Response C to: “The impact of New Zealand’s 2018 methamphetamine policy on residential contamination”

Dear Editor,

I write in further support of the concerns raised by the New Zealand Drug Foundation Te Puna Whakaiti Pāmamae Kai Whakapiri¹ in relation to the paper by Irfan et al., “The impact of New Zealand’s 2018 methamphetamine policy on residential contamination” (New Zealand Medical Journal, 14 August 2026).²

I agree that the issues regarding disclosure of the relationship between one of the authors and Safe & Healthy Home Solutions (SHHS), together with the reliance on SHHS as the sole provider of the dataset, warrant careful consideration by the Journal.

However, I wish to raise several additional concerns beyond the conflict of interest issue.

1. Misrepresentation of the policy history

The paper’s central narrative rests on the proposition that a policy change occurred in 2018 that effectively raised the threshold for methamphetamine contamination from 0.5µg/100cm² to 15µg/100cm². This is misleading and incorrect.

The 2018 report by the Office of the Prime Minister’s Chief Science Advisor (PMCSA) was a scientific risk assessment and advisory report.³ It was not itself a regulatory instrument and did not directly result in the policy changes implied by the paper, which have only recently come into effect (April 2026). The distinction is important because the study’s entire causal framing is built around a purported “2018 policy change”.

The paper also states that the Institute of Environmental Science and Research (ESR) proposed lower threshold values in 2019. The report referred to by the authors to support this proposition (though not cited in their reference list) appears to be a report completed in 2016,⁴ which was available and considered during the 2018 PMCSA review. The chronology presented therefore risks giving readers a misleading impression that the ESR subsequently produced evidence contradicting the 2018 conclusions. No mention is made of later ESR work, including subsequent reports that informed the current regulatory framework.

2. Lack of engagement with major contrary evidence

The paper cites literature that supports concerns about methamphetamine contamination while giving little or no attention to major reviews and reports that reached substantially different conclusions.

In particular, the paper does not adequately engage with:

• the 2018 PMCSA review;
• the broader toxicological basis of that review;
• subsequent ESR analyses;^5,6 and
• later evidence reviews that informed the current regulatory settings.

A scholarly paper arguing for substantial policy change should be expected to engage with the most relevant contrary evidence and explain why previous assessments were incorrect. In my view, this standard has not been met.

3. Transparency and reproducibility of the dataset

The paper provides insufficient information to permit independent evaluation of the principal findings.

For example, the paper does not provide sufficient detail regarding:

• the number of swabs collected per property;
• whether sampling protocols were consistent over time;
• whether the same locations were sampled during repeat visits;
• the degree of standardisation between testing providers;
• how property-level contamination values were calculated from individual samples; and
• the distribution of changes observed across properties.

As a result, it is impossible for readers to determine whether the reported change reflects genuine changes in contamination levels, changes in sampling protocols, changes in testing behaviour or the influence of a relatively small number of highly contaminated properties.

The paper therefore presents a statistical result that cannot be meaningfully interrogated using the information provided.

4. Selection bias is potentially sufficient to explain the findings

The authors acknowledge that testing is voluntary and that properties suspected of contamination may have been more likely to be tested following the purported policy change. This is not a minor limitation. If lower-risk properties ceased to be tested while testing became concentrated among properties already suspected of heavy contamination, an increase in average contamination among tested properties would be expected even in the absence of any underlying increase in contamination across the housing stock. The paper nevertheless repeatedly interprets the findings as evidence that the policy itself led to increased contamination. I do not believe the analysis presented is capable of supporting such a causal inference.

5. Health claims substantially exceed the evidence presented

The paper contains no data on:

• human exposure;
• biological uptake;
• toxicology;
• health outcomes;
• healthcare utilisation; and
• child health.

Nevertheless, the abstract, discussion and conclusions repeatedly imply consequences for health, healthcare costs, child wellbeing and related outcomes. These claims go well beyond what can be inferred from the data presented. The paper is ultimately a study of contamination measurements in a selected group of properties. It is not a study of exposure and it is not a study of health effects.

6. Failure to distinguish contamination

Finally, I am concerned that the paper repeatedly treats contamination and health risk as though they are interchangeable. Even accepting the reported increase of approximately 1.9µg/100cm² at face value, the paper provides no evidence that such a change has any toxicological or public-health significance. The central scientific question is not whether methamphetamine can be detected on surfaces, but whether the measured levels translate into meaningful exposure and adverse health outcomes. The paper does not provide evidence addressing that question.

Conclusion

For these reasons, I support the New Zealand Drug Foundation Te Puna Whakaiti Pāmamae Kai Whakapiri’s request that the Journal undertake a broader editorial review of the article.

My concerns extend beyond the disclosure issue and relate to:

• accuracy of the policy history presented;
• omission of key contrary evidence;
• lack of transparency regarding the data and contamination metrics;
• unsupported causal claims; and
• health conclusions that are not supported by the data collected.

I would be happy to provide any further information that may assist the Journal in its consideration of these matters.

Kind regards,
Anne Bardsley