Answers are grounded in this paper and cite their sources.
Background Palm oil production combines manual plantation work with mechanised mill processing, creating heterogeneous occupational hazards. This secondary analysis of a previously reported sample adds recorded injury profiles by assigned work activity and a temporally cautious exploratory model of factors associated with self-reported occupational accidents. Methods In this cross-sectional analysis, 200 of 358 workers at one Indonesian company were selected by proportional stratified random sampling; 191 were analysed after nine were excluded because direct observation of personal protective equipment (PPE) use could not be completed. The outcome was a self-reported accident in the preceding 12 months. The operational definition permitted commuting accidents, but interview clarification indicated that none occurred; event location was not retained as an analytic variable. PPE supervision was observed on one working day, after some reported events may have occurred, and was treated as a contemporaneous indicator. Prior accident history was excluded from the primary logistic model because its timing could not be established. Variable selection was repeated within 1,000 bootstrap resamples, and adjusted prevalence ratios (aPRs) used modified Poisson regression. Results Accidents were reported by 80/191 workers (41.9%). Prevalence among workers assigned to each activity ranged from 14.3% in security to 60.0% in maintenance. Open wounds/lacerations (52.5%), upper-extremity injuries (47.5%), and minor injuries (62.5%) predominated; all four recorded burns occurred in processing workers. Tenure < 5 years (aPR 2.13, 95% CI 1.61–2.83; adjusted odds ratio [aOR] 9.86, 95% CI 3.35–29.01) and poor PPE supervision (aPR 1.89, 95% CI 1.30–2.77; aOR 4.10, 95% CI 1.94–8.69) were associated with reported accidents, whereas the derived high-risk work-unit category was not (aPR 1.42, 95% CI 0.81–2.49). The optimism-corrected area under the curve was 0.735 and the calibration slope 0.755. Tenure was retained in 98.1% of bootstrap selections and PPE supervision in 69.2%. Conclusions Reported accident prevalence differed among workers assigned to different activities; injury distributions are descriptive because subgroup counts were small. Shorter tenure and contemporaneously observed poor PPE supervision were associated with reported accidents, without establishing temporal or causal relationships. The exploratory model showed moderate discrimination with evidence of overfitting and requires external and prospective evaluation before any operational use.
No takes yet. Share an insight, caveat, or question.
Waren et al. (2026) studied this question.
Synapse has enriched 10 related papers. Consider them for comparative context: