Skip to main content

Advertisement

SpringerLink
Log in

Strategies of Managing Repeated Measures: Using Synthetic Random Forest to Predict HIV Viral Suppression Status Among Hospitalized Persons with HIV

  • Original Paper
  • Published:
AIDS and Behavior Aims and scope Submit manuscript

Abstract

The HIV/AIDS epidemic remains a major public health concern since the 1980s; untreated HIV infection has numerous consequences on quality of life. To optimize patients’ health outcomes and to reduce HIV transmission, this study focused on vulnerable populations of people living with HIV (PLWH) and compared different predictive strategies for viral suppression using longitudinal or repeated measures. The four methods of predicting viral suppression are (1) including the repeated measures of each feature as predictors, (2) utilizing only the initial (baseline) value of the feature as predictor, (3) using the last observed value as the predictors and (4) using a growth curve estimated from the features to create individual-specific prediction of growth curves as features. This study suggested the individual-specific prediction of the growth curve performed the best in terms of lowest error rate on an independent set of test data.

This is a preview of subscription content, log in via an institution to check access.

Access this article

Log in via an institution

Price excludes VAT (USA)
Tax calculation will be finalised during checkout.

Instant access to the full article PDF.

Institutional subscriptions

Fig. 1
Fig. 2

Similar content being viewed by others

Data Availability

The data used in the paper are available from the National Drug Abuse Treatment’s Clinical Trials Network Data Share.

Code Availability

Not applicable.

References

  1. Fauci AS. The AIDS epidemic–considerations for the 21st century. N Engl J Med. 1999;341(14):1046–50.

    Article  CAS  PubMed  Google Scholar 

  2. UNAIDS. Global HIV & AIDS statistics—2020 fact sheet. https://www.unaids.org/en/resources/fact-sheet

  3. HIV Care Continuum: U.S. Statistics. https://www.hiv.gov/federal-response/policies-issues/hiv-aids-care-continuum

  4. Centers for Disease Control and Prevention (CDC). Estimated HIV incidence and prevalence in the United States, 2014–2018. HIV Surveillance Supplemental Report 2020; 25(No. 1). https://www.cdc.gov/hiv/pdf/library/reports/surveillance/cdc-hiv-surveillance-supplemental-report-vol-25-1.pdf.

  5. Baker JV, Henry WK, Neaton JD. The consequences of HIV infection and antiretroviral therapy use for cardiovascular disease risk: shifting paradigms. Curr Opin HIV AIDS. 2009;4(3):176–82.

    Article  PubMed  PubMed Central  Google Scholar 

  6. Basavaraj KH, Navya MA, Rashmi R. Quality of life in HIV/AIDS. Indian J Sex Trans Dis AIDS. 2010;31(2):75–80.

    Article  CAS  Google Scholar 

  7. Walker N, Grassly NC, Garnett GP, Stanecki KA, Ghys PD. Estimating the global burden of HIV/AIDS: what do we really know about the HIV pandemic? Lancet. 2004;363(9427):2180–5.

    Article  PubMed  Google Scholar 

  8. Fauci AS, Redfield RR, Sigounas G, Weahkee MD, Giroir BP. Ending the HIV epidemic: a plan for the United States. JAMA. 2019;321(9):844–5.

    Article  PubMed  Google Scholar 

  9. 90–90–90: An Ambitious Treatment Target To Help End The AIDS Epidemic: UNAIDS. https://www.unaids.org/en/resources/909090.

  10. An Q, Prejean J, Hall HI. Racial disparity in US diagnoses of acquired immune deficiency syndrome, 2000–2009. Am J Prev Med. 2012;43(5):461–6.

    Article  PubMed  Google Scholar 

  11. Barash ET, Hanson DL, Buskin SE, Teshale E. HIV-infected injection drug users: health care utilization and morbidity. J Health Care Poor Underserved. 2007;18(3):675–86.

    Article  PubMed  Google Scholar 

  12. Metsch LR, Feaster DJ, Gooden L, Matheson T, Stitzer M, Das M, et al. Effect of patient navigation with or without financial incentives on viral suppression among hospitalized patients with HIV infection and wubstance use: a randomized clinical trial. JAMA. 2016;316(2):156–70.

    Article  PubMed  PubMed Central  Google Scholar 

  13. Vital signs: HIV prevention through care and treatment-United States. MMWR Morbidity and Mortality Weekly Report. 2011;60(47):1618–23.

  14. Chapin-Bardales J, Rosenberg ES, Sullivan PS. Trends in racial/ethnic disparities of new AIDS diagnoses in the United States, 1984–2013. Ann Epidemiol. 2017;27(5):329-34.e2.

    Article  PubMed  PubMed Central  Google Scholar 

  15. Hall HI, Gray KM, Tang T, Li J, Shouse L, Mermin J. Retention in care of adults and adolescents living with HIV in 13 U.S. areas. J Acquir Immune Defic Syndr (1999). 2012;60(1):77–82.

    Article  Google Scholar 

  16. Moore RD, Keruly JC, Bartlett JG. Improvement in the health of HIV-infected persons in care: reducing disparities. Clin Infect Dis. 2012;55(9):1242–51.

