Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2017 Dec 3.
Published in final edited form as: Drug Alcohol Depend. 2017 Jun 16;178:115–118. doi: 10.1016/j.drugalcdep.2017.04.033

Substance use among HIV-infected patients in Rio de Janeiro, Brazil: agreement between medical records and the ASSIST questionnaire

Iona K Machado a,b, Paula M Luz c, Jordan E Lake a, Rodolfo Castro c,d, Luciane Velasque d, Jesse L Clark a, Valdilea G Veloso c, Beatriz Grinsztejn c, Raquel B De Boni c
PMCID: PMC5712472  NIHMSID: NIHMS919843  PMID: 28646713

Abstract

Background

Substance use assessment is a challenge in busy clinical settings that may adversely affect HIV-infected persons. This study aimed to evaluate agreement between the medical chart and a standardized substance use screening questionnaire.

Methods

Of adults (n=1050) in HIV care in Rio de Janeiro who completed the World Health Organization’s Alcohol, Smoking and Substance Involvement Screening Test (ASSIST) we randomly selected 200 participants for medical chart review. Lifetime use of tobacco, alcohol, marijuana, and cocaine agreement between the medical record and ASSIST was evaluated using Kappa statistics. Sensitivity and specificity of chart information were also calculated.

Results

The median age was 42.4 years, 60.3% were male and 49.5% were white. Prevalence of lifetime use reported in ASSIST was 55.3% (tobacco), 79.4% (alcohol), 23.1% (marijuana), and 20.7% (cocaine). Any information on lifetime use was found in the medical chart for tobacco (n=180, 90.5%), alcohol (n=183, 92.0%), marijuana (n=143, 71.8%) and cocaine (n=151, 75.9%). The Kappa statistic, sensitivity and specificity of the medical chart accurately identifying lifetime substance users per ASSIST were respectively 0.60, 0.71, and 0.91 for tobacco; 0.22, 0.75, and 0.51 for alcohol; 0.58, 0.51, and 0.98 for marijuana; and 0.73, 0.75, and 0.96 for cocaine.

Conclusion

Considering inaccuracies in the medical chart, the implementation of brief, standardized substance use screening is recommended in HIV care settings.

Keywords: substance use, HIV/AIDS, concordance, medical record, questionnaire

1. Introduction

Brazilian surveys estimated that 15–17% of the adult population currently use tobacco, 24–54% consume alcohol weekly, while 2.5% reported marijuana use and 1.7% cocaine use in the last 12 months (IBGE, 2014; INPAD, 2014). Previous studies conducted at the Instituto Nacional de Infectologia Evandro Chagas (INI) of the Fundação Oswaldo Cruz (Fiocruz) in Rio de Janeiro (Brazil) found that 19.5– 29.9% of individuals under care for HIV were smokers, and 30.1%, 3.9% and 3.5% reported alcohol, marijuana and cocaine use, respectively, in the last 3 months (Machado et al., 2017; Torres et al., 2014). The higher prevalence of substance use among HIV–infected persons compared to the general population is worrisome. Considering international data, treatment for smoking should become a priority in HIV care because smoking decreases life expectancy (Reddy et al., 2016), maybe even more than HIV itself (Helleberg et al., 2013). In addition, problematic alcohol (Azar et al., 2010; Baum et al., 2010; Braithwaite et al., 2007; Gonzalez et al., 2011; Rehm et al., 2010) and cocaine use (Baum et al., 2009; Cook et al., 2008) have been associated with poorer HIV outcomes, while the findings literature on marijuana use are mixed (Bonn-Miller et al., 2014; Gonzalez et al., 2011).

Despite these findings, and as in other medical settings (Mitchell et al., 2012), substance use may not be regularly assessed during routine HIV care (Korthuis et al., 2008; Metsch et al., 2008). The paucity of quality information (Korthuis et al., 2010) in the medical chart may underestimate the prevalence of substance use disorders (Bai et al., 2014), and may be a missed opportunity to treat substance use and improve HIV outcomes. Given the importance of substance use assessment and the scarcity of data among Brazilian HIV-infected persons, we aimed to assess the agreement between substance use information recorded in the medical chart and reported in the ASSIST questionnaire, as well as sensitivity and specificity of the medical chart to detect substance use, among HIV-infected persons in HIV care in Rio de Janeiro, Brazil.

