Code 722211) workers. This study aimed to examine the effects of floor surface characteristics, slip-resistant shoes and floor cleaning frequency on the rate of slipping in fast-food restaurant workers.

Methods

This study was conducted in 36 limited-service restaurants in the states of Connecticut, Massachusetts, New York, Pennsylvania, Tennessee and Wisconsin in the USA. These restaurants belonged to three major chains and had similar main menu items: hamburgers and French-fried potatoes. Several approaches were used to recruit the restaurants for the study. These included approaching chains, stores or franchisees that had previously been receptive to research studies by the investigative team members, approaching restaurant trade associations, direct solicitation of stores or franchisees, and outreach via the loss control department of a large worker's compensation insurance company.

A total of 475 workers were recruited from these restaurants in the years 2007 and 2008. The study was approved by the Institutional Review Board of the Liberty Mutual Research Institute for Safety and the Office of Human Research Administration at the Harvard School of Public Health.

Enrolment procedure

Once permission to enrol a restaurant was received, members of the study team met onsite with the restaurant manager to explain the research study, administer a baseline manager survey and set up an appointment to enrol and survey the restaurant's employees. Restaurant managers were given fliers advertising the study with the date of the survey team's upcoming visit to post in their employee break area. On the scheduled date, participants were enrolled, and surveys were conducted in the restaurant. Restaurant workers not working on the day of enrolment were encouraged to come to the restaurant sometime during that day, with their work shoes, if they were interested in participating in the study. The survey materials were made available in three languages: English, Spanish and Portuguese. Enrolment procedures have been previously described in detail.23

Main exposures

Floor surface characteristics

Floor surface characteristics were measured on the day of the baseline survey at each restaurant. Floor surface characteristics were measured on the floor as found by the research team. However, any loose, gross contamination, such as food, was removed before making roughness and COF measurements. Floor surface roughness and COF were used to assess the slipperiness of the floor surface.

Surface roughness

Floor surface roughness was measured using a stylus profilometer in each restaurant kitchen in eight functional locations: Front Counter, Drive-Through, Sandwich Assembly, Fryer, Grill, Sink, Cooler/Freezer and Ice Machine.25 Three tiles were selected at each location. The first selected tile was directly in front of the working area, where the whole tile was at least 15 cm away from either the cooking equipment or any other vertical structures such as a wall, assembly table, etc. The other two tiles were selected on either side of the first tile, separated by one tile, in the line of traffic. Five roughness measurements were taken on each tile. Roughness parameters, Ra (the arithmetical average of the surface heights from the mean line) and Rpm (the average of maximum heights above the mean line) were calculated. These parameters are described in detail by Chang et al.26 Surface roughness parameters were averaged within the restaurants to calculate mean restaurant level measures for each restaurant. Floor surface roughness could not be measured in one of the restaurants due to a lack of cooperation from the manager on duty.

COF

COF was measured on the same tiles as the floor surface roughness using a Brungraber Mark II slip metre with Neolite test liners. Two measurements were taken on each tile in the direction of traffic and parallel to the equipment.27 Floor COF was measured according to the F-1677-96 standard method published by the American Society for Testing and Materials28 along with protocol refinements recommended by Chang.29 Measurement results were averaged at the restaurant level to calculate each restaurant's mean COF.

Slip-resistant shoes

Participants were asked to remove their right shoe for measurements and photographs. The brand of the shoe was noted, and photographs of the sole and top of the shoe were taken. Since clear classification criteria for slip-resistant shoes could not be found in the literature, shoes were classified as slip-resistant if the manufacturer claimed them to be by putting a ‘slip resistant’ label on the sole. Participants were asked how often they wore that pair of shoes at work. A total of 397 participants reported wearing the same shoes every day at work (94%).

Floor cleaning frequency

Restaurant managers provided information about usual frequency of daily kitchen floor cleaning at baseline. ‘Kitchen’ was defined as all food preparation, service and storage areas between the front counter and the back door.

Outcome

Slipping

A study team member carefully explained the definition of a slip to the study participants by stating that, ‘A slip is simply a loss of traction of your foot—you can slip without falling.’ After completing the baseline survey, participants were asked to report their slip experience every week for the following 12 weeks. They were not required to keep a daily log and had to remember the incidents until the subsequent reporting day. Each week participants reported the number of slips and the number of hours they worked in the previous week. The rate of slipping was the primary outcome of interest (total number of slips reported/total number of hours worked during follow-up).

If participants experienced one or more slips, they also reported whether the slip resulted in a fall or an injury for up to four slips. A slip was categorised as a major slip if it resulted in a fall and/or an injury.

