A Research Guide for
HIV-Associated Neurocognitive Disorder

Understanding HAND: The Spectrum, the Criteria, and the Debate

"HAND" stands for HIV-Associated Neurocognitive Disorder. It is an umbrella term covering three conditions of increasing severity, defined in 2007 by an international working group meeting in Frascati, Italy — which is why they are universally called the Frascati criteria (Antinori and colleagues, Neurology, 2007;69:1789–1799).

Understanding these three categories — and understanding why they are now under serious revision — will help you interpret anything you are told about your own results.

The three Frascati categories, in plain language

1. Asymptomatic Neurocognitive Impairment (ANI)

The test scores are low, but daily life is unaffected. Specifically: performance at least 1 standard deviation below the mean in 2 or more cognitive domains, with no detectable impact on work, self-care, or independent living.

This is the largest category and the most controversial one. Someone with ANI is, by definition, functioning normally. They noticed nothing. Often, the only reason they know about it is that they were enrolled in a research study or given a screening test. Whether it is helpful to give such a person a diagnostic label at all is a live and legitimate question — some researchers argue ANI predicts later decline and should be flagged; others argue it mostly captures statistical noise and normal human variability, and that labelling people "impaired" causes anxiety, stigma, insurance problems, and self-doubt without offering any treatment.

What is known: one influential analysis (Grant and colleagues, Neurology 2014) found that people classified with ANI were at meaningfully increased risk of progressing to symptomatic impairment compared with people who tested normal. So the category is not meaningless. But it is also not a diagnosis of brain damage, and it is emphatically not dementia.

2. Mild Neurocognitive Disorder (MND)

The test scores are low, and it is starting to affect daily life — mildly. Same test threshold (at least 1 SD below the mean in 2+ domains), but now with at least mild interference in day-to-day functioning: taking longer to do familiar tasks, needing lists where you never used to, making more errors at work, finding complex activities more effortful.

People with MND are still independent. They drive, work, manage their own money and medications — but with more effort and more compensation than before. This is the category most people with genuine cognitive complaints fall into, and it is the category where practical strategies (see the Symptom & Daily-Function Management section) make the biggest difference to quality of life.

3. HIV-Associated Dementia (HAD)

Marked impairment on testing, and substantial interference with daily life. Specifically: performance at least 2 standard deviations below the mean in 2 or more domains, with marked functional impairment — needing help with medications, finances, transportation, or basic activities.

This is now rare. In the pre-ART era, HAD was an AIDS-defining illness that affected a large fraction of people with advanced disease, often progressed over months, and was frequently fatal. Today, in populations with good access to treatment, it affects on the order of 1–2% or fewer — and when it does occur, it occurs overwhelmingly in people who were diagnosed very late, who have not been on treatment, or who have had prolonged treatment interruptions.

It is also, importantly, partially reversible. Starting effective ART in someone with HAD frequently produces substantial cognitive improvement, sometimes dramatic. This is not a one-way door.

Why the criteria are being rewritten

Since about 2011, a steady stream of critiques has accumulated against the Frascati approach. The core problems:

• Over-diagnosis. Defining impairment as "1 SD below the mean on 2 of ~5–7 domains" catches a large number of perfectly healthy people. Analyses using simulated normal data have shown that the criteria can generate "impairment" rates approaching those reported in HIV cohorts — from noise alone. Gisslén, Price and Nilsson made this argument explicitly in a 2011 paper titled, pointedly, "The definition of HIV-associated neurocognitive disorders: are we overestimating the real prevalence?"
• The normative data problem. Cognitive test norms were largely developed in white, well-educated, English-speaking populations. Applying them to people with different educational histories, first languages, or cultural backgrounds systematically produces false "impairment." This is a serious equity issue and it disproportionately affects the populations most affected by HIV globally.
• Causal ambiguity. "HAND" implies HIV caused it. But the criteria only require that other causes are not better explanations — a much weaker standard. In practice, many people labelled with HAND have impairment driven by depression, substance use, vascular disease, or aging, with HIV playing a small or no role. Lumping all of these together as one disease obscures the mechanism and blocks progress on treatment.
• Legacy versus active injury. Someone who suffered brain injury during a period of untreated advanced HIV in 1996 and has been stable ever since is in a completely different situation from someone whose brain is being actively injured now. Frascati does not distinguish them. The clinical implications are entirely different.
• It was never meant for the clinic. Frascati was explicitly designed as a research nosology — a way to standardize study populations. It then drifted into clinical use, where it was never validated.