    Article  PubMed  PubMed Central  Google Scholar 

  17. Matthews GV, Dore GJ. HIV and hepatitis C coinfection. J Gastroenterol Hepatol. 2008;23(7 Pt 1):1000–8.

    Article  PubMed  Google Scholar 

  18. Chen JY, Feeney ER, Chung RT. HCV and HIV co-infection: mechanisms and management. Nat Rev Gastroenterol Hepatol. 2014;11(6):362–71.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  19. Kumar R, Singla V, Kacharya S. Impact and management of hepatitis B and hepatitis C virus co-infection in HIV patients. Trop Gastroenterol. 2008;29(3):136–47.

    PubMed  Google Scholar 

  20. Vulnerable groups and key populations at increased risk of HIV: The World Health Organization. http://www.emro.who.int/asd/health-topics/vulnerable-groups-and-key-populations-at-increased-risk-of-hiv.html.

  21. van Leth F, Andrews S, Grinsztejn B, Wilkins E, Lazanas MK, Lange JM, et al. The effect of baseline CD4 cell count and HIV-1 viral load on the efficacy and safety of nevirapine or efavirenz-based first-line HAART. AIDS (London, England). 2005;19(5):463–71.

    Article  PubMed  Google Scholar 

  22. Battegay M, Nüesch R, Hirschel B, Kaufmann GR. Immunological recovery and antiretroviral therapy in HIV-1 infection. Lancet Infect Dis. 2006;6(5):280–7.

    Article  CAS  PubMed  Google Scholar 

  23. Shoko C, Chikobvu D. A superiority of viral load over CD4 cell count when predicting mortality in HIV patients on therapy. BMC Infect Dis. 2019;19(1):169.

    Article  PubMed  PubMed Central  Google Scholar 

  24. Dubey A. Machine learning approaches in drug development of HIV/AIDS. Int J Mol Biol Open Access. 2018;3(1):23–5. https://doi.org/10.15406/ijmboa.2018.03.00044.

    Article  Google Scholar 

  25. Réda C, Kaufmann E, Delahaye-Duriez A. Machine learning applications in drug development. Comput Struct Biotechnol J. 2020;18:241–52.

    Article  PubMed  Google Scholar 

  26. Tran T, Luo W, Phung D, Gupta S, Rana S, Kennedy RL, et al. A framework for feature extraction from hospital medical data with applications in risk prediction. BMC Bioinformatics. 2014;15(1):425.

    Article  PubMed  PubMed Central  Google Scholar 

  27. Butcher B, Smith BJ. Feature engineering and selection: a practical approach for predictive models. Am Stat. 2020;74(3):308–9.

    Article  Google Scholar 

  28. Pasha SJ, Mohamed ES. Novel Feature Reduction (NFR) Model With machine learning and data mining algorithms for effective disease risk prediction. IEEE Access. 2020;8:184087–108.

    Article  Google Scholar 

  29. Luo D, Wang F, Sun J, Markatou M, Hu J, Ebadollahi S, editors. Sor: Scalable orthogonal regression for non-redundant feature selection and its healthcare applications. Proceedings of the 2012 SIAM International Conference on Data Mining; 2012: SIAM.

  30. Breiman L. Random forests. Mach Learn. 2001;45(1):5–32.

    Article  Google Scholar 

  31. NIH (National Institute on Drug Abuse). Linkage to Hepatitis C Virus Care among HIV/HCV Co-infected Substance Users. https://www.drugabuse.gov/about-nida/organization/cctn/ctn/research-studies/linkage-to-hepatitis-c-virus-care-among-hivhcv-co-infected-substance-users.

  32. Johnson MO, Neilands TB, Dilworth SE, Morin SF, Remien RH, Chesney MA. The role of self-efficacy in HIV treatment adherence: validation of the HIV Treatment Adherence Self-Efficacy Scale (HIV-ASES). J Behav Med. 2007;30(5):359–70.

    Article  PubMed  PubMed Central  Google Scholar 

  33. Cunningham WE, Andersen RM, Katz MH, Stein MD, Turner BJ, Crystal S, et al. The impact of competing subsistence needs and barriers on access to medical care for persons with human immunodeficiency virus receiving care in the United States. Med Care. 1999;37(12):1270–81.

    Article  CAS  PubMed  Google Scholar 

  34. Cunningham WE, Hays RD, Williams KW, Beck KC, Dixon WJ, Shapiro MF. Access to medical care and health-related quality of life for low-income persons with symptomatic human immunodeficiency virus. Med Care. 1995;33(7):739–54.

    Article  CAS  PubMed  Google Scholar 

  35. Yudko E, Lozhkina O, Fouts A. A comprehensive review of the psychometric properties of the drug abuse screening test. J Subst Abuse Treat. 2007;32(2):189–98.

    Article  PubMed  Google Scholar 

  36. Svetnik V, Liaw A, Tong C, Culberson JC, Sheridan RP, Feuston BP. Random forest: a classification and regression tool for compound classification and QSAR modeling. J Chem Inf Comput Sci. 2003;43(6):1947–58.