Participants were given a choice of reporting their weekly experience by telephone using the interactive voice response system, by an internet-based survey or by filling out and mailing printed survey forms. Figure 1 presents a flow diagram of cohort recruitment and loss to follow-up.

Figure 1

Flow diagram of cohort recruitment and loss to follow-up.

Worker and job characteristics

Demographic information about each participant was collected including, age, gender, education and primary language. Information about weight and height was used to calculate the body mass index. Questions about job characteristics included job tenure, primary locations of work and number of hours worked per week.

Courtney et al reported increasing age and female gender to be associated with reduced risk of slipping,30 and Fjeldstad et al found a higher prevalence of falls in obese people.31 Job tenure, ethnicity and education level have been shown to be associated with workplace injuries.32–34 However, their effects on slips and falls have been less well examined. Work experience, cultural differences and education level may affect the type of job assigned to a worker and may also affect reporting of slips. Therefore, to account for the effects of these potential confounders, we controlled for age, gender, BMI, job tenure, primary language and education level in the multivariable model.

Data analysis

Restaurants recruited in the study are clustered within chains, and workers are clustered within restaurants. To account for clustering of participants within restaurants, a multivariable Poisson generalised estimating equation model with compound symmetry covariance structure35 36 was used initially to assess the association between the rate of self-reported slipping, and floor surface characteristics, slip-resistant shoes and frequency of floor cleaning. As there were only three chains, two indicator variables for chains were included in the regression model to account for clustering of restaurants within chains. The Poisson model showed significant overdispersion, and therefore, negative binomial generalised estimating equation models were used to estimate crude and adjusted rate ratios for slipping. If main effects were not significant at a 0.05 level in the multivariable model, they were dropped from the model. However, all the potential confounders were selected a priori and were included in the multivariable model regardless of the significance level. Two-way interactions were evaluated on both the multiplicative and additive scales.37 Rate ratios for the main effects and their 95% CI based on robust SE estimates are presented.

In addition, the associations between the main effects and the rate of major slips were also examined. The proportion of major slips among all the slips was compared among those wearing slip-resistant shoes and those who were not, to examine whether the threshold for detecting slips was different in the two groups. All statistical analyses were performed using the SAS system version 9.1.

Results

Information about the total number of employees was available for 35 out of 36 restaurants. These 35 restaurants employed 1163 employees, of which 466 participated in the study (40%). On average, 9.8 weeks of data were collected for each worker (median=11, total 4317, figure 1), and 422 reported at least 1 week of follow-up data. Table 1 presents demographic information about participants who provided at least 1 week of follow-up data and those who did not. Participants with no follow-up data were younger, more likely to be male, less likely to be English-speaking and less likely to have some college education than participants with follow-up data.

Table 1

Demographic characteristics of participants with no follow-up, at least 1 week of follow-up and with complete follow-up

The mean age of participants who reported at least 1 week of follow-up data was 31.5 years (range=15–78), and 22% were 19 years old or younger. More than two-thirds of the participants were female (68%). The primary language of 89% of participants was English, 9% Spanish and 2% Portuguese. Participants reported working an average of 34 h per week, and the mean job tenure in their restaurant was 37 months (median=18). Most of the participants reported working at multiple locations within the restaurant; on average, they worked in five of the eight kitchen areas.

The mean Ra and Rpm had a positively skewed distribution, and the mean COF had a slightly negatively skewed distribution (table 2). The mean COF ranged from 0.45 to 0.86, a difference of about 0.4 between the restaurant with the highest COF and the restaurant with lowest COF. Twenty participants were not wearing their usual work shoes at the time of survey and were categorised as non slip-resistant shoe users (default category). A total of 286 participants were wearing slip-resistant shoes (68%, table 2). A floor-cleaning frequency of twice daily was most frequently reported (11/36, 31%).

Table 2

Distribution of floor surface roughness (Ra and Rpm), coefficient of friction, floor cleaning frequency, use of slip-resistant shoes, rate of slipping and rate of major slipping

Participants with at least 1 week of follow-up data were included in the analysis (n=422). The total number of slips reported during the follow-up was 1168, and the total number of hours worked was 105 240 h, resulting in an overall rate of slipping of 0.44 slips per 40 work hours. The mean of individual slipping rate was 0.69 slips per 40 work hours (median=0.14, range=0.00–22.22) or 34.5 slips per full-time employee per year (table 2). The mean of individual slipping rate varied among the restaurants from 0.02 to 2.49 slips per 40 work hours. Of the slips for which fall and injury information was reported (n=1099), 17 resulted in a fall and an injury, 43 resulted in a fall without an injury, and 23 resulted in an injury without a fall. These 83 slips were classified as major slips (7.6%, 83/1099). The mean rate of major slips was 0.052 per 40 work hours or 2.6 major slips per full-time employee per year (table 2). For 49 participants, at least one slip was classified as a major slip (49/422, 12%).