The 2023 international consensus (Nightingale et al.)

In June 2023, the International HIV-Cognition Working Group published consensus recommendations in Nature Reviews Neurology (Nightingale S, Ances B, Cinque P, et al. "Cognitive impairment in people living with HIV: consensus recommendations for a new approach." Nat Rev Neurol 2023;19:424–433). It is the single most important document for understanding where this field is going, and it was adopted by the European AIDS Clinical Society (EACS) in 2023.

The key changes, translated out of jargon:

• Symptoms matter again. Classification should combine cognitive symptoms (changes you or someone close to you has actually noticed), objective test performance, and abnormalities on neurological investigation — not test scores alone. This directly attacks the ANI problem: a person with no symptoms and no functional change should generally not be labelled as having a disorder.
• A new term for genuine HIV-caused injury: HIV-associated brain injury (HABI). This is reserved for cases where there is real evidence that HIV is the cause — not just a statistical association. It can be further characterized as active (ongoing injury, potentially responsive to intervention) or legacy (historical damage, now stable).
• Name the actual cause when you can. If someone's cognitive difficulty is being driven by depression, or alcohol, or cerebrovascular disease, the diagnosis should say so. Calling it "HAND" when the cause is untreated depression is not just imprecise — it delays the treatment that would help.
• Low test performance is not automatically brain injury. The consensus explicitly acknowledges that low scores may reflect social, educational, and language factors rather than any neurological process.
• Cognitive symptoms count even without functional impact. If you have noticed a change — even one that hasn't yet cost you anything practical — that is clinically meaningful and worth investigating.

What is actually happening in the brain

Briefly, and without overclaiming — because the honest answer is that this is not fully settled:

• HIV enters the brain early, carried across the blood-brain barrier inside infected immune cells (a "Trojan horse" mechanism), likely within days to weeks of infection.
• HIV does not directly infect neurons. It infects and persists in support cells — microglia (the brain's resident immune cells), perivascular macrophages, and possibly astrocytes. Neurons are damaged indirectly.
• The damage is largely inflammatory. Infected and activated immune cells in the brain release inflammatory signals and viral proteins (such as gp120 and Tat) that injure the connections between neurons — the synapses and dendrites — rather than killing the neurons outright. That is one reason improvement is possible: injured connections can sometimes recover.
• The brain is a reservoir. Even with fully suppressive ART and an undetectable blood viral load, HIV DNA persists in brain cells, and low-grade immune activation continues in the central nervous system in many people. This persistent, smouldering inflammation is the leading candidate explanation for why milder impairment continues to occur despite good treatment. Recent work has confirmed that intact, potentially replication-competent proviruses can be found in brain tissue even in virally suppressed people.
• The regions affected are typically the deep structures — basal ganglia and white matter — which is exactly why the symptom pattern is one of slowness and executive difficulty rather than the memory-first pattern of Alzheimer's disease.
• Vascular injury is now a major contributor. HIV accelerates small-vessel disease. As the population with HIV ages, cerebrovascular disease is becoming a bigger driver of cognitive change than the virus itself in many people.

Diagnosis: Cognitive Testing and Ruling Out Other Causes

A proper evaluation of new cognitive symptoms in a person with HIV has two halves, and the second half is more important than the first.

Half one: establish objectively whether there is a measurable cognitive problem, and characterize its pattern.
Half two: systematically rule out every treatable cause that is not HIV.

A clinic that does only half one and then hands you a HAND diagnosis has done half a job. The most common failure mode in this whole field is attributing cognitive symptoms to HIV without excluding the things that are far more likely and far more fixable.

Step 1: Screening

Most HIV clinics start with a brief screening tool. You should know what these are and what they are worth.

