Report: Drug overdoses in Aotearoa 2026
In 2025, 169 people died from preventable overdose. Behind every statistic in this report is immeasurable pain and loss. Real people who have left behind unfulfilled dreams, grieving loved ones, and broken hearts
Accidental overdose deaths are preventable. But for the last decade we have failed to reduce them.
On average we lose three people every week.
In 2025, accidental overdose meant
6,659 years of potential life lost*
over 39 years per person.
We need a change of approach to turn things around.
A comprehensive, health-based approach to drugs that includes specialist overdose prevention services, an expansion of harm reduction approaches, access to naloxone, and better preparedness for rapidly changing drug supplies.
You can find out more about how this can be achieved in our report Safer drug laws for Aotearoa New Zealand: Evidence to inform regulatory change.
There is currently no national standard for tracking overdose numbers and no national body tasked with counting them. Data uncertainties are a big part of the gap in our overdose response, and improving overdose surveillance should be a high-priority issue for our health system.
*Calculated as average years remaining based on life expectancy for a male at average overdose death age (46.2 years; Stats NZ), multiplied by the provisional total number of fatal accidental overdose deaths in 2025.
Acknowledgements
We want to acknowledge the teams that assisted us in accessing the data, including the Office of the Chief Coroner (Ministry of Justice), and National Collections Data Services (Te Whatu Ora).
In particular, we want to acknowledge Chris Lewis from the National Collections Data Services team for his continued guidance.
Jump to:
Overdose deaths in Aotearoa
Coronial data
This section examines coronial data on fatal, accidental overdoses in New Zealand between 2016 and 2025, and Te Whatu Ora mortality data covering the period of 2011 to 2022.
We provide further information about why and how we used the two datasets in this report below.
This data includes accidental drug overdose deaths and excludes intentional deaths. A small proportion of cases covered are deaths where intent was undetermined. We have excluded other drug-related death causes, such as communicable diseases, short- or long-term health complications, and chronic health conditions that may occur as a result of drug use.
The number of deaths in 2025 appears to have increased in comparison to previous years. That said, we must interpret that increase with caution considering that nearly half of these are still provisional (active) cases.*
1,440 people died of accidental drug overdose between 2016 and 2025.
Accidental overdose deaths
2011–2025
*Coronial inquiries take time
Recent cases are more likely to be under active investigation. This means that the total numbers we present for recent years are provisional and are subject to change.
For 2025, nearly half of the cases we analysed were still considered active cases under coronial inquiry at the time of publishing. We have marked this year with an asterisk.
The graph above shows both Mortality Collection and Office of the Chief Coroner data. We explain the differences between the two datasets in the Data caveats section below.
No system currently exists which allows for accurate, timely reporting on accidental fatal overdose in Aotearoa.
This year, we have decided to use two datasets for reporting on overdose deaths: coronial data from the Office of the Chief Coroner, and Mortality Collection data from Te Whatu Ora. Each has its strengths and limitations, and we present both where practical, for transparency. Presenting both data sources also highlights the current data gaps, and we hope that it will help readers to better understand why such a system is needed in Aotearoa.
Coronial datasetDATA SOURCE Office of the Chief Coroner data management system. Obtained by request under the Coroners Act 2006. STRENGTHS At present, this dataset is the most timely way to track fatal overdoses. It includes even the cases in which the Coroner has not yet definitively determined the cause of death. LIMITATIONS Inclusion of cases under investigation may result in overcounting or undercounting relevant cases, especially in the most recent years. The data extraction method does not allow for definitively excluding cases where an accidental overdose was not the main cause of death. This may lead to random fluctuation year to year. The main contributing class of substance is determined based on our own prioritisation system (see Prioritisation), and not on the totality of evidence for each case, which may overcount cases that are ranked higher in the hierarchy, and undercount cases that are ranked lower. Some cases do not have toxicology data, and instead rely on other circumstances to determine the cause of death. Even when toxicological investigations have been done, some novel or rarely occurring substances may not be included in the testing. HOW DID WE USE IT IN THE REPORT? Because of its timeliness, we primarily relied on coronial data to identify recent trends in the drugs involved in overdoses, and the overall recent trajectory of overdose deaths. |
Mortality datasetDATA SOURCE Mortality Collection dataset held by Te Whatu Ora. Obtained under an Information Sharing Agreement. STRENGTHS The mortality dataset includes almost exclusively finalised cases with definitive findings, so this data is unlikely to change over time. It uses a standardised coding system, relying on all the circumstances identified in coronial findings to determine the definitive underlying cause of death. The data is provided in an information-rich format that allows more robust analyses. LIMITATIONS There is a significant reporting lag, which prevents timely assessment of emerging trends. Some cases do not have toxicology data, and instead rely on other circumstances to determine the cause of death. Even when toxicological investigations have been done, some novel or rarely occurring substances may not be included in the testing. HOW DID WE USE IT IN THE REPORT? We have provided mortality dataset information for a longer time period which starts earlier than the period used in our coronial data analysis. Where practical, we have presented mortality data on the same graphs to highlight the degree of uncertainty in classified overdose cases. We note that in most cases, despite some numerical variation, both datasets show very similar trends. We primarily relied on the mortality dataset for demographic (age, gender, ethnicity) analyses, as these have remained relatively consistent over time. |
Prioritisation
In the coronial dataset, in cases where multiple drug types were detected in toxicology results, we attributed each case to one drug class using the ranked prioritisation system below:
- Opioid drugs
- Synthetic cannabinoids
- Benzodiazepines
- Alcohol
- Stimulants
- Other drugs (including antipsychotics, SSRIs, GHB/GBL)
- Hallucinogens
To determine this order, we looked at drug mortality research and guidelines from multiple international sources1,2 and compared the findings with drug use patterns in New Zealand. This provided us with a framework to prioritise which drug classes were likely to lead to an overdose death. It is important to bear in mind that combining multiple drugs greatly increases the likelihood of death from overdose. In some cases, the final coroner-determined cause of death will differ from our allocated prioritisation of these coronial cases.
