Factors Associated with HIV Testing Among Participants from Substance Use Disorder Treatment Programs in the US: A Machine Learning Approach
Tóm tắt
HIV testing is the foundation for consolidated HIV treatment and prevention. In this study, we aim to discover the most relevant variables for predicting HIV testing uptake among substance users in substance use disorder treatment programs by applying random forest (RF), a robust multivariate statistical learning method. We also provide a descriptive introduction to this method for those who are unfamiliar with it. We used data from the National Institute on Drug Abuse Clinical Trials Network HIV testing and counseling study (CTN-0032). A total of 1281 HIV-negative or status unknown participants from 12 US community-based substance use disorder treatment programs were included and were randomized into three HIV testing and counseling treatment groups. The a priori primary outcome was self-reported receipt of HIV test results. Classification accuracy of RF was compared to logistic regression, a standard statistical approach for binary outcomes. Variable importance measures for the RF model were used to select the most relevant variables. RF based models produced much higher classification accuracy than those based on logistic regression. Treatment group is the most important predictor among all covariates, with a variable importance index of 12.9%. RF variable importance revealed that several types of condomless sex behaviors, condom use self-efficacy and attitudes towards condom use, and level of depression are the most important predictors of receipt of HIV testing results. There is a non-linear negative relationship between count of condomless sex acts and the receipt of HIV testing. In conclusion, RF seems promising in discovering important factors related to HIV testing uptake among large numbers of predictors and should be encouraged in future HIV prevention and treatment research and intervention program evaluations.
Tài liệu tham khảo
Centers for Disease Control and Prevention (CDC). HIV surveillance report 2013. 2015;25. http://www.cdc.gov/hiv/library/reports/surveillance/. Accessed 1 Nov 2016.
Hall HI, An Q, Tang T, et al. Prevalence of diagnosed and undiagnosed HIV infection—United States, 2008–2012. MMWR Morb Mortal Wkly Rep. 2015;26:657–62.
Centers for Disease Control and Prevention (CDC). Estimated HIV incidence among adults and adolescents in the United States, 2007–2010. HIV surveillance supplemental report 2012. 2012; http://www.cdc.gov/hiv/topics/surveillance/resources/reports/#supplemental, 17(No. 4).
Hall HI, Holtgrave DR, Maulsby C. HIV transmission rates from persons living with HIV who are aware and unaware of their infection. AIDS. 2012;26(7):893–6.
Branson BM, Handsfield HH, Lampe MA, et al. Revised recommendations for HIV testing of adults, adolescents, and pregnant women in health-care settings. MMWR. Recommendations and reports: morbidity and mortality weekly report. Recommendations and reports/Centers for Disease Control. 2006;55(RR-14):1–17; quiz CE11-14.
Centers for Disease Control and Prevention (CDC). Integrated prevention services for HIV infection, viral hepatitis, sexually transmitted diseases, and tuberculosis for persons who use drugs illicitly: summary guidance from CDC and the US Department of Health and Human Services. MMWR. Recommendations and reports: morbidity and mortality weekly report. Recommendations and reports/Centers for Disease Control. 2012;61(RR-5):1.
Centers for Disease Control and Prevention (CDC). HIV and substance use in the United States. 2015; http://www.cdc.gov/hiv/riskbehaviors/substanceuese.html. Accessed 11 Aug 2016.
Hess K, Hu X, Lansky A, Mermin J, Hall I. Estimating the lifetime risk of a diagnosis of HIV infection in the United States. 2016; http://www.croiconference.org/sessions/estimating-lifetime-risk-diagnosis-hiv-infection-united-states. Accessed 10 Sept 2016.
King KM, Nguyen HV, Kosterman R, Bailey JA, Hawkins JD. Co-occurrence of sexual risk behaviors and substance use across emerging adulthood: evidence for state- and trait-level associations. Addiction. 2012;107(7):1288–96.
Raj A, Saitz R, Cheng DM, Winter M, Samet JH. Associations between alcohol, heroin, and cocaine use and high risk sexual behaviors among detoxification patients. Am J Drug Alcohol Abuse. 2007;33(1):169–78.
Rosengard C, Anderson BJ, Stein MD. Correlates of condom use and reasons for condom non-use among drug users. Am J Drug Alcohol Abuse. 2006;32(4):637–44.
D’Aunno T, Pollack HA, Jiang L, Metsch LR, Friedmann PD. HIV testing in the nation’s opioid treatment programs, 2005-2011: the role of state regulations. Health Serv Res. 2014;49(1):230–48.