• The International HIV Dementia Scale (IHDS). A short, three-part test (memory registration, timed finger tapping, timed alternating hand sequence, memory recall) scored out of 12. It was designed to be usable anywhere in the world, including settings without neuropsychologists, and it deliberately emphasizes motor speed — which is affected early in HIV. It is reasonably good at picking up more severe impairment and poor at picking up mild impairment.
• The Montreal Cognitive Assessment (MoCA). Widely used, 30 points, takes about 10 minutes. It was designed for Alzheimer's-type mild cognitive impairment, and it is imperfect for the HIV pattern — it under-weights processing speed. Still useful, and increasingly common.
• Simple symptom questions. The EACS guidelines have long recommended asking three direct questions about memory, attention, and mental slowing. If you answer yes to any, that triggers further evaluation. This "symptom-first" approach is very much in line with the 2023 consensus.
• Brief computerized batteries (such as CogState and similar tools) are used in some centers.

Important caveat about screening: guidelines do not currently recommend screening everyone with HIV who has no symptoms. Universal screening of asymptomatic people generates a lot of false positives, a lot of anxiety, and very little actionable information — because there is no treatment to offer someone with a low score and no symptoms beyond what you would already be doing (suppress the virus, manage risk factors). EACS recommends screening people who have symptoms, or who have specific risk features. If you are worried, that is a symptom, and you qualify.

Step 2: Formal neuropsychological testing

If screening is abnormal or symptoms persist, the next step is a full neuropsychological evaluation — a multi-hour session with a neuropsychologist involving pencil-and-paper and computerized tests across multiple cognitive domains.

The Frascati framework requires assessment of at least five domains. Typically these include:

• Verbal fluency / language — e.g., naming as many animals as you can in 60 seconds
• Attention and working memory — e.g., repeating digit sequences forwards and backwards
• Speed of information processing — e.g., symbol-digit substitution tasks; this domain is often the earliest and most affected in HIV
• Executive function — e.g., trail-making tests, card sorting, tasks requiring rule-switching
• Learning and memory — word lists and figure recall, tested both immediately and after a delay, with and without cues (the cueing distinction is what separates a retrieval problem from an encoding problem)
• Motor skills — e.g., grooved pegboard, finger tapping
• Sensory-perceptual function in some batteries

Your scores are converted to standardized scores by comparing them to a normative sample. This is the step where things can go wrong. The quality of the norms determines the validity of the result. Good practice uses demographically corrected norms (adjusted for age, education, sex, and ideally race/ethnicity and language). Poor practice uses off-the-shelf norms from a population unlike you. If you were educated outside the country you are being tested in, or English is not your first language, or your schooling was interrupted, ask explicitly what norms are being used.

Step 3: The confounder workup — this is the important part

This is what a rigorous evaluation should cover. Use this as a checklist and ask about anything that hasn't been addressed.

Mood and mental health

• Depression. The overlap between depression and cognitive impairment is enormous. Depression causes slowed thinking, poor concentration, memory complaints, apathy, and loss of drive — an almost perfect mimic of the HIV pattern. Depression is also more common in people with HIV than in the general population. It should be screened for formally (with a validated tool such as the PHQ-9) in every single person who presents with cognitive complaints. And critically: depression is treatable, and treating it often improves the cognitive symptoms.
• Anxiety and PTSD. Both impair attention and working memory. Rates of trauma exposure are elevated in many populations affected by HIV.
• Apathy syndrome. Sometimes present without depressed mood; can be a direct consequence of frontal-subcortical dysfunction, but should not be assumed to be.

Sleep

• Obstructive sleep apnea. Common, under-diagnosed, and a potent cause of daytime cognitive dysfunction. Treatable with CPAP, with real cognitive benefit. If you snore, are told you stop breathing, wake unrefreshed, or fall asleep during the day, ask for a sleep study.
• Insomnia. Extremely common in people with HIV. Chronic sleep restriction reliably degrades attention, processing speed, and working memory. Cognitive behavioral therapy for insomnia (CBT-I) works and should be tried before sedatives — sedatives themselves impair cognition.

Substances

• Alcohol. Chronic heavy use causes cognitive impairment directly and via thiamine deficiency. This is a huge and frequently missed contributor.
• Methamphetamine and cocaine. Both cause lasting cognitive deficits and both interact badly with HIV in the brain. Methamphetamine use in particular is associated with substantially worse cognitive outcomes.
• Opioids and benzodiazepines. Directly sedating and cognitively impairing.
• Cannabis. Heavy regular use impairs attention and memory, particularly in younger users. Effects are largely reversible with sustained abstinence.