    Article  CAS  PubMed  Google Scholar 

  37. Pan Y, Liu H, Metsch LR, Feaster DJ. Factors associated with HIV testing among participants from substance use disorder treatment programs in the US: a machine learning approach. AIDS Behav. 2017;21(2):534–46.

    Article  PubMed  PubMed Central  Google Scholar 

  38. Salzberg SL. C4.5: programs for machine learning by J. Ross Quinlan. Morgan Kaufmann Publishers, Inc., 1993. Mach Learn. 1994;16(3):235–40.

    Article  Google Scholar 

  39. Kingsford C, Salzberg SL. What are decision trees? Nat Biotechnol. 2008;26(9):1011–3.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  40. Vidhya A. Tree Based Algorithms: A Complete Tutorial from Scratch (in R & Python) https://www.analyticsvidhya.com/blog/2016/04/tree-based-algorithms-complete-tutorial-scratch-in-python/

  41. Boehmke B, Greenwell BM. Hands-on machine learning with R. Boca Raton: Chapman and Hall/CRC; 2019.

    Book  Google Scholar 

  42. Ishwaran H, Malley JD. Synthetic learning machines. BioData Mining. 2014;7(1):28.

    Article  PubMed  PubMed Central  Google Scholar 

  43. Wallace ML, Buysse DJ, Redline S, Stone KL, Ensrud K, Leng Y, et al. Multidimensional sleep and mortality in older adults: a machine-learning comparison with other risk factors. J Gerontol A. 2019;74(12):1903–9.

    Article  Google Scholar 

  44. Friedman JH. Greedy function approximation: a gradient boosting machine. Ann Stat. 2001;29:1189–232.

    Article  Google Scholar 

  45. Rubin DB. Inference and missing data. Biometrika. 1976;63(3):581–92.

    Article  Google Scholar 

  46. Pedersen AB, Mikkelsen EM, Cronin-Fenton D, Kristensen NR, Pham TM, Pedersen L, et al. Missing data and multiple imputation in clinical epidemiological research. Clin Epidemiol. 2017;9:157–66.

    Article  PubMed  PubMed Central  Google Scholar 

  47. Tang F, Ishwaran H. Random forest missing data algorithms. Stat Anal Data Mining. 2017;10(6):363–77.

    Article  Google Scholar 

  48. McManus KA, Srikanth K, Powers SD, Dillingham R, Rogawski McQuade ET. Medicaid expansion’s impact on Human Immunodeficiency Virus outcomes in a nonurban Southeastern Ryan White HIV/AIDS program clinic. Open Forum Infect Dis. 2020. https://doi.org/10.1093/ofid/ofaa595.

    Article  PubMed  PubMed Central  Google Scholar 

  49. Eligibility of Medicaid Department of Health and Human Services; 2020 https://www.medicaid.gov/medicaid/eligibility/index.html.

  50. Derogatis LR, Melisaratos N. The Brief Symptom Inventory: an introductory report. Psychol Med. 1983;13(3):595–605.

    Article  CAS  PubMed  Google Scholar 

  51. Recklitis CJ, Parsons SK, Shih MC, Mertens A, Robison LL, Zeltzer L. Factor structure of the brief symptom inventory–18 in adult survivors of childhood cancer: results from the childhood cancer survivor study. Psychol Assess. 2006;18(1):22–32.

    Article  PubMed  Google Scholar 

  52. Andreu Y, Galdón MJ, Dura E, Ferrando M, Murgui S, García A, et al. Psychometric properties of the Brief Symptoms Inventory-18 (Bsi-18) in a Spanish sample of outpatients with psychiatric disorders. Psicothema. 2008;20(4):844–50.

    PubMed  Google Scholar 

  53. Trivedi MH, Wisniewski SR, Morris DW, Fava M, Gollan JK, Warden D, et al. Concise Health Risk Tracking scale: a brief self-report and clinician rating of suicidal risk. J Clin Psychiatry. 2011;72(6):757–64.

    Article  PubMed  Google Scholar 

  54. Maisto SA, Carey MP, Carey KB, Gordon CM, Gleason JR. Use of the AUDIT and the DAST-10 to identify alcohol and drug use disorders among adults with a severe and persistent mental illness. Psychol Assess. 2000;12(2):186–92.

    Article  CAS  PubMed  Google Scholar 

  55. Heatherton TF, Kozlowski LT, Frecker RC, Fagerström KO. The Fagerström Test for Nicotine Dependence: a revision of the Fagerström Tolerance Questionnaire. Br J Addict. 1991;86(9):1119–27.

    Article  CAS  PubMed  Google Scholar 

  56. Mor V, Laliberte L, Morris JN, Wiemann M. The Karnofsky Performance Status Scale. An examination of its reliability and validity in a research setting. Cancer. 1984;53(9):2002–7.

    Article  CAS  PubMed  Google Scholar 

  57. Schag CC, Heinrich RL, Ganz PA. Karnofsky performance status revisited: reliability, validity, and guidelines. J Clin Oncol. 1984;2(3):187–93.