2. Materials and methods

We conducted a cross-sectional analysis at the INI/Fiocruz. A random sample of 200 participants was selected from 1050 HIV-infected adults (≥18 years of age) who are followed in the INI’s HIV cohort (Grinsztejn et al., 2009, 2007; Moreira et al., 2011) and completed the World Health Organization’s (WHO) Alcohol, Smoking, and Substance Involvement Screening Test vs3.1 (ASSIST) in Portuguese for the Brazilian population (Henrique et al., 2004). The sample size to detect a difference between a null Kappa=0.4 and hypothetical Kappa=0.6 with 80% power (alpha=0.05), considering the substance with lowest prevalence found in ASSIST (cocaine use, 20%) is n=187 (R Statistical Software version 3.2.2). One author (IKM) extracted documented references to endorsement or denial of tobacco, alcohol, marijuana, cocaine, and unspecified illicit drug use. Words searched for included: ‘bebidas alcoólicas’ (alcoholic beverages), ‘alcóol’ (alcohol), ‘cachaça’ (a popular type of liquor), ‘destilados’ (liquors), ‘cerveja’ (beer), ‘vinho’ (wine), ‘etilista’ (alcoholic), ’etoh’ (ethyl alcohol), ‘maconha’ (marijuana), ‘crack’, ‘cocaína’ (cocaine), and the non-specific phrase ‘drogas ilicitas’ (illicit drugs). Because denial of illicit substance use, such as marijuana and cocaine use, may be recorded non-specifically as no illicit drug use, charts with no specific marijuana or cocaine use information and explicitly recorded no illicit drug use were categorized as no marijuana or cocaine use for statistical calculations. The entire medical chart, including all physician, psychologist, and nutrition notes, of each participant was reviewed from the first appointment in the cohort to when the medical chart review was conducted, in October 2015 after the administration of ASSIST (August 2013 to September 2015). Because only 18 participants had information about any substance use within three months prior to ASSIST administration, any endorsement of substance use in the medical chart was compared to lifetime use per ASSIST.

Prevalence of both lifetime and current problematic substance use as per ASSIST are described. For each substance, lifetime use was a positive answer to the ASSIST question, “In your life, which of these substances have you used” with a list containing “tobacco,” “alcoholic beverages,” “marijuana,” and “cocaine, crack”. ASSIST scores were calculated per standardized instructions and problematic use of each substance was defined as an ASSIST score corresponding to moderate or high risk of experiencing health problems from the current pattern of use, ≥11 for alcohol and ≥4 for tobacco, marijuana, and cocaine (WHO, 2016). Demographic information was retrieved from INI’s HIV cohort database.

Agreement between medical charts and ASSIST information on lifetime use of each substance was evaluated by proportion agreement and Kappa statistics (κ). Considering ASSIST as the gold standard, sensitivity and specificity of medical chart information on lifetime use were also calculated. All statistical analyses were performed with R Statistical Software version 3.2.2.

3. Ethical Considerations

Studies were approved by the INI Institutional Review Board (CAAE 17844113.2.0000.5262 and CAAE 0032.0.009.000-10) and individuals signed an informed consent form. As this analysis used a de-identified dataset for purposes within the scope of the original consent form, this work was exempt from additional review by the University of California, Los Angeles Institutional Review Board.

4. Results

The study sample (n=199) had a median age of 42.4 years, was 60.3% male (n=120), 49.5% white (n=98), 68% heterosexual (n=134), and 55.6% of low education (never entered high school, n=110). The prevalence of lifetime use of tobacco, alcohol, marijuana and cocaine was 55.3% (110/199), 79.4% (158/199), 23.1% (46/199), and 20.1% (40/199) respectively. The prevalence of lifetime polysubstance use (use of ≥2 substances) was 59.3% (118/199). Problematic use of tobacco, alcohol, marijuana, and cocaine in the last 3 months was reported by 18.6% (37/199), 2.0% (4/199), 4.5% (9/199), and 5.0% (10/199) as captured by ASSIST (Table 1). Ten participants (10/199) screened positively for problematic substance use of more than one substance.

Table 1.

Number of participants with lifetime use, no lifetime use, or no information found in medical chart by lifetime (1a) and problematic use (1b) per ASSIST (INI FIOCRUZ, 2015).

a. ASSIST Medical Chart
Lifetime Use (n) No information Lifetime use No lifetime use No illicit drug use**
Tobacco Yes (110) 8 72 30 --
No (89) 11 7 71 --
Alcohol Yes (158) 12 109 37 --
No (41) 4 18 19 --
Marijuana Yes (46) 25 19 2 16
No (153) 151 2 0 104
Cocaine Yes (40) 15 24 1 7
No (159) 149 5 5 108
Unspecified Illicits* --- 74 2 124
b. ASSIST Medical Chart %***
Problematic Use (n) No information Lifetime use No lifetime use
Tobacco Yes (37) 0 34 3 92%
No (69) 8 34 27
Alcohol Yes (4) 0 4 0 100%
No (152) 12 104 36
Marijuana Yes (9) 3 6 0 67%
No (37) 22 13 2
Cocaine Yes (10) 3 7 0 70%
No (30) 12 17 1
Open in a new tab
*