Medications

A careful medication review should look for:

• Anticholinergic drugs — older antihistamines (diphenhydramine, found in many OTC sleep aids), certain bladder medications, tricyclic antidepressants, some antipsychotics. Cumulative anticholinergic burden is a well-established cause of cognitive impairment and should be actively reduced.
• Benzodiazepines and "Z-drugs" (zolpidem etc.).
• Opioids.
• Some anticonvulsants (topiramate is notorious for word-finding difficulty).
• Efavirenz — the antiretroviral most consistently associated with CNS side effects: vivid dreams, dizziness, mood change, and in some people cognitive complaints. Most modern regimens no longer contain it, but many people who have been in care for a long time were exposed to it, and a few remain on it.
• Dolutegravir and other integrase inhibitors — associated in some studies with insomnia, mood changes, and neuropsychiatric side effects in a minority of people. The data are mixed, effects are usually mild, and the drugs are excellent antiretrovirals. But if you developed insomnia, anxiety, or brain fog shortly after starting one, that is worth raising.

Blood tests

A standard workup should include:

• Complete blood count, kidney and liver function, electrolytes
• Thyroid function (TSH) — hypothyroidism causes slowed thinking and is trivially treatable
• Vitamin B12 (and often folate) — deficiency causes cognitive impairment and is treatable; it is also more common in people with HIV
• Syphilis serology — neurosyphilis is a great mimic, is more common in people with HIV, and is curable with antibiotics. This test is not optional.
• Hepatitis C — associated with cognitive impairment, and now curable
• HbA1c / glucose — diabetes and insulin resistance affect cognition
• Lipids
• Testosterone in men with fatigue, low mood, and low drive
• CD4 count and plasma HIV RNA (viral load) — obviously

Brain imaging

An MRI of the brain is usually indicated when cognitive symptoms are significant, new, or progressing. It is looking for:

• Alternative diagnoses — tumors, CNS lymphoma, opportunistic infections (toxoplasmosis, progressive multifocal leukoencephalopathy), strokes, hydrocephalus
• Vascular disease — small-vessel white matter changes, old silent strokes, microbleeds
• Atrophy patterns — generalized shrinkage, particularly of deep structures, is common in HIV-associated injury; a specific pattern of hippocampal atrophy might point toward Alzheimer's disease instead

Note that MRI in HIV-associated impairment is often non-specific — it frequently shows some white matter change and atrophy without giving a definitive answer. Its main value is excluding other things. That is still enormously worthwhile.

Lumbar puncture (spinal tap)

This is where the most specific, most actionable finding in this entire field lives: CSF viral escape.

A lumbar puncture involves inserting a thin needle between the vertebrae in the lower back to sample the cerebrospinal fluid (CSF) that surrounds the brain and spinal cord. Done properly it takes 20–30 minutes, is uncomfortable rather than agonizing, and carries a modest risk of a temporary headache afterwards.

When it should be considered: the European AIDS Clinical Society and the British HIV Association both recommend sampling CSF in people who have cognitive impairment and an undetectable plasma viral load, when no other cause has been found. In practice, the threshold should be low for anyone with new, progressive, or unexplained neurological symptoms.

What it can find:

• CSF viral escape — HIV detectable in the CSF despite an undetectable blood viral load, or CSF viral load more than 1 log₁₀ higher than plasma. This means HIV is replicating independently in a compartment your blood tests cannot see. The virus there may have developed drug resistance mutations that differ from those in your blood. Reported frequency varies widely — under 10% in people without symptoms, and up to around 25% in people who are neurologically symptomatic. It is genuinely treatable: CSF resistance genotyping guides a regimen change, and neurological improvement often follows.
• Opportunistic CNS infections — cryptococcus, toxoplasma, JC virus (PML), CMV, tuberculosis
• Neurosyphilis
• CNS lymphoma (via cytology and flow cytometry)
• Alzheimer's biomarkers (amyloid and tau) in older individuals where that is a consideration — though interpretation in people with HIV is an active research question, not yet settled clinical practice

Genetics and Risk Factors

Let's start with the practical bottom line, because this section is where a lot of people go looking for an explanation and find mostly uncertainty.

