Explore the vast range of titles available.

ObjectiveVast amounts of injury narratives are collected daily and are available electronically in real time and have great potential for use in injury surveillance and evaluation. Machine learning algorithms have been developed to assist in identifying cases and classifying mechanisms leading to injury in a much timelier manner than is possible when relying on manual coding of narratives. The aim of this paper is to describe the background, growth, value, challenges and future directions of machine learning as applied to injury surveillance.MethodsThis paper reviews key aspects of machine learning using injury narratives, providing a case study to demonstrate an application to an established human-machine learning approach.ResultsThe range of applications and utility of narrative text has increased greatly with advancements in computing techniques over time. Practical and feasible methods exist for semiautomatic classification of injury narratives which are accurate, efficient and meaningful. The human-machine learning approach described in the case study achieved high sensitivity and PPV and reduced the need for human coding to less than a third of cases in one large occupational injury database.ConclusionsThe last 20 years have seen a dramatic change in the potential for technological advancements in injury surveillance. Machine learning of ‘big injury narrative data’ opens up many possibilities for expanded sources of data which can provide more comprehensive, ongoing and timely surveillance to inform future injury prevention policy and practice.

Introduction of engineered nanoparticles (ENPs) into traditional surface coatings (e.g., paints, lacquers, fillers) may result in new exposures to both workers and consumers and possibly also a new risk to their health. During finishing and renovation, such products may also be a substantial source of exposure to ENPs or aggregates thereof. This study investigates the particle size distributions (5.6 nm–19.8  μ m) and the total number of dust particles generated during sanding of ENP-doped paints, lacquers, and fillers as compared to their conventional counterparts. In all products, the dust emissions from sanding were found to consist of five size modes: three modes under 1  μ m and two modes around 1 and 2  μ m. Corrected for the emission from the sanding machine, the sanding dust, was dominated by 100–300 nm size particles, whereas the mass and surface area spectra were dominated by the micrometer modes. Adding ENPs to the studied products only vaguely affected the geometric mean diameters of the particle modes in the sanding dust when compared to their reference products. However, we observed considerable differences in the number concentrations in the different size modes, but still without revealing a clear effect of ENPs on dust emissions from sanding.

PurposeThis study aims to identify the role of age structure in occupational choices and the classification of the occupations based on the age structure of individuals in the Indian labour market.Design/methodology/approachThis study uses the first Periodic Labour Force Survey, 2017–18. The occupational classifications are based on the standardised scores for age groups and their occupations. Further, a multinomial logistic regression model has been used to estimate social and economic factors in determining the age-based occupational classifications.FindingsThe authors found age structure an essential factor in determining occupational choices. Hence, occupations in the Indian labour market have been grouped into seven categories, accordingly. In addition, social and economic factors of individuals and households do have a significant influence on the selection of age-based occupational classifications.Research limitations/implicationsThe study is limited to the occupational classification based on the age structure of individuals without any industry effects. The findings suggest that policymakers must adopt occupation-specific policies considering the age structure of individuals.Originality/valueEarlier studies are limited to the dynamics of age either on the basis of specific age groups (younger or older) or on the industrial classification in a disaggregated way. They also lack a rich approach in analysing the occupational classification considering age structure, especially in the Indian labour market. The study adds value when the role of age structure is identified in occupational choices in the Indian labour market, and hence, a novel classification of occupations into seven categories is proposed.

Studies examining workers’ diet according to smaller occupational groups within “large occupational categories” are sparse. The aim of this study was to examine the potential differences in workers’ diets based on the classification of workers into smaller occupational groups that comprise “large occupational categories”. The subjects of this study were working fathers who had participated in the Japan Environment and Children’s Study (N = 38,656). Energy and nutrient intake were calculated based on data collected from the Food Frequency Questionnaire. Occupations were classified according to the Japanese Standard Occupational Classification. Logistic regression analyses were performed to examine the adherence to current dietary recommendations within smaller occupational groups. In particular, significant differences were observed among the categorical groups of “professional and engineering workers”, “service workers”, and “agricultural, forestry, and fishery workers”. In “professional and engineering workers”, teachers showed higher odds of adherence to calcium intake recommendations compared with nurses (OR, 2.54; 95% CI, 2.02–3.14; p < 0.001). In “agricultural, forestry, and fishery workers”, agriculture workers showed higher odds of adherence to calcium (OR, 2.15; 95% CI, 1.46–3.15; p < 0.001) and vitamin C (OR 1.90, 95% CI 1.31–2.74, p = 0.001) intake recommendations compared with forestry and fishery workers. These findings may be beneficial from a research perspective as well as in the development of more effective techniques to improve workers’ diet and health.