Overdose deaths by gender
Men are more likely to die of an accidental drug overdose than women in New Zealand. In the mortality dataset, between 2016 and 2022, two thirds (66.7%) were male, and the remaining third (33.3%) were female. There is currently no data for gender diverse people.
Overdose deaths by gender
Mortality dataset: 2016-2022
Overdose deaths by drug type
Opioids are responsible for the largest number of overdoses. This includes illicit, prescribed, and diverted opioids. In the mortality dataset, between 2011 and 2022, opioids were deemed to have been the cause of nearly half (45.6%) of all fatal overdose cases. Coronial data shows a similar pattern (graph presented below).
Deaths by substance type: Active and closed coronial cases: 2016-2025
Changes over time
This graph depicts the number of cases for each substance group between 2016 and 2025.
Deaths by substance type
Active and closed coronial cases 2016-2025
Support our work to prevent overdoses in Aotearoa
Opioids
e.g., morphine, methadone, heroin, codeine, fentanyl, nitazenes
Opioid overdose deaths: Coronial and mortality datasets
- According to the coronial data, opioids are by far the biggest contributor to overdose deaths in New Zealand, featuring in 41.5% of cases between 2016 and 2025.
- 41% of closed opioid overdose cases involved more than one opioid (for example, codeine and tramadol).
- The risk of overdose from opioids increases when a person also takes another depressant drug. Between 2016 and 2025, 92.9% of closed opioid overdose cases also involved at least one other depressant drug (e.g., alcohol, a benzodiazepine, or another opioid).
- The most commonly featured opioids, in order, are: codeine, morphine†, methadone, and tramadol.
†Many opioids are metabolised to morphine in the body.
Synthetic cannabinoids
Synthetic cannabinoid overdose deaths: Coronial and mortality datasets
- Synthetic cannabinoid overdose deaths started to increase in 2017 and peaked in 2018, when New Zealand experienced a synthetic cannabinoid overdose crisis.
- Synthetic cannabinoid overdoses have since decreased and remained relatively stable.
- Between 2016 and 2025, 5.9% of closed cases featured a synthetic cannabinoid.
Benzodiazepines
e.g., alprazolam (Xanax), diazepam, etizolam, bromazolam
Benzodiazepine overdose deaths: Coronial and mortality datasets
- Between 2016 and 2025, 33.5% of all closed cases featured one or more benzodiazepines.
- Diazepam was the most common benzodiazepine featured in closed overdose cases, appearing 260 times.
- Benzodiazepines are most dangerous when mixed with other depressant drugs. Of the closed cases that involved benzodiazepines, nearly all (95.3%) also involved at least one other depressant drug (e.g., alcohol or codeine).
- Most benzodiazepines featured are prescription benzodiazepines. However, a novel benzodiazepine was detected in 17 coronial cases between 2016 and 2025.
Stimulants
e.g., methamphetamine, methylphenidate, synthetic cathinones
Stimulant overdose deaths: Coronial and mortality datasets
- While case numbers are still small, stimulant overdose deaths in 2025 remained higher than in the period before 2023.
- 58.8% of cases that contained a stimulant drug contained an amphetamine (e.g., methamphetamine).
- Cocaine (or its metabolite) featured in 24 cases – 16 of which occurred between 2023 and 2025.
