Response to peer review comments on “Methamphetamine contamination in
residential environments: analysis of evidence related to human health effects
(Cressey and Fowles, 2020)” Reviewer 1 (Leo Schep)
Comment
There is an emphasis on the literature Jackie Wright has published on the risks of toxicity
following exposure to low concentrations of methamphetamine (MA) in houses previously
contaminated by this drug (either through clandestine manufacture or recreational use). This
report has placed undue emphasis on her work.
A substantial amount of her work is based on a lifestyle property in rural Victoria that
previously had a clandestine lab operating. The new owners were purportedly contaminated
from surfaces, notably in the house, that had relatively high concentrations of MA. A section
of her investigations was published in MMWR in 2017.[1] where hair samples were assayed
for evidence of MA which were compared against purported signs and symptoms of toxicity.
Of note were the positive concentrations from the two youngest children, coded CSM7 and
CSM8. They had MA concentrations, incorporated into the hair, of 460 and 330 pg/mg
respectively.
She concluded:
“The data from this case study of a single family present evidence of adverse health
effects and reflect exposure to methamphetamine that occurred while living in a
home with environmental methamphetamine surface contamination levels in the
range of 11.7–26.0 μg/100 cm2.”
However, evidence of MA toxicity in hair is anticipated in people either exposed to this drug
during its actual manufacture, or following its recreational use. Bassindale, for example,
showed hair concentrations in children living in premises where children were exposed to the
manufacture of MA, had evidence of this drug in their hair ranging from 100 to 131,000
pg/mg.[2] The concentrations of hair samples from 2 of the children reported in Ms Wright’s
study are also within values anticipated with occasional recreational users.[3, 4]
Furthermore, the reported signs and symptoms of the new owners of that property were not
objectively assessed, such as by a clinician, but rather were self-reported or by the
caregiver. This is problematic.
Another report published in 2020 by this author[5], includes the results of this study along
with other cases. Again, there is self-reporting and correlated with hair analysis. Of interest
was the result from CS16, who purportedly had a hair concentration of 2,500 pg/mg. Again,
this is in the range of drug users and is very unlikely to have occurred following passive
exposure to surfaces in the owner’s house.
Contrary to Ms Wright’s reports, relatively low concentrations of MA on surfaces in houses,
particularly those where people have previously have smoked MA, are unlikely to lead to
host contamination and have measurable concentrations of this drug in their hair.
Response The ‘undue emphasis’ on the work of Ms Wright is largely due to the fact that this was the
predominant body of work on the human health ef ects of methamphetamine, published
within the time frame that it was agreed that ESR should consider. However, it should be
noted that the work of Ms Wright was not used in the substantive analyses carried out in the
ESR report. The reviewer’s comments place considerable focus on the concentrations of
methamphetamine in hair. This aspect of the Wright study was given lit le attention in the
ESR report, as it was considered to be uninformative.
The reviewer comments predominantly on Ms Wright’s article in MMWR. While ESR are
familiar with this article, it was not the subject of review in the ESR report.
The reviewer notes the self-reporting in the Wright (2020) study. This was also note in the
ESR report.
The final sentence of the reviewer’s comment appears to be a personal opinion,
unsupported by evidence or reference to evidence.
On balance, ESR sees no specific amendment that could be made to the report on the basis
of this comment.
Comment The authors should be commended for deriving a more realistic RfD of 3.0 rather than 0.3
µg/kg/day.
However, the derivation of this RfD, is based, in part, on the study conducted by Chapman in
1961,[6] and was derived from weight losses in pregnant women. By definition, an RfD
(reference dose) is the dose that humans can be exposed to daily, over their lifetime,
without
any harmful effects. It is assumed, therefore, that the measure of weight loss change is
associated with this drug is a potential harmful effect. However, In the United States, this
drug can be prescribed for short-term treatment of refractory obesity. Weight loss linked to
this drug is not a harmful effect. Harmful effects are more linked to the adverse adrenergic
and dopaminergic effects (such as adverse cardiovascular and psychotic signs and
symptoms).[7]
Furthermore, in the methods section of Chapman’s paper states:
Al patients were given a routine vitamin-mineral prenatal supplement. The
recommended diet stressed the use of protein and fresh vegetables,
but did not limit
quantity or proportions of food ingested. The six-meal-a-day plan was advised
repeatedly.