    Article  CAS  PubMed  Google Scholar 

  58. Coates J, Anne Swindale and Paula Bilinsky. Household Food Insecurity Access Scale (HFIAS) for measurement of household food access: Indicator Guide (v. 3). Washington, DC: Food and Nutrition Technical Assistance Project, Academy for Educational Development, August 2007.

  59. Baker DW, Williams MV, Parker RM, Gazmararian JA, Nurss J. Development of a brief test to measure functional health literacy. Patient Educ Couns. 1999;38(1):33–42.

    Article  CAS  PubMed  Google Scholar 

  60. Kalokhe AS, Paranjape A, Bell CE, Cardenas GA, Kuper T, Metsch LR, et al. Intimate partner violence among HIV-infected crack cocaine users. AIDS Patient Care STDS. 2012;26(4):234–40.

    Article  PubMed  PubMed Central  Google Scholar 

  61. Paranjape A, Liebschutz J. STaT: a three-question screen for intimate partner violence. J Women’s Health. 2003;12(3):233–9.

    Article  Google Scholar 

  62. Thompson HS, Valdimarsdottir HB, Winkel G, Jandorf L, Redd W. The Group-Based Medical Mistrust Scale: psychometric properties and association with breast cancer screening. Prev Med. 2004;38(2):209–18.

    Article  PubMed  Google Scholar 

  63. Ware J Jr, Kosinski M, Keller SD. A 12-Item Short-Form Health Survey: construction of scales and preliminary tests of reliability and validity. Med Care. 1996;34(3):220–33.

    Article  PubMed  Google Scholar 

  64. Han C, Pulling CC, Telke SE, Huppler HK. Assessing the utility of five domains in SF-12 Health Status Questionnaire in an AIDS clinical trial. AIDS (London, England). 2002;16(3):431–9.

    Article  PubMed  Google Scholar 

  65. Mannheimer SB, Matts J, Telzak E, Chesney M, Child C, Wu AW, et al. Quality of life in HIV-infected individuals receiving antiretroviral therapy is related to adherence. AIDS Care. 2005;17(1):10–22.

    Article  CAS  PubMed  Google Scholar 

  66. SF-12v2 Scoring Web Service https://staging.qualitymetric.com/api2/amihealthy.asmx?op=SF12v2Scoring.

  67. McLellan AT, Luborsky L, Cacciola J, Griffith J, Evans F, Barr HL, et al. New data from the Addiction Severity Index. Reliability and validity in three centers. J Nerv Mental Dis. 1985;173(7):412–23.

    Article  CAS  Google Scholar 

  68. McLellan AT, Luborsky L, O’Brien CP, Woody GE, Druley KA. Is treatment for substance abuse effective? JAMA. 1982;247(10):1423–8.

    Article  CAS  PubMed  Google Scholar 

  69. Bohn MJ, Babor TF, Kranzler HR. The Alcohol Use Disorders Identification Test (AUDIT): validation of a screening instrument for use in medical settings. J Stud Alcohol. 1995;56(4):423–32.

    Article  CAS  PubMed  Google Scholar 

Download references

Funding

Funding for this study and by the National Institute on Drug Abuse under the following awards: UG1DA013720, UG1DA013035, UG1DA013034, UG1DA013727, UG1DA020024, UG1DA013732, UG1DA015831, UG1DA015815, and U10DA020036. Support from the University of Miami Center for HIV and Research on Mental Health (CHARM) (P30MH116867) and Center for AIDS Research (CFAR) (P30AI07396) is also acknowledged.

Author information

Authors and Affiliations

  1. Department of Public Health, Miller School of Medicine, University of Miami, Coral Gables, USA

    Jingxin Liu, Yue Pan, Mindy C. Nelson & Daniel J. Feaster

  2. Department of Sociomedical Sciences, Mailman School of Public Health, Columbia University, New York, USA

    Lauren K. Gooden & Lisa R. Metsch

  3. Miller School of Medicine, University of Miami, Coral Gables, USA

    Allan E. Rodriguez

  4. HIV Center for Clinical and Behavioral Studies, New York State Psychiatric Institute and Columbia University Irving Medical Center, New York, NY, USA

    Susan Tross

  5. Division of Infectious Diseases, Department of Medicine, Emory University School of Medicine and Hubert Department of Global Health, Rollins School of Public Health, Emory University, Atlanta, GA, USA

    Carlos del Rio

  6. Center for the Clinical Trials Network, National Institute on Drug Abuse, National Institutes of Health, Bethesda, MD, USA