Alcohol
Alcohol features in many overdose cases, particularly those where more than one substance is detected. Alcohol can add to the risk of overdose death, particularly when it is mixed with other depressant drugs such as opioids, benzodiazepines, or synthetic cannabinoids.
According to mortality data, between 2016 and 2022, alcohol was involved in more than half (53.6%) of all overdose deaths.
Alcohol overdose deaths: Coronial and mortality datasets
Multiple substances and fatal overdoses
Most fatal overdoses in New Zealand involve more than one drug or medicine. Between 2016 and 2025, more than one substance was involved in 85.4% of closed cases. Five or more different substances were involved in just over half (50.3%) of the closed cases.
Number of substances detected (closed cases): 2016–2025
Depressants are substances that slow down the nervous system, and may cause the heart or a person’s breathing to slow down. These substances include benzodiazepines, opioids, alcohol, and others. Mixing depressants can be particularly dangerous due to their additive effects. A depressant drug was involved in 84.1% of closed cases. More than one depressant was involved in 61.7% of closed cases.
Number of depressants detected: Closed coronial cases 2016-2025
Novel psychoactive substances
Novel psychoactive substances (NPS) are manufactured to mimic the effects of well-known illicit drugs. This is a wide group of substances that includes synthetic cathinones (‘bath salts’), synthetic cannabinoids (‘synnies’), novel benzodiazepines (e.g., etizolam, bromazolam), and novel synthetic opioids (e.g., nitazenes).
NPS can be difficult to test for, and some are not part of routine toxicology screening.
NPS appeared in 9.2% of closed cases from 2016–2025, featuring 196 times in total (some cases involved more than one).
Synthetic cannabinoids were most common, but novel benzodiazepines, novel opioids, and novel stimulants also featured in cases.
Medicines and fatal overdose
Medicines are involved in many overdose deaths in New Zealand.
We don’t know whether these medicines were used as prescribed, diverted, or illicitly made. Either way, it is important to consider how they can interact with illicit drugs.
Closed cases commonly involved medicines. At least one medicine was involved in 73.5% of all closed cases, and 23.5% of closed cases involved five or more medicines.
Number of medicines detected Closed coronial cases: 2016–2025
Some medicines feature more heavily in overdose death cases.
Twenty different medicines featured over 50 times each in closed cases. Diazepam, paracetamol, zopiclone, and codeine were the most common medicines appearing in closed overdose cases.
Ten most commonly involved medicines: Closed coronial cases: 2016–2025
Non-fatal overdoses
Drug poisoning presentations: public hospital discharge data
A non-fatal overdose is a major predictive factor for a future fatal overdose.
Very little research has been done to determine the incidence of non-fatal overdose in New Zealand. In this section, we present hospitalisation data from the National Minimum Dataset (NMDS), provided to the NZ Drug Foundation under an Information Sharing Agreement with the Te Whatu Ora Data Services team.
This data is only the tip of the iceberg of non-fatal drug overdoses in Aotearoa. Our analysis only captures cases where people presented to hospital, were assessed and either admitted, transferred, or discharged. It does not include people who were treated by ambulance teams, received help from members of the community, or presented to after-hours clinics or GPs.
Many people who experience a non-fatal overdose don’t receive any help at all.
Some recover without treatment. Others die before getting help and are counted in fatal overdose figures.
Hospitalisation data caveats
The NMDS data we present covers all Te Whatu Ora regions and uses the same measures across all regions between 2016 and 2025. We only included cases of people aged 15 and over in the data.
We present cases where the ICD-10 diagnostic code for the primary diagnosis included poisonings with the following substances:
- Opioids
- Stimulants
- Cannabis (may include some synthetic cannabinoid cases)
- Hallucinogens
- Other narcotics and psychotropics (includes most synthetic cannabinoid cases)
- Benzodiazepines (and barbiturates: very small number of cases)
- Alcohol
We believe that the criteria for selection of cases are conservative and will likely result in undercounting of accidental drug poisoning cases. We removed all cases where there were any codes present that could indicate intentional self-harm or poisoning by assault. We did not include cases where the substance involved was unknown. Cases where poisonings were present as non-primary diagnosis only were also excluded.
Anecdotal evidence suggests that occasionally cases that should have been classified as accidental poisonings are misclassified as intentional self-harm.
This analysis assumes the validity of clinical diagnoses recorded and does not include toxicology data. Because of the limitations of ICD-10 coding and lack of toxicological data, this analysis is unable to provide information on NPS poisonings.
There may be small changes in the data we report on from year to year, due to data-extract coding variations.
It is important to remember that not all accidental poisonings covered in this analysis resulted from deliberate ingestion of a substance. Some cases may involve someone unknowingly consuming a substance, though this is rare among adults.
