Table 2.
Data extraction from clinical studies examining associations between CU and HIV status in humans
| Author/year | Sample characteristics (% male) | Cannabis use criteria | Cognitive assessments | Cognitive domains | Main findings |
|---|---|---|---|---|---|
| Attonito et al. 2014 | 370 PWH: > 50% CU + (49% M) | no. of days of CU in the past 3 months | AVLT, The Color Trails Test 2; Short Category Test | Executive function, attention, information-processing, memory | Higher number of CU days correlated with worse psychomotor speed and attention (p = 0.051) |
| Bedi et al. 2010 | 7 PWH: 100% current and chronic CU (≥ 2 × /week) (100% M) | 16-day dronabinol (5–10 mg) + 16-day placebo (4×/day) | DSS, RAT, DAT, immediate and delayed DRT, 10-min rapid information-processing task | Attention, information-processing, verbal memory | Dronabinol was associated with worse processing-speed, latency (after 9–16 day treatment), and more false alarms (days 1–8), but better acquisition compared to placebo; No effect on information-processing or recall |
| Byrd et al. 2011 | 1284 PWH: 924 CU + (67% M) | Interview-Substance Abuse Module for frequent use (≥5×) | WAIS-III DS, WAIS-III SS, TMA, SMT, FMT, WCST, TMB, COWAT, CF, WAIS-III LNS, PASAT, GPT | Global cognition [executive function, attention, information processing, learning, memory, verbal-fluency, motor]; individual subdomain scores also reported | Lifetime “dosage” of cannabis was weakly associated with better verbal-fluency (p<0.1) |
| Chang et al. 2006 | PWH: 21 C − (81% M) 21 C + (86% M) PWoH: 30 C − (80% M) 24 C + (83% M) |
C + : regular CU or history of chronic CU C − : < 1 joint per month or no use |
Timed gait, RAVLT, GPT, Stroop, New adult reading test revised, TMA, TMB, SDM, CalCAP, Tests for working memory, and visual discrimination and response inhibition (degraded words with distracters, response reversal/visual scanning, and form discrimination tasks) | Executive function, information processing, verbal memory, motor function | CU, irrespective of HIV diagnosis, was associated with better executive function, information-processing and motor skills, but no effects on any cognitive function after age was included as a covariate |
| Christopher-Hayes et al. 2021 | PWH: 22 C − (82% M) 18 C + (78% M) PWoH: 20 C − (55% M) 21 C + (57% M) |
≥ 4 × CU/month; no other illicit substance use; no CU on test day | GPT, HVLT-R, Phonemic and semantic verbal-fluency, WAIS-III DS, WAIS-III SS, TMA, Stroop, TMB | Executive function, attention, information-processing, learning, memory, verbal-fluency | No effects or interaction between cannabis and HIV on any cognitive functions; main effect of HIV on reaction time in visuospatial task (PWH slower than PWoH) |
| Cristiani et al. 2004 | PWH: 46 asympt. C − 79 asympt. C + 29 sympt. C − 55 sympt. C + PWoH: 24 C − 49 C + No gender data |
Past 12 month self-report; ≥ 1 × CU/week | WAIS-R, Selective Reminding Test, Wechsler Memory Scale-Revised, Verbal Concept Attainment Test, WCST, Verbal and figural fluency, TMA, TMB, GPT, PASAT | Global cognition [executive function, attention, information-processing, learning, memory, language, motor] | Cannabis related to worse overall impairment; driven by interaction effect on delayed memory; CU worsened delayed memory, particularly for those with more severe HIV disease) |
| Crook et al. 2021 | PWH: 91 past CUD (79% M) 47 past CUD + (69% M) |