Explicit statement of “illicit drug use” extracted from the medical chart

**

n participants with no marijuana or cocaine information and explicit denial of illicit drug use per the medical chart

***

% of problematic users with lifetime use documented in the medical chart

The medical chart did not provided any tobacco and alcohol use information for 9.5% (19/199) and 8.5% (16/199) of participants, while 27.6% (55/199) and 24.1% (48/199) lacked information on marijuana and cocaine use. The chart failed to capture 8.1% (3/37), 33.3% (3/9), and 30% (3/10) of problematic tobacco, marijuana, and cocaine users respectively (Table 1). Four participants reporting lifetime tobacco use and two reporting lifetime alcohol use were excluded from the calculation of problematic use because of ASSIST data incompleteness.

The agreements between the medical chart and ASSIST for tobacco (n=180), alcohol (n=183), marijuana (n=143) and cocaine (n=150) were 79%, 70%, 86%, and 91% with respective Kappa statistics of 0.60, 0.22, 0.58, and 0.73 (Table 2). The sensitivity of the medical chart accurately identifying lifetime substance users per ASSIST was 0.71 for tobacco, 0.75 for alcohol, 0.51 for marijuana, and 0.75 for cocaine with respective specificities of 0.91, 0.51, 0.98, and 0.96. (Table 2).

Table 2.

Agreement, sensitivity, specificity, and Kappa statistic of medical chart information by substance (INI FIOCRUZ, 2015).

Tobacco Alcohol Marijuana Cocaine
N 180 183 143 150
ASSIST Prevalence of Majority Response 0.57 (Yes) 0.80 (Yes) 0.74 (No) 0.79 (No)
Agreement 0.79 0.70 0.86 0.91
Sensitivity (95% CI) 0.71 (0.61, 0.79) 0.75 (0.67, 0.81) 0.51 (0.36, 0.67) 0.75 (0.58, 0.88)
Specificity (95% CI) 0.91 (0.83, 0.96) 0.51 (0.36, 0.67) 0.98 (0.93, 0.99) 0.96 (0.90, 0.98)
Kappa 0.60 (0.48, 0.71) 0.22 (0.05, 0.39) 0.58 (0.40, 0.75) 0.73 (0.59, 0.87)
Open in a new tab

Wilson-score confidence intervals reported

5. Discussion

The prevalences of lifetime tobacco, alcohol, marijuana, and cocaine use were 55.3%, 79.4%, 23.1%, and 20.7%, respectively, per ASSIST with moderate agreement with the medical chart for tobacco (0.60) and marijuana (0.58), poor agreement for alcohol (0.22), and good agreement for cocaine (0.73). As the agreement of lifetime alcohol use (70%) was lower than the prevalence of alcohol use per ASSIST (80%), the medical chart is a poorer predictor of alcohol use than assuming all charts indicate lifetime alcohol use. This poor agreement is reflected in the 30% (55/183) of participants who had conflicting information between the medical chart and ASSIST. That about two thirds (37/55) denied using alcohol to their health care provider but reported it on ASSIST may reflect social desirability bias or a permissive alcohol culture where alcohol use is not considered potentially problematic by patients or by health providers. As a converse example, the agreement between medical chart and ASSIST for lifetime cocaine use (91%) was higher than the prevalence of the majority response "no" (79%), indicating that the medical chart is a useful predictor of lifetime cocaine use. Excepting alcohol, the sensitivities of the medical chart accurately identifying lifetime substance use per ASSIST were lower (0.51–0.75) than the respective specificities (0.91–0.98), implicating that an additional effort is needed to identify substance use in HIV care. These results underscore the value of the ASSIST instrument in systematically collecting information that may be minimized or neglected during the regular medical interview (such as current substance use) but is crucial for planning the needs of mental health and addiction treatment in this population.

Limitations of this study include the use of convenience sampling (n=1050), which may not accurately represent the entire INI cohort nor other HIV-infected populations. Moreover, patients at INI may take part in research protocols that include substance use screening. This information becomes part of their medical chart, and therefore can affect their care, but does not necessarily reflect that patients’ own providers screened for substance use at a regular HIV clinic visit.

