181
Views
1
CrossRef citations to date
0
Altmetric
Methods in Addiction Research

Many leveled ordinal models for frequency of alcohol and drug use

ORCID Icon & ORCID Icon
Pages 566-575 | Received 14 Nov 2022, Accepted 07 May 2023, Published online: 26 Jun 2023
 

ABSTRACT

Background: The numbers of days people consume alcohol and other drugs over a fixed interval, such as 28 days, are often collected in surveys of substance use. The presence of an upper bound on these variables can result in response distributions with “ceiling effects.” Also, if some peoples’ substance use behaviors are characterized by weekly patterns of use, summaries of substance days-of-use over longer periods can exhibit multiple modes.

Objective: To highlight advantages of ordinal models with a separate level for each distinct survey response, for bounded, and potentially multimodal, count data.

Methods: We fitted a Bayesian proportional odds ordinal model to longitudinal cannabis days-of-use reported by 443 individuals who used illicit drugs (206 female, 214 male, 23 non-binary). We specified an ordinal level for each unique response to allow the exact numeric distribution implied by the predicted ordinal response to be inferred. We then compared the fit of the proportional odds model with binomial, negative binomial, hurdle negative binomial and beta-binomial models.

Results: Posterior predictive checks and the leave one out information criterion both suggested that the proportional odds model gave a better fit to the cannabis days-of-use data than the other models. Cannabis use among the target population declined during the COVID-19 pandemic in Australia, with the odds of a member of the population exceeding any specified frequency of cannabis use in Wave 4 estimated to be 73% lower than in Wave 1 (median odds ratio 0.27, 90% credible interval 0.19, 0.38).

Conclusion: Ordinal models can be suitable for complex count data.

Resumen

Introducción: A través de encuestas podemos documentar los días en que una persona consume alcohol y otras drogas en intervalos fijos de tiempo, por ejemplo, veintiocho días. La presencia de un límite superior en estas variables puede dar lugar a distribuciones de respuesta con “efectos de techo”. Además, si el consumo de sustancias se caracteriza por patrones semanales, los resúmenes de los días de consumo durante periodos más largos pueden presentar múltiples modalidades.

Objetivos: Destacar las ventajas de los modelos ordinales con un nivel separado para cada respuesta en la encuesta, para datos de recuento acotados y potencialmente multimodales.

Métodos: Se aplicó un modelo bayesiano de probabilidades proporcionales ordinales a los días de consumo longitudinal de cannabis declarados por 443 consumidores de drogas ilegales (206 mujeres, 214 hombres, 23 no binarios). Se especificó un nivel ordinal para cada respuesta única con el fin de poder inferir la distribución numérica exacta implicada por la respuesta ordinal predicha. A continuación se comparó el ajuste del modelo de probabilidades proporcionales con los modelos binomial, binomial negativo, binomial negativo con obstáculos y binomial beta.

Resultados: Las comprobaciones predictivas posteriores y el criterio de información de exclusión sugirieron que el modelo de probabilidades proporcionales se ajustaba mejor a los datos de días de consumo de cannabis que los demás modelos. El consumo de cannabis entre la población diana disminuyó durante la pandemia de COVID-19 en Australia, estimándose que las probabilidades de que un miembro de la población superara cualquier frecuencia especificada de consumo de cannabis en la Cuarta Ola eran un 73% menores que en la Primera Ola (mediana de la Odds ratio 0,27; intervalo de credibilidad del 90%: 0,19; 0,38).

Conclusión: Los modelos ordinales pueden ser adecuados para datos de recuento complejos.

Acknowledgments

We thank the people that participated in the ADAPT survey. Rachel Sutherland advised on the operational details of the study and Greta Baillie assisted with provision of the survey data. Special thanks to Andrea Zocco for the Spanish translation of the abstract. Comments from four anonymous reviewers improved the manuscript. CD acknowledges support from an Australian Research Council Future Fellowship (FT210100260). The National Drug and Alcohol Research Centre is supported by funding from the Australian Government Department of Health under the Drug and Alcohol Program.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Supplemental data

Supplemental data for this article can be accessed online at https://doi.org/10.1080/00952990.2023.2213868

Additional information

Funding

CD acknowledges support from an Australian Research Council Future Fellowship (FT210100260). The National Drug and Alcohol Research Centre is supported by funding from the Australian Government Department of Health under the Drug and Alcohol Program.

Log in via your institution

Log in to Taylor & Francis Online

PDF download + Online access

  • 48 hours access to article PDF & online version
  • Article PDF can be downloaded
  • Article PDF can be printed
USD 65.00 Add to cart

Issue Purchase

  • 30 days online access to complete issue
  • Article PDFs can be downloaded
  • Article PDFs can be printed
USD 987.00 Add to cart

* Local tax will be added as applicable

Related Research

People also read lists articles that other readers of this article have read.

Recommended articles lists articles that we recommend and is powered by our AI driven recommendation engine.

Cited by lists all citing articles based on Crossref citations.
Articles with the Crossref icon will open in a new tab.