Hospital presentations for drug poisonings
In 2025, there were 579 hospital presentations for drug poisonings. The most commonly implicated substance class was opioids, followed by stimulants and benzodiazepines.
Drug poisoning hospitalisations by substance class 2016-2025
10-year trends in drug poisonings resulting in hospital presentation
Hospitalisation trends for most substances appear have been stable or falling over the past ten years. Stimulant presentations are the exception, and have been rising since 2022.
Wastewater testing data shows significant recent increases in cocaine and methamphetamine use, so we expect this rise to continue.
Drug poisoning hospitalisations by substance class 2016-2025
Hospital presentations for drug poisonings over time by ethnicity
The number of presentations has been falling for non-Māori New Zealanders over the last five years. However, presentations have remained stubbornly high for Māori.
Drug poisoning hospitalisations by ethnicity 2016-2025
Deaths among people seen at New Zealand hospitals for drug poisoning
Between 2016 and 2025, 44 (0.68%) of 6,437 hospital presentations for drug poisoning resulted in death.
This low rate is not surprising, as people who reach hospital usually get timely care. Their poisonings may also have been less severe than those of people who died before getting help.
Young people are far more likely to be hospitalised for drug poisoning
Relative to population size, 15–24-year-olds had the highest rate of poisoning resulting in hospitalisation, much higher than any other age group.
Annual drug poisoning hospitalisations by age group
2016–2025
Māori experience a higher rate of drug poisoning compared to non-Māori
Relative to the population size, Māori had twice the rate of drug poisonings that resulted in hospital visits compared to non-Māori in New Zealand.
Annual drug poisoning hospitalisations by ethnicity
2016–2025
Summary and recommendations
The number of accidental overdose deaths in New Zealand remains stubbornly high, and may be on the rise. While the most recent figures are provisional, the 169 possible overdose deaths in 2025 warrant serious concern.
Māori continue to bear the greatest burden, suffering fatal overdoses at over twice the rate of non-Māori.
At the same time, people aged 45–54 have a higher rate of accidental fatal overdose than other age groups.
We continue to see worrying trends. These include a concerning proportion of stimulant-related deaths and hospitalisations. We also see some novel substances appearing in coronial data for the first time.
While our classification method likely overestimates the number of recent cases where alcohol was the cause of death, the number of recent alcohol-related overdoses may have increased. Alcohol continues to be involved in more than half of overdose cases.
Accidental overdoses are preventable.
We urgently need to shift our approach in order to reduce harm, prevent fatal overdoses, and prepare ourselves for major changes in drug supply.
Our recommendations
1. Develop and resource a comprehensive overdose prevention programme in Aotearoa, including targeted interventions for groups at elevated risk of fatal overdose (e.g., Māori and people aged 45–54)
This should include implementing the recommendations from our overdose prevention plan (PDF, 469 KB). There are some key actions that require urgent implementation:
- Interventions should target both those who have presented to hospital for drug poisoning and those who have never been in contact with the health system about their substance use. This is part of a set of interventions that the government has committed to in the Action Plan to Prevent and Reduce Substance Harm 2026 – 2029, and it should be rolled out as soon as possible.
- As opioids continue to be the leading underlying cause of death in accidental overdose cases, a funded take-home naloxone programme must be rolled out to avoid these preventable deaths.
2. Build a national overdose surveillance system
Reports like ours cannot replace a sustainable, standardised, and timely mechanism to track overdose deaths and non-fatal overdoses. A dedicated surveillance system is crucial for us to respond effectively and to help shape policies and services that will prevent overdose mortality.
3. Ensure our legislation and regulations enable action, including passing laws that encourage people to call for help when at risk of serious harm, like Good Samaritan overdose laws
Politicians must enable communities to protect each other from harm. Beyond major drug law reform, there are many examples of minor legislative and regulatory changes that will save lives, such as a Good Samaritan amendment or allowing non-commercial distribution of safer drug-use equipment.
Implementing interventions like overdose prevention centres or enabling legal trip sitting will save lives in Aotearoa New Zealand.
Help us build a future free from drug harm
Share:
Topics
Drugs
Recent news
Election 2026: Where the parties stand on reducing drug harm
There is common ground among parties on reducing drug harms and supporting addiction treatment, suggesting progress can be made in the next term.
SMART Recovery funding to grow local peer-led addiction support in communities
Government money for mental health and addiction, announced today, will go straight into building local capacity to support people who want to address substance use.
Drug Foundation supports focus on early intervention in new govt strategy
The NZ Drug Foundation is pleased to see a focus on early intervention to reduce drug harm in the Government’s new mental health and wellbeing strategy