6. Conclusions

Both current and lifetime substance use prevalence, particularly that of marijuana and cocaine, is underreported in the medical chart and information was not complete enough to provide the diagnosis of substance use disorders. The medical chart was an accurate predictor of lifetime tobacco, marijuana, and cocaine but not alcohol use as identified by ASSIST. Given the potential impact of substance use on HIV and non-HIV outcomes, our results support standardized substance use screening as a regular part of HIV/AIDS care.

Highlights.

  • Information from the medical chart did not allow the diagnosis of substance use disorders.

  • The medical chart was a more specific than sensitive indicator of lifetime substance use

References

  1. Azar MM, Springer SA, Meyer JP, Altice FL. A systematic review of the impact of alcohol use disorders on HIV treatment outcomes, adherence to antiretroviral therapy and health care utilization. Drug Alcohol Depend. 2010;112:178–93. doi: 10.1016/j.drugalcdep.2010.06.014. [DOI] [PMC free article] [PubMed] [Google Scholar]
  2. Bai JR, Mukherjee DV, Befus M, Apa Z, Lowy FD, Larson EL. Concordance between medical records and interview data in correctional facilities. BMC Med. Res. Methodol. 2014;14:50. doi: 10.1186/1471-2288-14-50. [DOI] [PMC free article] [PubMed] [Google Scholar]
  3. Baum MK, Rafie C, Lai S, Sales S, Page B, Campa A. Crack-cocaine use accelerates HIV disease progression in a cohort of HIV-positive drug users. J. Acquir. Immune Defic. Syndr. 2009;50:93–9. doi: 10.1097/QAI.0b013e3181900129. [DOI] [PubMed] [Google Scholar]
  4. Baum MK, Rafie C, Lai S, Sales S, Page JB, Campa A. Alcohol Use Accelerates HIV Disease Progression. AIDS Res. Hum. Retroviruses. 2010;26:511–518. doi: 10.1089/aid.2009.0211. [DOI] [PMC free article] [PubMed] [Google Scholar]
  5. Bonn-Miller MO, Oser ML, Bucossi MM, Trafton JA. Cannabis use and HIV antiretroviral therapy adherence and HIV-related symptoms. J Behav Med. 2014;37:1–10. doi: 10.1007/s10865-012-9458-5. [DOI] [PubMed] [Google Scholar]
  6. Braithwaite RS, Conigliaro J, Roberts MS, Shechter S, Schaefer A, McGinnis K, Rodriguez MC, Rabeneck L, Bryant K, Justice AC. Estimating the impact of alcohol consumption on survival for HIV+ individuals. AIDS Care. 2007;19:459–466. doi: 10.1080/09540120601095734. [DOI] [PMC free article] [PubMed] [Google Scholar]
  7. Cook JA, Burke-Miller JK, Cohen MH, Cook RL, Vlahov D, Wilson TE, Golub ET, Schwartz RM, Howard AA, Ponath C, Plankey MW, Levine AM, Levine A, Grey DD. Crack cocaine, disease progression, and mortality in a multicenter cohort of HIV-1 positive women. AIDS. 2008;22:1355–63. doi: 10.1097/QAD.0b013e32830507f2. [DOI] [PMC free article] [PubMed] [Google Scholar]
  8. Gonzalez A, Barinas J, O’Cleirigh C. Substance use: impact on adherence and HIV medical treatment. Curr. HIV/AIDS Rep. 2011;8:223–34. doi: 10.1007/s11904-011-0093-5. [DOI] [PubMed] [Google Scholar]
  9. Grinsztejn B, Veloso VG, Friedman RK, Moreira RI, Luz PM, Campos DP, Pilotto JH, Cardoso SW, Keruly JC, Moore RD. Early mortality and cause of deaths in patients using HAART in Brazil and the United States. AIDS. 2009;23:2107–14. doi: 10.1097/QAD.0b013e32832ec494. [DOI] [PMC free article] [PubMed] [Google Scholar]
  10. Grinsztejn B, Veloso VG, Pilotto JH, Campos DP, Keruly JC, Moore RD. Comparison of clinical response to initial highly active antiretroviral therapy in the patients in clinical care in the United States and Brazil. J. Acquir. Immune Defic. Syndr. 2007;45:515–20. doi: 10.1097/QAI.0b013e3180decb6a. [DOI] [PubMed] [Google Scholar]