    Raul N. Mandler

  7. Soffer Clinical Research Ctr, 1120 NW 14th St, Room 1059, Miami, FL, 33136-2107, USA

    Jingxin Liu

Authors
  1. Jingxin Liu
    View author publications

    You can also search for this author in PubMed Google Scholar

  2. Yue Pan
    View author publications

    You can also search for this author in PubMed Google Scholar

  3. Mindy C. Nelson
    View author publications

    You can also search for this author in PubMed Google Scholar

  4. Lauren K. Gooden
    View author publications

    You can also search for this author in PubMed Google Scholar

  5. Lisa R. Metsch
    View author publications

    You can also search for this author in PubMed Google Scholar

  6. Allan E. Rodriguez
    View author publications

    You can also search for this author in PubMed Google Scholar

  7. Susan Tross
    View author publications

    You can also search for this author in PubMed Google Scholar

  8. Carlos del Rio
    View author publications

    You can also search for this author in PubMed Google Scholar

  9. Raul N. Mandler
    View author publications

    You can also search for this author in PubMed Google Scholar

  10. Daniel J. Feaster
    View author publications

    You can also search for this author in PubMed Google Scholar

Contributions

JL: methodology, formal analysis, writing-original draft, writing—review and editing, visualization. YP, MCN: formal analysis, writing-review and editing, visualization. LKG; LRM; AER; ST; CdR; RNM: writing-review and editing. DJF: conceptualization, methodology, formal analysis, writing- original draft, writing—review and editing, supervision.

Corresponding author

Correspondence to Jingxin Liu.

Ethics declarations

Conflict of interest

Not applicable.

Ethical Approval

Initial data collection was performed under IRB approvals. This secondary analysis was exempt due to the use of deidentified data.

Consent to Participate

All participants in the surveys were explained study procedures and signed informed consent documents prior to participation in this research.

Consent for Publication

Not applicable.

Additional information

Publisher's Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Electronic supplementary material

Below is the link to the electronic supplementary material.

Supplementary file1 (DOCX 302 KB)

Appendix

Appendix

Coding of Features (Variables Used for Prediction)

Time Invariant Predictors

  1. 1.

    Age Age at the time of baseline assessment; treated as a continuous variable and rounded up to nearest integer with one decimal place.

  2. 2.

    Gender Male (1), female (2), transmale (female to male) (3), or transfemale (male to female) (4).

  3. 3.

    Race/Ethnicity Race—American Indian or Alaska Native (1); Asian (2); Black or African American (3); Native Hawaiian or Pacific Islander (4); White (5); Other (6); Ethnicity—Hispanic or Latino (1); Not Hispanic or Latino (2).

  4. 4.

    Asian ethnic background participants were asked, “Which of the following best describes your Asian ethnic background?”; response options were Chinese (1); Filipino (2); Japanese (3); Vietnamese (4); Korean (5); Indian (6); Pakistani (7); Other (8).

  5. 5.

    Black ethnic background participants were asked, “Which of the following best describes your Black ethnic background?”; response options were African American (1); Dominican (2); Haitian (3); Jamaican (4); Cuban (5); African (6); Other (7).

  6. 6.

    English language participants were asked, “Is English your second language?”, response options were no (0) or yes (1).

  7. 7.

    Education Middle school or less (1); Some high school, no diploma (2); High school diploma/GED or equivalent (3); Junior college (4); Technical/trade/vocation school (5); Some college (6); College graduate (7); Graduate or professional school (8).

  8. 8.

    Marital status Married (1); Widowed (2); Divorced (3); Separated (4); Never married (5).

  9. 9.

    Employment status Working full-time (1); Working steady part-time (2); Working only sometimes (3); Temporarily laid off, sick leave or maternity leave (4); Unemployed (5); Retired (6); Disabled (7); Unpaid child care or house work (8); Student (9); Currently incarcerated (10).

  10. 10.

    Working hours Participants were asked, “Regardless of full-time or part-time status, how many hours per week on average do you work?”; responses were numeric numbers range between 3 to 60.

  11. 11.

    Annual personal income participants were asked, “What was your total personal income in the last year from all sources?”; response options were $0 (1); $1 to $5000 (2); $5001 to 10,000 (3); $10,001 to $20,000 (4); $20,001 to $30,000 (5); $30,001 to $ 40,000 (6); $ 40,001 to $50,000 (7); more than $50,000.

  12. 12.

    Annual family income participants were asked, “What is your best estimate of the total income of all family members from all legal sources, before taxes, in last calendar year?”; response options were $0 (1); $1 to $5000 (2); $5001 to 10,000 (3); $10,001 to $20,000 (4); $20,001 to $30,000 (5); $30,001 to $ 40,000 (6); $ 40,001 to $50,000 (7); more than $50,000.

  13. 13.

    Health Insurance Status participants were asked, “Are you covered by health insurance or some other kind of health care plan?”; response options were no (0) or yes (1).

  14. 14.

    Health Insurance type participants were asked, “What kind of health insurance or health care coverage do you have?”; twelve specific types were given: Private Health Insurance; Medicare; Medi-gap; Medicaid; SCHIP (CHIP/children's health insurance program); Military health care (Tricare/VA/champ-VA); Indian health service; State-sponsored health plan; Other government program; Single service plan (e.g., dental, vision, prescriptions); ADAP; Other insurance; response options for each type were no (0) or yes (1).

  15. 15.

    Incarceration (ever) Participants were asked, “Have you ever been in jail, prison, or a correctional facility?”; response options were no (0) or yes (1).

  16. 16.

    Incarceration (recent) Participants were asked, “In the last 6 months, have you ever been in jail, prison, or a correctional facility?”; response options were no (0) or yes (1).

  17. 17.