Substance abuse and mental health services Administration, Center for Behavioral Health Statistics and Quality. The N-SSATS report: HIV services offered by substance abuse treatment facilities. Rockville, M.D: Substance Abuse and Mental Health Services Administration; 2010.
Metsch LR, Feaster DJ, Gooden L, et al. Implementing rapid HIV testing with or without risk-reduction counseling in drug treatment centers: results of a randomized trial. Am J Public Health. 2012;102(6):1160–7.
Metcalf CA, Douglas JM Jr, Malotte CK, et al. Relative efficacy of prevention counseling with rapid and standard HIV testing: a randomized, controlled trial (RESPECT-2). Sex Transm Dis. 2005;32(2):130–8.
Skinner HA. Assessment of substance abuse: drug abuse screening test. 2nd ed. Durham: Macmillan Reference USA; 2001.
Skinner HA. The drug abuse screening test. Addict Behav. 1982;7(4):363–71.
NIAAA. Helping patients who drink too much: a clinician’s guide. Maryland: Bethesda; 2005.
Koblin BA, Husnik MJ, Colfax G, et al. Risk factors for HIV infection among men who have sex with men. Aids. 2006;20(5):731–9.
Metsch LR, McCoy CB, McCoy HV, et al. HIV-related risk behaviors and seropositivity among homeless drug-abusing women in Miami, Florida. J Psychoact Drugs. 1995;27(4):435–46.
Brafford LJ, Beck KH. Development and validation of a condom self-efficacy scale for college students. J Am Coll Health. 1991;39(5):219–25.
DeHart DD, Birkimer JC. Trying to practice safer sex: development of the sexual risks scale. J Sex Res. 1997;34(1):11–25.
Rush AJ, Trivedi MH, Ibrahim HM, et al. The 16-item quick inventory of depressive symptomatology (QIDS), clinician rating (QIDS-C), and self-report (QIDS-SR): a psychometric evaluation in patients with chronic major depression. Biol Psychiatry. 2003;54(5):573–83.
Breiman L. Random forests. Mach Learn. 2001;45(1):5–32.
Ishwaran H, Kogalur UB, Blackstone EH, Lauer MS. Random survival forests. Ann Appl Stat. 2008;2:841–60.
Ishwaran H, Kogalur UB. Random forests for survival, regression and classification (RF-SRC), R package version 1.6.1. 2015.
Xu, Ruo. “Improvements to random forest methodology”. Graduate Theses and Dissertations. 2013;Paper 13052.
Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat. 2001;29:1189–232.
Ishwaran H. Variable importance in binary regression trees and forests. Electron J Stat. 2007;1:519–37.
Chen X, Ishwaran H. Random forests for genomic data analysis. Genomics. 2012;99(6):323–9.
Shi M, He J. SNRFCB: sub-network based random forest classifier for predicting chemotherapy benefit on survival for cancer treatment. Mol BioSyst. 2016;12(4):1214–23.
Xiao LH, Chen PR, Gou ZP, et al. Prostate cancer prediction using the random forest algorithm that takes into account transrectal ultrasound findings, age, and serum levels of prostate-specific antigen. Asian J Androl. 2016.
Stekhoven DJ, Bühlmann P. MissForest—non-parametric missing value imputation for mixed-type data. Bioinformatics. 2012;28(1):112–8.
Noar SM, Cole C, Carlyle K. Condom use measurement in 56 studies of sexual risk behavior: review and recommendations. Arch Sex Behav. 2006;35(3):327–45.
Fonner VA, Kennedy CE, O’Reilly KR, Sweat MD. Systematic assessment of condom use measurement in evaluation of HIV prevention interventions: need for standardization of measures. AIDS Behav. 2014;18(12):2374–86.
Schroder KE, Carey MP, Vanable PA. Methodological challenges in research on sexual risk behavior: II. Accuracy of self-reports. Ann Behav Med. 2003;26(2):104–23.
Segal MR, Barbour JD, Grant RM. Relating HIV-1 sequence variation to replication capacity via trees and forests. Stat Appl Genet Mol Biol. 2004;3(1):1–18.
Xu S, Huang X, Xu H, Zhang C. Improved prediction of coreceptor usage and phenotype of HIV-1 based on combined features of V3 loop sequence using random forest. J Microbiol. 2007;45(5):441–6.
Dybowski JN, Heider D, Hoffmann D. Prediction of co-receptor usage of HIV-1 from genotype. PLoS Comput Biol. 2010;6(4):e1000743.