The question therefore arises: Were the changes in weight associated solely due to the
therapeutic doses of MA or were they also influenced by their uncontrolled dietary intake? To
my knowledge, this weakness in his study has not been discussed or considered when
people have derived an appropriate RfD value for MA.
Response Dr Schep has commented on the use of the Chapman study on previous occasions and is
on record as saying “It has problems, but I can see why they used that paper and I wouldn't
criticise them for that, because it is dif icult to try to establish a dose that is of concern.” (RNZ
interview, 2018).
Dr Schep comments in his current review that “Weight loss linked to this drug is not a
harmful effect”. However, in toxicological studies a weight loss or even a decrease in weight
gain is viewed as an adverse effect and in many cases may be the only adverse effect
characterised. To make a case that involuntary weight loss, secondary to a CNS stimulant
effect, should be discounted as a toxicological POD (point of departure), would be to call into
question hundreds of animal studies and would also open up the notion that other unsought
CNS effects, including euphoria, anxiolytic effects, etc, should also not be considered
adverse.
Dr Schep further comments on the uncontrolled dietary intakes of the study participants.
While this is not an ideal aspect of the study, the study did involve real human beings and,
as such, controlling all background factors is often not possible. The fact that reduced weight
gain was monotonically associated with methamphetamine dose under the double-blinded
protocol argues against an uncontrolled aspect of the study as the basis for the weight loss
reported. A sentence has been added to the report to highlight the uncontrolled nature of the
participants’ food intake.
Reviewer 2 (Barry Borman)
Comment Wright et al (2002) (pgs 26-7):
• ESR acknowledge the study “has many shortcomings”, which are described on pg
26. The report says “this study is currently the best available assessment of adverse
health effects”. However, these methodological deficiencies severely compromise
the validity of the study, and therefore lit le credence should be given to its findings.
• ESR say that Wright et al (2002) “did not identify a clear biological gradient” and that
the “principles of toxicology require such a gradient.”
• Assessing Wright et al (2002) against the criteria suggested by Hil (1965) that might
be suggestive of “causality”, show that there is lit le evidence of a causal
relationship.
Response It is not immediately clear what changes reviewer 2 would like to see in the ESR report. We
believe that the approach taken to assessing the study of Wright (2020) was suitably
objective and we consider that the conclusions drawn were suitably cautious. In a situation
where evidence is sparse, as in the current situation, rejecting evidence because it has
shortcomings wil often leave you with no evidence at all. In such circumstance, evidence
should not so much be rejected as ‘weighted’. We believe our comments have indicated the
level of weight that should be attached to the evidence from this study.
The report has not been amended on the basis of this comment.
Comment
Other studies (pgs 28-9): there is no critical evaluation of the study methodology, and
therefore, it is not known if the results might be plausible of due to some bias
• Burgess et al (1996) –- how were the labs selected?
• Witter et al (2007) - how were the personnel selected, and symptoms were self-
reported?
• Sternquist (2012) – no information is provided on the selection process or
confirmation of the symptoms
• Thrasher et al (2009) – these are case reports
Response The intent was not to formally review these papers, as they are outside the date range
agreed in the contract with HUD. However, the bul et points were considered individually, to
determine if improvements could be made to the report.
Bullet point 1 (BP1). The study interviewed officials who had investigated clandestine
laboratories. The laboratories were not the subject matter of this study and, hence, no
laboratories were selected. The proportion of investigations that involved methamphetamine
laboratories have been added to give greater context.
BP2. Have added further clarification. The reviewer notes that the symptoms were self-
reported – this is certainly not unusual in epidemiological studies and the intent of this
comment on the part of the reviewer is uncertain.
BP3. It is assumed that Sternquist (2012) is a reference to Ross and Sternquist (2012).
Details of the selection process and the medical assessment have been added.
BP4. As stated by the reviewer and the ESR report, the publication of Thrasher et al (2009)
is a summary of case reports. It is uncertain what action the reviewer is suggesting here.