Past CUD measured by DSM-IV interview; excluded current CUD | WAIS-III LNS, PASAT, TMA, WAIS-III DS, WAIS-III SS, TMB, WCST, COWAT, HVLT, BVMT, GPT | Executive function, informa-tion-processing, attention, learning, memory, verbal-fluency, motor | CUD + associated with better processing-speed, visual learning, memory, delayed recall, and dominant hand motor ability compared to CUD− |
| Dastgheyb et al. 2021 | 929 PWH 717 PWoH 100% F |
Self-reported CU | HVLT-R, LNS, TMA, TMB, Stroop, SDMT, COWAT; Animal fluency; GPT | Neuropsychological “profiles” created using the following domains: executive function, attention, learning, memory, language, motor | Current CU associated with lower likelihood of impairment “profile” consisting of learning, information-processing and executive function in women with HIV |
| Flannery et al. 2021 | PWH: 27 C − (67% M) 32 C + (91% M) PWoH: 22 C − (50% M) 28 C + (57% M) |
≥ 1 × CU/week for 3 months; ≥ 20 × in the past year | EAT (go/no-go motor inhibition paradigm) during MRI; Cognitive Failures Questionnaire | Executive function | Greater lifetime CU associated with worse inhibition-related neural activity in PWoH but not PWH |
| Flannery et al. 2022 | PWH: 28 C − (67% M) 32 C + (91% M) PWoH: 22 C − (50% M) 24 C + (57% M) |
≥ 1 × CU/week for 3 months; ≥ 20 × in the past year | WAIS-IV, WCST, IGT, HVLT-R, BVMT, GPT, MST | Executive function, attention, information-processing, learning, memory, and motor | CU was associated with increased resting state functional connectivity in regions associated with executive function in PWH and PWoH, but no effect of CU on cognitive functions |
| Flannery et al. 2022 | PWH: 28 C − (67% M) 32 C + (91% M) PWoH: 22 C − (50% M) 24 C + (57% M) |
≥ 1 × CU/week for 3 months; ≥ 20 × in the past year | EAT (go/no-go motor inhibition paradigm) during MRI | Executive function | CU was associated with better executive function in PWH (error-awareness) relative to PWH who do not use cannabis |
| Gomez et al. 2017 | 138 neurocognitively normal (NN) PWH/C − 67 NN PWH/C + 69 NCI PWH/C − 14 NCI PWH/C + (84–90% M/group) |
Patient chart review for the past cannabis abuse | Game of Dice Task (GDT), SDM, TMT-2/4, GPT, HVLT, WCST | Executive function | CU predicted worse executive function (GDT net score) |
| Gonzalez et al. 2011 | PWH: 25 C − 17C + PWoH: 21 C − 23 C + Mostly men no gender data; polysubstance users |
DSM-IV + Kreek-McHugh-Schluger-Kellog scale; lifetime history of cannabis dependence | Rotary pursuit task (RPT), Star mirror tracing task (SMT), weather prediction task (WPT) | Procedural learning of motor skills | History of cannabis dependence was not associated with procedural learning of motor skills; Motor skills performance on the RPT and SMT adversely affected among PWH with history of polysubstance dependence |
| Haney et al. 2005 | 15 PWH/C + with muscle loss (80% M) 15 PWH/C + no muscle loss (100% M) 100% current CU + ≥ 2×/ week for 4 weeks |
Acute dronabinol (0, 10, 20, 30 mg) or cannabis (0.0, 1.8%, 2.8%, 3.9% THC) | DSS, RAT, DAT, Immediate and a delayed DRT | Attention, learning, memory | CU associated with no effects on performance in either group compared to placebo; 20-mg dronabinol associated with worse memory and attention in clinical muscle loss group; 30-mg dronabinol was associated with worse memory in normal muscle mass group |
| Haney et al. 2007 | 10 PWH/C + (90% M): 100% current cannabis users (≥ 2 ×/week) for the past 4 weeks |
Chronic (4 day; QUID) dronabinol (0, 5, 10 mg) or cannabis (THC 0%, 2%, 3.9%) | DSS, RAT, DAT, immediate and a delayed DRT, 50-item visual analogue scale | Attention, information-processing, learning, memory | Cannabis or dronabinol was associated with no effects on cognitive performance compared to placebo |
