Since 2010, paracetamol has been the most common agent used in Hong Kong for deliberate self-harm by overdose and poisoning.1 It is readily available over-the-counter and is commonly prescribed by doctors. There are more than 900 registered pharmaceuticals that contain paracetamol in Hong Kong. Paracetamol overdose can result in delayed, sometimes life-threatening, liver injury and dosedependent damage. N-acetylcysteine (NAC) is well known to be an effective antidote that can prevent liver injury if administered in time. The decision to give NAC can be facilitated by plotting the serum paracetamol concentration against time since ingestion on the Rumack-Matthew nomogram. In the article written by Chan et al, 4 the 150-treatment line has been evaluated and identified a failure rate of 0.45%. All four index patients developed chemical hepatitis that responded to supportive treatment. The incidence of 150-treatment line failures in the US has been reported as 1% to 3% and thought to be predominantly due to inaccurate ingestion history.5 Looking closely at the four cases presented in Chan et al's study, 4 two of them presented late, and in most cases there was an apparent discrepancy between the dose taken and the achieved paracetamol level. Similar to the US experience, an inaccurate ingestion history might explain treatment-line failure for some of these cases. Further evidence from more robust studies is needed before a recommendation can be made to lower the treatment threshold to the UK standard. Obtaining an accurate history from patients who deliberately self-harm is known to be difficult. Patients may be unwilling or unable to provide accurate information to the clinician. According to the author's own experience in managing such patients, history taking must be done tactfully and sometimes repeatedly from different sources of information. An astute physician should make the decision to give NAC after analysing all the available evidence including the best-gathered history, the clinical presentation, and the remaining treatmenttime window, together with the serum paracetamol level. Laboratory tests may help but can never replace clinical skill, clinical judgement, and experience in patient management.Previously, the responsibility for managing such time-critical overdosed patients was often borne by interns and junior residents. The quality of care provided may not have been optimal. Over the past 10 years emergency physicians and trainees have received intensive training in the management of toxicology cases based on updated evidence and standards. In addition, groups of interested emergency physicians have formed toxicology teams to oversee and support the management of poisoning patients in individual hospitals. This model of care improves patient outcome and shortens the length of stay for medical treatment.6,7 Such improvements might explain the observed good outcome for Chan et al's cohort 4 of patients with paracetamol overdose.