Kratom use in the United States: Response to Grundmann et al.

Abstract

We thank Grundmann et al. for their comments [1]. As pointed out, the cross-sectional study design is not intended to infer a causal relationship between kratom use and substance use disorders or risks of drug abuse [2]. Our study found that 21% of kratom users self-report behaviors classified as substantial or severe on the Drug Abuse Screening Test (DAST-10), a validated tool to assess these behaviors. Grundmann et al. accurately characterize prior studies as finding that a subset of kratom users report their reason for use as self-treatment of opioid drug dependence; however, the authors erroneously suggest that these reports are inconsistent with our finding of high DAST-10 scores in a portion of users. Reasons for use and general drug abuse behaviors are distinct, and while investigating reasons for kratom use is invaluable, it does not necessarily predict drug abuse behaviors. A survey by Garcia-Romeu et al. found that < 3% of users met moderate or severe DSM-5 criteria for kratom-related substance use disorder; however, as stated in the article's limitations, recruitment through websites of interest to kratom users probably resulted in selection bias that emphasized users ‘positively inclined towards kratom’ [3]. These results differ significantly from ours, but given the different drug abuse assessment scales and sampling methods, the prevalence of substance abuse behaviors and substance use disorders among kratom users requires further study. Grundmann et al. incorrectly characterize our survey as sampling only non-medical users. The Non-Medical Use of Prescription Drugs (NMURx) program samples the population of adults aged 18 years and older in the United States. Further, it is incorrect to suggest that prior surveys on websites frequented by kratom users are capable of estimating use in a general population, particularly in the absence of statistical controls. Estimates of prevalence must use appropriate sampling methods and statistical adjustments to account for non-probabilistic elements of on-line recruitment. The NMURx program has shown results congruent with national probability surveys [4]. Estimates of national prevalence by additional survey programs will further illuminate kratom's popularity. We strongly agree that further studies are warranted to clarify potential benefits and risks of use. Discourse is crucial regarding kratom's potential efficacy as a harm reduction drug and the decision for federal scheduling, but it is important to recognize that this discussion is currently based on retrospective and cross-sectional data. Furthermore, much of the current data were funded or obtained by the American Kratom Association, a group advocating for continued unrestricted access. Our article did not address possible safety profile differences between kratom and classical opioids. Rather, our intent was to provide the first estimates of prevalence at the national level, contextualized with additional data on demographics and drug abuse profiles, to further inform the national discussion. None.