Use of slip-resistant shoes, mean COF and floor-cleaning frequency were significantly associated with rate of slipping in the unadjusted regression models (table 3). In the multivariable model that adjusted for age, gender, BMI, education, primary language, job tenure and restaurant chain, slip-resistant shoes and mean COF were significantly associated with rate of slipping. The use of slip-resistant shoes was associated with a 54% reduction in the reported rate of slipping (95% CI 37% to 64%), and the rate of slipping decreased by 21% (95% CI 5% to 34%) for each 0.1 increase in mean COF. The interaction effect of slip-resistant shoes and mean COF was not significant in either the multiplicative scale or the additive scale. Floor-cleaning frequency and roughness parameters were not significant in the multivariable model and were dropped from the model.

Table 3

Rate ratios (RR) and their 95% CIs from univariate and multivariable regression models and sensitivity analysis modelling the rate of major slips

In the analysis examining associations with major slips, both slip-resistant shoes and mean COF retained their protective effects. The use of slip-resistant shoes was associated with a 40% reduction in the reported rate of major slipping that did not reach statistical significance (rate ratio (RR)=0.60, 95% CI 0.24 to 1.48), and the rate of major slipping decreased by 54% (RR=0.46, 95% CI 0.31 to 0.66) for each 0.1 increase in the mean COF. About 9% (8.9%) of all slips were major slips among those wearing slip-resistant shoes and 6.4% among those not wearing slip-resistant shoes. The difference between the two groups was not statistically significant.

Discussion

Although same-level falls contribute to a significant proportion of occupational injuries, and slipping is a major initiating event leading to same-level falls, few studies have examined the risk factors for slipping outside the laboratory environment. This study examined the effects of floor surface characteristics, slip-resistant shoes and floor-cleaning frequency on the rate of slipping in limited-service restaurant workers. The results indicated that increasing mean COF at the restaurant level and use of slip-resistant shoes were associated with a reduced rate of slipping.

Many laboratory studies have reported a higher COF to be associated with a decreased risk of slipping.20 38–40 Their suggestions regarding safe COF for walking ranged from 0.2 to 0.6. However, the effects of COF outside the laboratory environment have rarely been examined. The US Occupational Safety and Health Administration (OSHA) does not have any standards that mandate a particular COF for walking/working surfaces. As part of its non-mandatory guidelines, OSHA recommends a COF of 0.5 as a guide to achieve proper slip resistance and mentions that a higher COF may be required for certain tasks such as carrying, pushing and walking on a ramp.41 The Americans with Disabilities Act recommends a COF of 0.6 or greater on flat surfaces.42

This study, similar to the cross-sectional study by Courtney et al, found a high rate of slipping in restaurants with an overall average COF of 0.67 and restaurant level average COF ranging from 0.45 to 0.86.30 Moreover, an increase in the average COF was also associated with a decrease in the rate of slipping in restaurants with an average COF within or above the recommendations regarding slip-resistant floor surfaces. Courtney et al also found a decreased risk of slipping with increasing COF in this range.30 Both of the findings suggest that 0.5 may not be an absolute cut-off COF value above which floors can be considered safe for walking. Higher COF may provide additional safety against slipping in an environment with a high prevalence of floor contamination.

Different shoe sole materials and tread patterns have different slip-resistant properties,43 and slip resistance can vary widely among common footwear materials. Menz et al found that men's Oxford shoes exhibited higher average COF values than women's fashion shoes; however, none of these shoes could be considered safe on wet surfaces.44 While there are many ‘slip-resistant’ shoes on the market, very few published studies have examined the association between slip-resistant shoes and risk of slipping. The UK Health and Safety Laboratory recently published a report on the slip potential of common ‘occupational footwear’ based on ramp tests in the laboratory environment.45 They reported that among shoes marketed as ‘slip-resistant,’ 33% posed low, 33% posed moderate and 33% posed high slip potential on quarry tiles, a common type of floor surface found in the limited-service restaurants, with glycerol contamination. They also found that almost all shoes that are not marketed as slip-resistant posed high slip potential in this environment. To date, only Bell et al have reported an effective slip, trip and fall prevention intervention programme in hospital employees where slip-resistant shoes were a part of the comprehensive approach.46 Courtney et al did not find a significant association between slip-resistant shoes and risk of slipping.30 They classified shoes as slip-resistant based on tread pattern, which they suggested may have led to misclassification. The present study found that slip-resistant shoes were associated with more than 50% reduction in the rate of slipping, suggesting that they may have substantial potential for slip and fall prevention.