| Heaton et al. 2023 | 402 PWH 76% M |
Past or present CUD | DVT, WAIS-III DS, WAIS-III SS, TMA, SMT, FMT, HCT, WCST, TMB, COWAT, CF, WAIS-R DS, WAIS-III LNS, PASAT, GPT | Global cognition [executive function, attention, information-processing, learning, verbal-fluency] | History of CUD was associated with worse neurocognitive decline (decrease from baseline after 12 years) |
| Kallianpur et al. 2020 | PWH (85% M): 4 C − 25 recent C + 23 remote C + PWoH (73%M): 11 C − 9 recent C + 35 remote C + |
Self-report of CU; recent use (use in the last 12 months), remote use (> 12 months ago); also reported 12-month CU frequency |
CalCAP, RAVLT, RCF Copy and Recall, TMA, TMB, WAIS-R, WAIS-III LNS, GPT, Verbal-fluency test, Animal and Boston Naming tests, Delis-Kaplan Executive Function System, Stroop, Timed gait | Executive function, information-processing, learning, memory | Occasional CU was associated with better executive function in PWH compared to non-use; duration of CU positively correlated with psychomotor speed and executive function in PWH but not PWoH; recent CU was associated with worse psychomotor speed compared to remote CU in PWoH |
| Lorkiewicz et al. 2018 | 215 PWH 65% M |
Lifetime CU history; current use: no. of days of use in the past 20 days); lifetime CU (no. of years use;≥3×/week) |
Montreal Cognitive Assessment (memory and attention) | Attention, memory | Current CU associated with no effects on cognitive function; no significant association between lifetime CU and cognitive function |
| Meade et al. 2018 | PWH: 29 C − (72% M) 20 C + (75% M) PWoH: 19 C + (68% M) 25 C − (68% M) |
≥ 4 days of CU in the past month and ≥ 1 year of regular CU | Stroop, Counting stroop (during fMRI) | Executive function | CU was associated with no effects on cognitive interference, but an HIV X cannabis interaction effect on fMRI activity in task-dependent brain regions. PWH/C + had the greatest activity in these regions |
| Murdoch et al. 2023 | PWH: 33 C + 12 C − /cocaine + 22 C + /cocaine + 43 C − /cocaine − (58–91% M/group) |
Lifetime history of regular use and ≥ 12 days of use in the past 90 days | PASAT, WAIS-IV DS, WAIS-IV LNS, TMA, TMB, WAIS-IV coding, Stroop, HVLT-R, CF, GPT | Global cognition [executive function, speed of information-processing, attention, learning, memory, verbal-fluency, motor function] | CU was associated with no effect on global score |
| Naveed et al. 2022 | 581 PWH-159 use cannabis 79% M |
Self-reported CU | TMA, TMB, WCST, WAIS-III DS, WAIS-III SS WAIS-III LNS, PASAT, BVMT-R, HVLT-R, COWAT, GPT | global cognition [executive function, speed of information-processing, attention and working memory, learning, memory, verbal-fluency, motor function] | Lifetime CU associated with worse neurocognitive decline |
| Okafor et al. 2019 | PWH: 498 C − 290 C + PWoH: 755 C − 377 C + 100% M |
Self-reported CU in the past 6 months; CU years = total days used during study; categorized “monthly,” “weekly,” “daily” | TMA, TMB, SDM | Executive functioning, attention, information-processing, psychomotor speed | Daily and monthly CU associated with worse processing-speed compared to non-users in PWH; no association between cumulative cannabis years and cognitive function in PWH men; each additional 5 years of CU was associated with worse in processing-speed in men without HIV |
| Rogers et al. 2023 | PWH (82–92% M): 187 Meth − /C − 68 Meth − /C + 82 PWH/Meth + /C − 135 Meth +/C + |