Slips, trips, and falls are leading causes of musculoskeletal injuries in firefighters. Researchers have hypothesized that heat stress is the major contributing factor to these fireground injuries. To examine the effect of environmental conditions, including hot and ambient temperatures, and exercise on functional and physiological outcome measures, including balance, rectal temperature, and perceived exertion. Randomized controlled clinical trial. Laboratory environmental chamber. A total of 13 healthy, active career firefighters (age = 26 ± 6 years [range = 19-35 years], height = 178.61 ± 4.93 cm, mass = 86.56 ± 16.13 kg). Independent variables consisted of 3 conditions (exercise in heat [37.41°C], standing in heat [37.56°C], and exercise in ambient temperature [14.24°C]) and 3 data-collection times (preintervention, postintervention, and postrecovery). Each condition was separated from the others by at least 1 week and lasted a maximum of 40 minutes or until the participant reached volitional fatigue or a rectal temperature of 40.0°C. Firefighting-specific functional balance performance index, rectal temperature, and rating of perceived exertion. Exercise in the heat decreased functional balance, increased rectal temperature, and altered the perception of exertion compared with the other intervention conditions. A bout of exercise in a hot, humid environment increased rectal temperature in a similar way to that reported in the physically active population and negatively affected measures of functional balance. Rather than independently affecting balance, the factors of exercise and heat stress appeared to combine, leading to an increased likelihood of slips, trips, and falls.

BackgroundBayesian methods show promise for classifying injury narratives from large administrative datasets into cause groups. This study examined a combined approach where two Bayesian models (Fuzzy and Naïve) were used to either classify a narrative or select it for manual review.MethodsInjury narratives were extracted from claims filed with a worker's compensation insurance provider between January 2002 and December 2004. Narratives were separated into a training set (n=11,000) and prediction set (n=3,000). Expert coders assigned two-digit Bureau of Labor Statistics Occupational Injury and Illness Classification event codes to each narrative. Fuzzy and Naïve Bayesian models were developed using manually classified cases in the training set. Two semi-automatic machine coding strategies were evaluated. The first strategy assigned cases for manual review if the Fuzzy and Naïve models disagreed on the classification. The second strategy selected additional cases for manual review from the Agree dataset using prediction strength to reach a level of 50% computer coding and 50% manual coding.ResultsWhen agreement alone was used as the filtering strategy, the majority were coded by the computer (n=1,928, 64%) leaving 36% for manual review. The overall combined (human plus computer) sensitivity was 0.90 and positive predictive value (PPV) was >0.90 for 11 of 18 2-digit event categories. Implementing the 2nd strategy improved results with an overall sensitivity of 0.95 and PPV >0.90 for 17 of 18 categories.ConclusionsA combined Naïve-Fuzzy Bayesian approach can classify some narratives with high accuracy and identify others most beneficial for manual review, reducing the burden on human coders.

Background Traditionally, time-use data have been used to inform a broad range of economic and sociological research topics. One of the new areas in time-use research is the study of physical activity (PA) and physical activity energy expenditure (PAEE). Time-use data can be used to study PAEE by assigning MET values to daily activities using the Ainsworth Compendium of Physical Activities. Although most diarists record their daily activities accurately and in detail, they are only required to record their paid working hours, not the job-specific tasks they undertake. This makes it difficult to assign MET values to paid work episodes. Methods In this methodological paper, we explain how we addressed this problem by using the detailed information about respondents’ occupational status included in time-use survey household and individual questionnaires. We used the 2008 ISCO manual, a lexicon of the International Labour Organization of occupational titles and their related job-specific tasks. We first assigned a MET value to job-specific tasks using the Ainsworth compendium (2011) then calculated MET values for each of the 436 occupations in the ISCO-08 manual by averaging all job-specific MET values for each occupation. Results The ISCO-08 Major Groups of ‘elementary occupations’ and ‘craft and related trades workers’ are associated with high PAEE variation in terms of their job-specific MET values and together represented 21.6% of the Belgian working population in 2013. We recommend that these occupational categories should be prioritised for further in-depth research into occupational activity (OA). Conclusions We developed a clear and replicable procedure to calculate occupational activity for all ISCO-08 occupations. All of our calculations are attached to this manuscript which other researchers may use, replicate and refine.

OBJECTIVE: Occupational differences in COVID-19 are well documented, but the empirical evidence on potential reasons for these differences remains limited. Possible reasons include pre-existing health conditions. This study investigated occupational differences in COVID-19 disease severity and whether they can be attributed to pre-existing health conditions. METHODS: Our study used German health insurance data covering 3.17 million insured individuals (age 18–67 years), with details on COVID-19-related hospitalization and mortality in 2020 and 2021, information on occupation (regrouped into four classifications) and pre-existing health conditions (divided into seven disease groups). In addition to descriptive statistics, we estimated multivariable Cox regression models with varying sets of adjustments. RESULTS: We found clear occupational differences in COVID-19 hospitalization and mortality, with the highest risks for the production sector (especially manufacturing), commercial services (especially cleaning) and for low-skilled occupations. These findings persisted after adjusting for age, sex, and region, and also after mutual adjustment for other occupational classifications. We also found some evidence that the association between occupation and disease severity was partly explained by pre-existing conditions, especially in the case of low skill levels. CONCLUSIONS: Our findings provide support for occupational differences in COVID-19, where the occupational classifications under study were independently related to risk differences (eg, skill-level and job sector). Furthermore, we provide empirical evidence that differences by occupational skill levels are partly due to pre-existing conditions. This finding suggests that occupational inequalities in health increased during the pandemic, with those with poorer health who worked in disadvantaged occupations also being more likely to experience severe COVID-19 outcomes.