Underestimate of deaths due to specific opioids in England and Wales for a decade

IntroductionOpioids are involved in combination with other substances in most of the drug-related deaths in Europe, with instance of opioid-related death increasing over the past decade. For a death to be associated with the use of a specific opioid in England and Wales it must be specifically named as a cause of death. However, with increasing polypharmacy, ambiguous causes such as ‘multi-drug toxicity’ or ‘polydrug use’ are being increasingly used. This study aimed to understand how such ambiguous causes are causing underestimation of deaths due to specific opioids in England and Wales.  MethodsData from Office for National Statistics (ONS) and the National Programme on Substance Abuse Deaths(NPSAD) database were used. ONS receives 100% of drug-related deaths but can only attribute a death to a certain drug if it is specifically named on the death certificate. In contrast, NPSAD receives drug-related death reports from approximately 90% of coronial jurisdictions in England and Wales, but as NPSAD has access to full toxicology reports it is able to identify the implicated drugs in cases with ambiguous causes of death. The percentage of cases where implicated drugs were specifically named and where they were ambiguously implicated were calculated from NPSAD cases, with these proportions then extrapolated toONS figures to estimate the true number of deaths from specific opioids. The percentage of deaths due to specific opioids ‘missed’ by ONS was plotted against the median polypharmacy rate determined from toxicology reports submitted to the NPSAD with the correlation coefficients calculated. ResultsFrom triangulating with NPSAD records, ONS has consistently underestimated the number of deaths due to specific opioids (Table 1). The year-on-year underestimates significantly correlated with increasing rates of polypharmacy, with correlation coefficients for heroin/morphine (r=-0.843), methadone (r=-0.914) and dihydrocodeine (r=-0.773) displaying relatively strong negative correlations, tramadol (r=-0.636) showing a moderate negative correlation, and fentanyl (r=-0.433) a weak negative correlation. 

Opioid
ONS Reported Deaths
Actual Estimated Deaths
% Underestimate

Codeine
1674
3366
50.27

Dihydrocodeine
1106
1793
38.32

Methadone
5086
8006
36.47

Tramadol
2244
3467
35.28

Heroin/Morphine
12073
17797
32.16

Fentanyl
549
748
26.60

Table 1. The total deaths attributed to specific opioid from ONS and the estimated deaths extrapolated from NPSAD, 2011-2021. 
ConclusionONS is underestimating the number of deaths due to specific opioids. This means the current opioid-related epidemic in England & Wales is worse than it appears. It is important to secure accurate and informative figures for drug-related deaths to fully understand the true situation and for proper resource allocation to ensure harm reduction initiatives are appropriately funded to affect change.