Timeline follow back and DSM-IV Interview for CUD | DVT, WAIS-III DS, WAIS-III SS, TMA, SMT, FMT, HCT, WCST, TMB, COWAT, CF, WAIS-R DS, WAIS-III LNS, PASAT, GPT | Global cognition [executive function, information-processing speed, learning, memory, verbal-fluency, motor]; individual subdomains also reported | Meth + /C + performed better than Meth + /C− in executive function, learning, memory, working memory and better than Meth − /C − on verbal-fluency, but worse than Meth − /C − in learning and memory. Meth − /C + performed better than Meth − /C − on executive function, learning, memory and working memory |
| Saloner et al. 2019 | All participants 50–64 years 734 PWH 123 PWoH PWH: 84% M PWoH: no gender distribution |
Lifetime CUD | CF, Letter fluency, PASAT, WAIS-III LNS, WAIS-III SS, WAIS-III DS, WMS-III spatial span, TMA, Stroop, WCST, TMB, Halstead Category Test, HVMT-R, BVMT-R, SMT, FMT, GPT | Global cognition, [executive function, attention, information-processing, learning, verbal-fluency, motor] | Higher rates of lifetime CUD among “super ager” PWH (participants with-above average GDS cognitive performance relative to their age) compared to cognitively impaired individuals PWH and cognitively normal PWH |
| Schantell et al. 2022 | 33 PWoH/C − (48 %M) 32 PWoH/C + (59% M) 17 PWH/C − (47% M) 18 PWH/C + (72% M) |
≥ 2 ×CU/week for 6 months; other substance use< 1 ×/month | Eriksen flanker task during magnetoencephalography | Executive function | PWH/C − had larger flanker interference effect than PWoH/C + and PWoH/C −; PWoH/C + had smaller flanker interference effect relative to PWH/C + |
| Schouten et al. 2016 | 74 PWoH 103 PWH 100% M |
Daily to monthly self-reported CU | CF, LF, TMB, WCST, Stroop, TMA, DS, SS, PASAT, LNS, RAVLT, Visual reproduction test, GPT | Global cognition [executive function, attention, information-processing, memory, fluency, and motor] | Cognitive impairment detected in 17% of men with HIV; CU was associated with worse cognitive performance among men with HIV |
| Skalski et al. 2018 | PWH (67–75%): 42 C − 12EC + 15LC + EC + = early CU onset LC + = late CU onset (before/after age 18 respectively) |
C + : ≥ 10 days/month ≥ 1 year or ≥ 3 ×/week of binging or problematic regular use | HVLT-R, BVMT-R, TMA, TMB, GPT, COWAT, PASAT | Attention, information-processing, learning, memory, verbal-fluency, psychomotor ability | EC + more likely to have worse learning and memory compared to C − (learning, memory) but more likely to have better attention/working memory. No differences between LC + and C − |
| Thames et al. 2017 | PWH (83% M): 24 PWH/C − 24 PWH/C+ PWoH (52% M): 16 PWoH/C − 13 PWoH/C+ |
Average CU amount (in grams) smoked/day times no. of days/week of CU in the past month | WTAR, TMA, Stroop, WAIS-IV LNS, WAIS-IV DS, WAIS-IV SS, COWAT, BVMT-R, HVLT-R, TMB | Global cognition, [executive function, information-processing, attention, learning, memory, verbal-fluency]; individual subdomains also reported | Higher levels of CU associated with worse global cognition in PWoH, no effect in PWH, driven by processing-speed and memory; simple effects: low CU (<1.43 g/week) was associated with worse global cognition in PWH vs. PWoH; no between group effects observed when CU > 1.43 g/week |
| Thames et al. 2016 | PWH: 14 C − (85% M) 30 light C +(95% M) 31 heavyC + (54% M) PWoH: 32 C − (45% M) 12 light C + (75% M) 10 heavy C + (100% M) |
Light C + :2–14 ×/week> 12 months; Heavy C +: 18–90 ×/week > 12 months |
WTAR, TMA, Stroop, WAIS − IV LNS, WAIS-IV DS, WAIS-IS SS, COWAT, BVMT-R, HVLT-R TMB | Executive function, attention, information-processing, learning, memory, verbal-fluency | Moderate to heavyC + associated with worse global cognition, processing-speed, learning/memory and executive function compared to LightC + and C − ; PWH/ HeavyC + performed worse in learning and memory compared to all groups, PWH/LightC + performed better in verbal-fluency than PWoH/LightC + |