  11. Helleberg M, Afzal S, Kronborg G, Larsen CS, Pedersen G, Pedersen C, Gerstoft J, Nordestgaard BG, Obel N. Mortality Attributable to Smoking Among HIV-1 – Infected Individuals : A Nationwide, Population-Based Cohort Study. Clin. Infect. Dis. 2013;56:727–734. doi: 10.1093/cid/cis933. [DOI] [PubMed] [Google Scholar]
  12. Henrique IFS, De Micheli D, De Lacerda RB, De Lacerda LA, Formigoni MLODS. Validacao da Versao Brasileira do Teste de Triagem do Envolvimento com Alcool, Cigarro e Outras Substancias (ASSIST) Rev Assoc Med Bras. 2004;50:199–206. doi: 10.1590/s0104-42302004000200039. [DOI] [PubMed] [Google Scholar]
  13. IBGE. Pesquisa Nacional de Saúde 2013: percepção do estado de saúde, estilos de vida e doenças crônicas 2014 [Google Scholar]
  14. INPAD. II Levantamento Nacional de Álcool e Drogas (LENAD) - 2012, II Levantamento Nacional de ÁLcool e Drogas-LENAD. São Paulo 2014 [Google Scholar]
  15. Korthuis PT, Josephs JS, Fleishman JA, Hellinger J, Himelhoch S, Chander G, Morse EB, Gebo KA HIV Research Network. Substance abuse treatment in human immunodeficiency virus: the role of patient-provider discussions. J. Subst. Abuse Treat. 2008;35:294–303. doi: 10.1016/j.jsat.2007.11.005. [DOI] [PMC free article] [PubMed] [Google Scholar]
  16. Korthuis PT, Saha S, Chander G, Mccarty D, Moore RD, Cohn JA, Sharp VL, Beach MC. Substance Use and the Quality of Patient-Provider Communication in HIV Clinics. AIDS Behav. 2010 doi: 10.1007/s10461-010-9779-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
  17. Machado IK, Luz PM, Lake JE, Castro R, Velasque L, Clark JL, Veloso VG, Grinsztejn B, De Boni RB. Self-rated health and substance use among individuals in HIV care in Rio de Janeiro, Brazil: a cross-sectional study. Int. J. STD AIDS. 2017;0 doi: 10.1177/0956462417692278. 95646241769227. [DOI] [PMC free article] [PubMed] [Google Scholar]
  18. Metsch LR, Pereyra M, Colfax G, Dawson-Rose C, Cardenas G, McKirnan D, Eroglu D. HIV-positive patients’ discussion of alcohol use with their HIV primary care providers. Drug Alcohol Depend. 2008;95:37–44. doi: 10.1016/j.drugalcdep.2007.12.006. [DOI] [PubMed] [Google Scholar]
  19. Mitchell AJ, Meader N, Bird V, Rizzo M. Clinical recognition and recording of alcohol disorders by clinicians in primary and secondary care: meta-analysis. Br. J. Psychiatry. 2012;201:93–100. doi: 10.1192/bjp.bp.110.091199. [DOI] [PubMed] [Google Scholar]
  20. Moreira RI, Luz PM, Struchiner CJ, Morgado M, Veloso VG, Keruly JC, Grinsztejn B, Moore RD. Immune status at presentation for HIV clinical care in Rio de Janeiro and Baltimore. J. Acquir. Immune Defic. Syndr. 2011;57(Suppl 3):S171–8. doi: 10.1097/QAI.0b013e31821e9d59. [DOI] [PMC free article] [PubMed] [Google Scholar]
  21. Reddy KP, Parker RA, Losina E, Baggett TP, Paltiel AD, Rigotti NA, Weinstein MC, Freedberg KA, Walensky RP. Impact of Cigarette Smoking and Smoking Cessation on Life Expectancy Among People With HIV: A US-Based Modeling Study. J. Infect. Dis. 2016;214:jiw430. doi: 10.1093/infdis/jiw430. [DOI] [PMC free article] [PubMed] [Google Scholar]
  22. Rehm J, Baliunas D, Borges GLG, Graham K, Irving H, Kehoe T, Parry CD, Patra J, Popova S, Poznyak V, Roerecke M, Room R, Samokhvalov AV, Taylor B. The relation between different dimensions of alcohol consumption and burden of disease: an overview. Addiction. 2010;105:817–43. doi: 10.1111/j.1360-0443.2010.02899.x. [DOI] [PMC free article] [PubMed] [Google Scholar]
  23. Torres TS, Luz PM, Derrico M, Velasque L, Grinsztejn E, Veloso VG, Cardoso SW, Santini-Oliveira M, Grinsztejn B, De Boni RB, Kumar A. Factors associated with tobacco smoking and cessation among HIV-Infected individuals under care in Rio de Janeiro, Brazil. PLoS One. 2014;9:1–15. doi: 10.1371/journal.pone.0115900. [DOI] [PMC free article] [PubMed] [Google Scholar]
  24. World Health Organization. WHO - ASSIST V3.0 2016 [Google Scholar]

RESOURCES