    Living arrangements (recent) participants were asked, “In the last 6 months, indicated all the places you have lived”; eleven specific types were given: Homeless (living on the street, in a park, in a bus station, etc.); In a shelter; Transitional (time—limited) single—room occupancy hotel; Permanent single—room occupancy hotel; HIV/AIDS housing/group home; Drug treatment facility; Other residential facility or institution (e.g. health care facility, halfway house); Staying with family/friend; Rent an apartment/house (alone or with others); Own my home; In Jail; response options for each type were no (0) or yes (1).

  18. 18.

    Hospitalizations (last year) participants were asked, “How many times have you been hospitalized in the last 12 months?”, treated as a continuous variable and entered by participants, ranged from 0 to 15.

  19. 19.

    Hospitalizations (last 6 months) participants were asked, “How many times have you been hospitalized in the last 6 months?”, treated as a continuous variable and entered by participants, ranged from 0 to 12.

  20. 20.

    Alcohol or drug treatment participants were asked, “Have you ever participated in alcohol or drug treatment”, response options were no (0) or yes (1).

  21. 21.

    Psychological distress The Brief Symptom Inventory (BSI) survey was applied to measure participants’ psychological stress, which was first introduced by Derogatis [51] and validated by others [52, 53], 18 questions regarding mental health symptoms were asked, such as “Faintness or dizziness”, “Feeling no interest in thing”,. Based on their answers, total scores of each sub-category were calculated: Somatization; Obsession-Compulsion; Interpersonal Sensitivity; Depression; Anxiety; Hostility; Phobic Anxiety; Paranoid Ideation; Psychoticism and were dichotomized to not psychologically distressed (0) or yes (1).

  22. 22.

    Hepatitis C status participants were asked, “Have you ever been diagnosed with hepatitis C?”; response options were no (0) or yes (1).

  23. 23.

    Urine Drug screen participants’ urine samples were tested for existence of substances use, including amphetamine, marijuana, THC, methamphetamine, opiates, cocaine, MDMA ecstasy, oxycodone, methadone, barbiturate, buprenorphine, benzodiazepines; an indicator of whether used drugs—no (0) or yes (1) was created.

Time-Varying Predictors

  1. 24.

    Medication Adherence Self-Efficacy (past month) Score was obtained from the HIV Treatment Adherence Self-Efficacy Scale (HIV-ASES) survey [32]; participants were asked to scale 12 question such as “Stick to your treatment plan even when side effects begin to interfere with daily activities”; Based on their answers, overall total score, subscale of integrity and personal of the of Medication Adherence Self-Efficacy were calculated.

  2. 25.

    Access to Care (past 6 month) Score was obtained from the validated Access to Care survey [33, 34]; participants were asked to scale six statements on their experience about access to care, such as “If I need hospital care, I can get admitted without trouble”; total score was calculated.

  3. 26.

    Concise Health Risk Tracking (past week) the Concise Health Risk Tracking Scale [54] was applied, 12 questions including “I feel as if things are never going to get better” were asked, based on participants’ response to each question, the total score was calculated to measure the extent of potential Health risks.

  4. 27.

    Acquired immunodeficiency syndrome (AIDS) status Lab results were used identify patients’ AIDS status, those who has less than 200 CD4 cells/μl, or a CD4 percentage less than 14%, is considered to have AIDS, status was no (0) or yes (1).

  5. 28.

    Drug Abuse Severity the Dast-10 measurement [35] was applied to assess the drug use severity, which has been validated by Maisto et al. [55]. Participants were asked to scale 10 statements such as “Have you used drugs other than those required for medical reasons”; response options were no (0) or yes (1). The total score was calculated to measure the extent of drug use.

  6. 29.

    Nicotine dependence The Fagerstrom Test for Nicotine Dependence (FTND) survey was used to assess the intensity of physical addiction to nicotine. This instrument was modified and validated by Todd Heatherton, et al. [56]. An example question was “How soon after you wake up do you smoke your first cigarette?”;. the total score was calculated based on participants’ answers.

  7. 30.

    Karnofsky Functioning Status (Past Week) Scores were obtained from the validated Karnofsky Performance Status Scale (KPS) [57, 58], participants were asked to scale performance on normal activities and work, score ranged from 0 (dead) to 100% (no special care is needed)..

  8. 31.

    Food Access (past month) Household Food Insecurity Access Scale (HFIAS) [59] was used to measure food insecurity, participants were asked nine occurrence questions such as “did you worry that you would not have enough food? If yes, how often did this happen?”; response options were rarely (once or twice in the past 4 weeks) (1); sometimes (three to ten times in the past 4 weeks) (2); Often (more than ten times in the past 4 weeks) (3).. The total score of access to food was calculated and was categorized into four groups: food secure (1); mildly food insecure (2); moderately food insecure (3); severely food insecure (4).

  9. 32.

    Health Literacy score Test of Functional Health Literacy in Adults (TOFHLA) was used to measure the ability to understand medical materials, details were described by Baker DW et al. [60] participants answered three questions, such as “How often do you have someone help you read hospital materials?”; the total score of Health Literacy ranged from 0 to 12.

  10. 33.