Although we found slip-resistant shoes to be effective in reducing the rate of slipping, there are no definite criteria for labelling shoes as slip-resistant, and all shoes marketed as ‘slip-resistant’ may not provide similar slip resistance. Development of a reliable and valid test to measure the slip-resistance properties of footwear heel and sole and the specific slip-resistance required to label a shoe as ‘slip-resistant’ will help employers and safety professionals in choosing appropriate shoes for the work environment and including slip-resistant shoes as a part of an overall slip and fall prevention programme.

Very few studies have examined the effects of floor cleaning on the risk of slipping. Leclercq et al reported that cleaning a soiled surface does not immediately lead to an increase in its slip resistance.24 In this study, we found that increasing floor-cleaning frequency was associated with a reduced rate of slipping in the unadjusted analysis. However, this association was not significant in the multivariable analysis. Although effective floor cleaning is important for maintaining the slip resistance of the floor surface, ineffective floor cleaning and wet floor surfaces after mopping may increase the risk of slipping. More studies are needed to understand how floor-cleaning practices, such as adherence to floor-cleaning protocol, employee training and floor-cleaning equipment affect the risk of slipping in limited-service restaurants and other environments with high hazards of slips and falls.

Although many studies have reported roughness to be associated with the slip-resistance property of nominally flat floor surfaces,47 48 we did not find floor surface roughness measures to be associated with a reduced rate of slipping. Many restaurants in this study had floors with macroscopic patterns such as grit embedded in the tiles or tiles with raised patterns (figure 2). In the presence of these macroscopic patterns, the effect of roughness measured in micrometres on the slip-resistance property of the floor, and thus rate of slipping, may have been less important.

Figure 2

Tiles with raised pattern and grit.

Limitations and strengths

This is the first prospective study to examine the association between floor surface characteristics, use of slip-resistant shoes, floor-cleaning frequency and risk of slipping. Restaurants belonging to three major chains and from six different states participated in the study. The survey material was made available in three different languages, thus increasing the generalisability of the study findings. However, since 34 of the 36 restaurants were owned by large employers, some of the findings may not be generalisable to small employers. In particular, the distribution of COF and use of slip-resistant shoes may not be similar.

Individuals vary in their detection of minor slips.49 Random variation in slip detection is likely to bias the results towards null. However, a systematic difference in slip-detection threshold among those who wore slip-resistant shoes and those who did not may bias the results away from null and may be responsible for some of the observed findings. Participants who believed in the effectiveness of slip-resistant shoes may be more likely to use them and, because of this belief, may ignore minor slips, thus reporting a lower number of slips. Major slips would be less likely to be affected by the difference in threshold for reporting slips. If slip-resistant shoe users were more likely to ignore minor slips than non slip-resistant shoe users, the ratio of major to minor slips would be higher among slip-resistant shoe users. We did not find any significant difference in the proportion of major slips among those who were wearing slip-resistant shoes and those who were not. Therefore, it is less likely that the entire effect of slip-resistant shoes could be explained by the difference in threshold for detecting and reporting slips.

We have used slips as an outcome, as its high incidence makes the examination of the effects of risk factors using a 12-week prospective study design feasible. Although slipping is an important precursor to fall-related and other injuries, factors affecting the rate of slipping may or may not be associated with risk of falls or related injuries. When outcome was limited to slips that resulted in a fall or an injury, we found that both COF and slip-resistant shoes retained their protective effects.

Conclusion

This is one of the first epidemiological studies to examine the effects of floor surface characteristics, slip-resistant shoes and floor-cleaning frequency on rate of slipping. Results suggest the use of slip-resistant shoes and measures to increase COF as preventive interventions to reduce slips and falls, and the injuries resulting from them. More studies are needed to examine how floor-cleaning methods and macroscopic patterns on floor surfaces affect the rate of slipping.

References

Footnotes

  • See Commentary, p 238

  • Competing interests None.

  • Ethics approval Ethics approval was provided by the Institutional Review Board of the Liberty Mutual Research Institute for Safety and the Office of Human Research Administration at the Harvard School of Public Health.

  • Provenance and peer review Not commissioned; externally peer reviewed.

Linked Articles