| Wang et al. 2020 | PWH: 23 C − (96% M) 21 C + (95% M) PWoH: 24 C − (88% M) 22 C + (86% M) |
Chronic CU (> 3 ×/week for > 2 years) | RAVLT, RCFT, D-KEFS, Stroop, Trail-making (number-letter switching); WAIS-IV, Fluency test, COWAT, GPT | Executive function, attention, information-processing, learning, memory, design and verbal-fluency, motor | Nonsignificant trend indicating PWH/C + performed better than PWH/C- in executive function and verbal-fluency, executive function, and information processing |
| Watson et al. 2020 | PWH: 573 C −(80% M) 106 C + (92% M) PWoH: 229 C − (60% M) 44 C + (79% M) |
C + : history of CUD and/or CU in the past year | COWAT, CF, WCST, TMA, TMB, Stroop, WAIS-III DS, WAIS-III SS, WAIS-III LNS, HVLT-R, BVMT-R, PASAT, GPT | Executive function, attention, information processing, learning, verbal-fluency, motor | PWH/C + associated with better performance in verbal-fluency, learning; no effect of cannabis in PWoH on any cognitive domain |
| Watson et al. 2021 | PWH (82–95% M): 105 C− 62 moderate C + 31 dailyC + PWoH: 65 C −(68% M) |
Moderate C + : 3 days of CU/week to 3 × of use within the past 6 months | HVLT-R, BVMT-R, COWAT, VF, TMA, TMB, WCST, HCT, PASAT, GPT, WAIS-III DS, WAIS-III SS, WAIS-III LNS, Stroop | Executive function, attention, information-processing, learning, verbal-fluency, motor | PWH/Daily C + had better global cognition compared to PWH/moderate C + and PWH/C−(ns); Similar trend in verbal-fluency, attention/working memory, processing-speed, learning and motor skills |
| Watson et al. 2023 | PWH (81–92% M): 191 C− 83 Occasional C + 23 Frequent C + |
Occasional: 1 ×/week to< 1 ×/month Frequent: 2–7 days/week |
HVLT-R, BVMT-R, COWAT, VF, TMA, TMB, WCST, HCT, PASAT, GPT, WAIS-III DS, WAIS-III SS, WAIS-III LNS, Stroop | Executive function, attention, information-processing, learning, verbal-fluency, motor | Occasional, but not frequent, CU was associated with better global cognition; driven by better performance in attention (significant), verbal-fluency (ns), and learning (ns). Recent THC use was associated with worse global cognition, driven by worse memory |
AVLT, auditory verbal learning test; WAIS-III DS, Wechsler Adult Intelligence Scale Third Edition-III digit symbol; WAIS-III SS, WAIS-III symbol search; TMA, Trail Making Test Part A; SMT, Story Memory Test; FMT, Figure Memory Test; WCST, Wisconsin Card Sorting Task; TMB, Trail Making Test Part B; COWAT , Controlled Oral Word Association Test; CF, category fluency; WAIS-III LNS, WAIS-III letter-number sequencing; PASAT, paced auditory serial addition task; GPT, Grooved Pegboard Test; RAVLT, Rey Auditory Verbal Learning Test; EAT, error-awareness task; Stroop, Stroop Color Interference Test; SDM, symbol digit modalities; CalCAP, California Computerized Assessment Package; HVLT-R, Hopkins Verbal Learning Test-Revised; WAIS-R, Wechsler Adult Intelligence Scale-Revised; BVMT-R, Brief Visuospatial Memory Test-Revised; MST, Mirror Star Tracing Test; TMT-2, TMT-4, Trail Making Test 2 and 4; HCT, Halstead Category Test; WTAR , Wechsler Test of Adult Reading; RCFT, Rey-Osterreith Complex Figure Test; DSS, 3-min digit–symbol substitution task; RAT , 3-min repeated acquisition task; DAT, 10-min divided attention task; DRT, immediate and delayed digit-recall task; D-KEFS, Delis-Kaplan Executive Function System; SUD, substance use disorder; QUID, four times daily