    Arrested (last 6-month) participants were asked, “Have you been arrested in the last 6 months”, response options were no (0) or yes (1).

  11. 34.

    Incarceration (last 6-months) participants were asked, “Have been incarcerated in the last 6 month?”, response options were no (0) or yes (1).

  12. 35.

    Intoxication Arrest History (ever) participants were asked, “Have you ever been arrested for public Intoxication?”, response options were no (0) or yes (1).

  13. 36.

    Driving under influence (DUI) experience (ever) participants were asked, “Have you ever driven under influence of alcohol or drugs?”, response options were no (0) or yes (1).

  14. 37.

    Illegal drug possession history (ever) participants were asked, “Have you ever used or possessed illegal drug?”, response options were no (0) or yes (1).

  15. 38.

    Drug paraphernalia (ever) participants were asked, “Have you ever had possession of drug paraphernalia?”, response options were no (0) or yes (1).

  16. 39.

    Abuse History as a child (ever) participants were asked, “As a child, were you ever beaten, physically attacked, or physically abused?”, “As a child, were you ever sexually attacked, raped, or sexually abused?”, response options were no (0) or yes (1). If any of the statements was yes, then ever been abused as a child was yes (1).

  17. 40.

    Abuse history as an adult (ever) participants were asked, “As an adult, have you ever been beaten, physically attacked, or physically abused?”, “Were you ever in a relationship where a sexual partner beat, physically attacked or physically abused you?”, “As an adult, have you ever been sexually attacked, raped, or sexually abused?”; “Were you ever in a relationship where a sexual partner sexually attacked, raped, or sexually abused you?”; response options were no (0) or yes (1). If any of the statements was yes, then ever been abused as an adult was yes (1).

  18. 41.

    Intimate Partner Violence the Intimate Partner Violence (IPV) tool was adapted [61] from a previously validated instrument, the STaT (Slapped, Threatened, or Throw things) [62] and was used to screen for Intimate Partner Violence experience. Participants were asked five questions such as “Have you ever been in a relationship where a sexual partner threatened you with violence?”; response options were no (0) or yes (1). If any of the statements was yes, then ever had interpersonal violence experience was yes (1).

  19. 42.

    Medical Mistrust Scores were obtained from The Group-Based Medical Mistrust Scale (GBMMS), which contained 12 items and was validated by Thompson et al. [63]. Participants were asked to scale statements such as “Doctors and health care workers sometimes hide information from patients who belong to my ethnic group”; response options were strongly disagree (1); disagree (2); neither (3); agree (4); strongly agree (5). The total score of medical mistrust was calculated.

  20. 43.

    Readiness for Substance Use Treatment participants were asked to scale 8 statements, “Treatment could be your last chance to solve your substance use problems”; “If you enter treatment, you will stay for a while”; “Treatment could really help you”; “You want to be in a treatment program”; “Most counselors in substance use treatment programs are "squares" who don't understand substance users”; “Substance use treatment programs have too many rules and regulations for me”; “I don't think I could trust many of the people who work in the substance use treatment programs”; “It takes too much time and effort to get into a substance use treatment program”; response options were strongly disagree (1); disagree (2); undecided (3); agree (4); strongly agree (5). The total score was calculated. Additionally, the subtotals of the former four and last four statements were calculated to measure the readiness, and negative attitude towards Substance Use Treatment.

  21. 44.

    Perceived Health Status (past month) A computer-assisted version of the SF-12 measure [64] was used to assess the perceived health status, this measurement was validated among HIV population [65, 66]; participants were asked to scale statements regarding their health such as “In general, would you say your health is”; example response options were excellent (1), very good (2), good (3), fair (4), poor (5);. Based on those responses, the norm-based scores for the physical functioning (PF), role physical (RP), bodily pain (BP), general health (GH), vitality (VT), social functioning (SF), role emotional (RE), mental health (MH) scales, as well as the physical component summary (PCS), mental component summary (MCS) were calculated by using the SF-12v2 Scoring Web Service [67].

  22. 45.

    Conflictual social interactions (past month) participants were asked, during the past four weeks, how much of the time have you “Had serious disagreements with your family about things that were important to you?”; “had serious disagreements with your friends about things that were important to you?”; How often was “Someone to love and make you feel wanted?”; “Someone to help with daily chores (child care, buying food, preparing meals) if you were sick?”; “Someone to help you buy medicines?”; “Someone to help with transportation?”; “Someone to give you money if you need it?”; response options were none of the time (1), a little of the time (2), some of the time (3), most of the time (4), all of the time (5). The total score, as well as the sub-scores of conflict and supportive social interaction were calculated.

  23. 46.

    ER utilization (past 6 months) participants were asked, “During the past 6 months, did you go to a hospital emergency room for emergency care? Include any visits to the emergency room, even if you were admitted to the hospital from there. Include emergency rooms of psychiatric hospital”; response options were no (0) or yes (1). If they answered yes, further question “how many different times did you go to a hospital emergency room for emergency care during the past 6 months, including psychiatric hospitals” was asked. The dichotomous indicator of utilized the ER service and the number of ER usage were calculated.

  24. 47.

