Assessing non-fatal intentional poisoning incidents as a national sentinel indicator for suicide mortality trends in Malaysia (2006 to 2021): a mechanistic and economic impact estimation

Abstract Background Suicide is a major public health concern in Malaysia. Relying solely on official mortality statistics often introduces significant reporting delays, hindering immediate public health response. In Malaysia, while the Suicide Mortality Rate (SMR) is the paramount metric for national health planning, official statistics suffer from a significant 1 to 2 year reporting lag, rendering prevention reactive. While toxicovigilance data does not capture all suicide modalities, this study aimed to assess the utility of non-fatal intentional poisoning data previously collected in near real-time by the National Poison Centre (NPC-USM) to be used as a sentinel indicator for tracking national distress and methodological shifts in suicide. Methods A 16 year ecological and time-series analysis was conducted using data from the National Poison Centre (NPC) and the Department of Statistics Malaysia (DOSM). Spearman’s correlation, Durbin-Watson statistics, and an Autoregressive Integrated Moving Average with Exogenous Predictor [ARIMAX (1,0,0)] model were utilized to address longitudinal autocorrelation. An economic impact estimation was formulated utilizing the official Malaysian Value of Statistical Life (VSL) proxy. Results Time-series analysis revealed a complex decoupling between non-fatal poisoning and fatal suicide outcomes. Overall correlation was weak (r s = 0.182), primarily due to a severe divergence occurring in the latter half of the study period. Between 2019 and the 2021 pandemic peak, non-fatal intentional poisoning attempts reported to the NPC plummeted by 29.2% (1,212 to 857 cases), whereas official DOSM suicide fatalities surged by 85.7%. The ARIMAX model (AIC = -1.12) and Durbin-Watson diagnostics (0.61) support this structural shift. Analysis of age group distribution revealed the 20 to 39 years demographic remains the strongest anchor (r s = 0.826). Conclusion The drastic decoupling of non-fatal poisoning and official mortality during the pandemic supports the hypothesis of an aggressive method substitution toward higher-lethality means. The NPC data functions as a high-sensitivity sentinel not for parallel overall mortality, but for tracking acute behavioural shifts in method availability and lethality intent. Integrating real-time toxicovigilance into national psychiatric surveillance architectures presents a potentially cost-effective paradigm for proactive public health intervention.