    Hospital utilization (past 6 months) participants were asked, “During the past 6 months, were you a patient in any hospital overnight or longer? Include psychiatric hospital”; response options were no (0) or yes (1). If they answered yes, further question “how many separate overnight hospital stays did you have during the past 6 months, including psychiatric hospital stays?” was asked. The dichotomous indicator of staying the hospital overnight and the count were calculated.

  25. 48.

    Respite Care utilization (past 6 months) participants were asked, “During the past 6 months, did you spend one or more nights in a respite care facility, personal care home, nursing home, rehabilitation center, or hospice facility?”; response options were no (0) or yes (1). If they answered yes, further question “how many separate stays in a nursing home or hospice facility did you have during the past 6 months?” was asked. The dichotomous indicator of using the respite care and the count were calculated.

  26. 49.

    Day hospitalization utilization (past 6 months) participants were asked, “During the past 6 months, did you attend any medical program where you spent the day there but went home at night? Include day hospitals, partial hospitalizations, or intensive outpatient programs for reasons other than substance abuse?”; response options were no (0) or yes (1). If they answered yes, further question “how many different programs like this did you go to?” was asked. The dichotomous indicator of attending the medical program and the count were calculated.

  27. 50.

    Medical Care utilization (past 6 months) participants were asked, “During the past 6 months, did you go to any hospital clinic, hospital outpatient department, community clinic or neighborhood health center for medical care, for example, to care for your HIV/AIDS or other physical problems?”; response options were no (0) or yes (1). If they answered yes, further question “how many different hospital clinics, outpatient departments, community clinics or neighborhood health centers did you visit for medical care during the past 6 months?” was asked. The dichotomous indicator of using the medical care center and the count were calculated.

  28. 51.

    Outpatient substance abuse treatment utilization (past 6 months) participants were asked, “How many days did you attend intensive outpatient substance abuse treatment in the past 6 months?”; the total counts was calculated.

  29. 52.

    Dental care utilization (past 6 months) participants were asked, “During the past 6 months, did you get any dental care?”; response options were no (0) or yes (1).

  30. 53.

    Support group utilization (past 6 months) participants were asked, “During the past 6 months, did you participate in any other support group, group counseling or self-help group for emotional, substance abuse or health issues? This would include groups led by an unpaid professional, for example clergy, or other provider”; response options were no (0) or yes (1). If they answered yes, further question, “how many group sessions did you attend with one of these providers to discuss substance use issues?”; the total number was calculated.

  31. 54.

    Discrimination participants were asked, “In a health care setting have you ever experienced discrimination, been prevented from doing something, or been hassled or made to feel inferior because of your: ‘HIV status?’; or ‘gender?’; or ‘sexual orientation or practices?’; or ‘race or ethnicity?’; or ‘drug use?’”; response options were no (0) or yes (1), based on participants’ answers, dichotomized discriminations scales were created.

  32. 55.

    Drug use the ASI (Addiction Severity Index) Lite survey was used to assess the addiction severity index [68, 69]; the composite score for alcohol use and drug use were calculated.

  33. 56.

    HIV Viral load the clinical labs were performed to measure patients’ HIV viral load, the number of copies per ml was noted if the Viral load > 200 copies/mL, otherwise the lab’s lower limit was noted.

  34. 57.

    Alcohol use (past year) was assessed using the Alcohol Use Disorders Identification Test (AUDIT) [70]. Participants were asked on the frequency about drinking, “How often do you have a drink containing alcohol?”; “How often do you have six or more drinks on one occasion?”; “How often during the last year have you found that you were unable to stop drinking once you started?”; “How often during the last year have you failed to do what was normally expected of you because of drinking?”; “How often during the last year have you needed a first drink in the morning to get yourself going after a heavy drinking session?”; “How often during the last year have you felt guilt or remorse after drinking?”; “How often during the last year have you been unable to remember what happened the night before because of drinking?”; response options were never (0); Less than monthly (1); Monthly (2); Weekly (3); Daily or almost daily (4). Participants were also asked, “How many drinks containing alcohol do you have on a typical day when drinking?”; response options were 1 or 2 (0); 3 or 4 (1); 5 or 6 (2); 7 to 9 (3); 4 or more (4). Besides, participants were asked, “Have you or someone else been injured as a result of your drinking?”; “Has a relative, friend, doctor, or other health worker, been concerned about your drinking or suggested you cut down?”; response options were no (0), yes but not in the last year (2), yes during the last year (4). The total score for alcohol use, scores of subscales including hazardous alcohol consumption, harmful alcohol consumption, and alcohol dependence were calculated.

Rights and permissions

Springer Nature or its licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Liu, J., Pan, Y., Nelson, M.C. et al. Strategies of Managing Repeated Measures: Using Synthetic Random Forest to Predict HIV Viral Suppression Status Among Hospitalized Persons with HIV. AIDS Behav 27, 2915–2931 (2023). https://doi.org/10.1007/s10461-023-04015-1

Download citation

  • Accepted:

  • Published:

  • Issue Date:

  • DOI: https://doi.org/10.1007/s10461-023-04015-1

Keywords

Search

Navigation