What the genetics research has actually looked at

• APOE ε4. This is the best-known genetic risk factor for Alzheimer's disease. Studies have asked whether it also increases the risk of cognitive impairment in people with HIV. The results have been inconsistent — some studies found an association, particularly in older people; others found none. It is not recommended as a clinical test, and knowing your APOE status would not change your management.
• CCR5 and CCR2 variants. These are the co-receptors HIV uses to enter cells. Variants have been studied for effects on CNS disease with mixed results.
• MBL, TNF-α, and other immune/inflammatory gene variants. Studied; associations reported; none replicated robustly enough to be clinically useful.
• Host genetic ancestry. Some cohort analyses have found differences by ancestry, but these are extremely difficult to disentangle from differences in access to care, timing of diagnosis, socioeconomic factors, education quality, and the appropriateness of cognitive test norms. Interpreting them as biological is not currently justified.

The honest summary: no gene has emerged as a strong, reproducible determinant. This is likely because the condition is not a single disease with a single mechanism — it is a common endpoint of many different processes.

The risk factors that actually matter

Here is the same information organized by how much you can do about it.

Not modifiable now (but historically important)

• Nadir CD4 count. The lowest your CD4 count ever fell. Below 200 is a consistent risk marker; below 100 more so. This reflects a period when your immune system was overwhelmed and injury may have occurred. You cannot change it. But it should not be read as a prophecy — many people with a very low nadir have entirely normal cognition decades later.
• Duration of untreated infection.
• History of an AIDS-defining illness, particularly a CNS one.
• Age. Rising, unavoidably, and increasingly the dominant factor.

Highly modifiable — and this is where your effort should go

1. Viral suppression. Taking ART consistently. This is the whole ballgame. Interruptions are damaging.
2. Blood pressure. High blood pressure is the strongest modifiable risk factor for cognitive decline in the general population and there is no reason it is weaker in people with HIV. Get it measured properly and treated to target. This is arguably the second most important thing after ART adherence.
3. Diabetes and insulin resistance. Poorly controlled diabetes damages small blood vessels, including in the brain.
4. Cholesterol. Note that some antiretrovirals affect lipids; this is a conversation to have with your provider.
5. Smoking. People with HIV smoke at higher rates than the general population, and smoking is a potent vascular risk factor. In many cohorts, smoking now costs people with HIV more years of life than HIV does. Quitting is the highest-yield health intervention available to most smokers.
6. Alcohol. Reducing heavy drinking improves cognition.
7. Methamphetamine and stimulants. Cessation matters enormously.
8. Depression. Treating it improves cognitive symptoms and adherence simultaneously.
9. Physical inactivity. Aerobic exercise has the best evidence of any lifestyle factor for cognition in the general population, and small studies in people with HIV are encouraging.
10. Hepatitis C. Now curable with 8–12 weeks of direct-acting antivirals. If you have it, treat it.
11. Untreated hearing loss. Increasingly recognized as a significant, modifiable contributor to cognitive decline in the general population. Get your hearing tested. Hearing aids are not a vanity item.
12. Social isolation. A real risk factor, and one that HIV-related stigma actively worsens.
13. Obstructive sleep apnea. Treatable, with cognitive benefit.

Managing HAND: The Central Role of ART (and the Truth About CNS Penetration)

This is the section people skip to. It is also, in a sense, the shortest one to summarize:

Part 1: Antiretroviral therapy — the foundation

Effective ART, taken consistently, achieving and maintaining an undetectable plasma viral load, is the single most important thing you can do for your brain. The evidence for this is not subtle:

• Starting ART improves cognition. In the ACTG A5199 international study (NCT00096824), which followed people starting first-line ART across seven resource-limited countries, the odds of neurocognitive impairment fell by roughly 12% for every 24 weeks on treatment. Moderate and severe impairment fell substantially. This is a real, measurable, clinically important benefit.
• HIV-associated dementia largely disappeared when combination ART became widely available. The near-elimination of a formerly common, devastating, AIDS-defining illness is about as strong a piece of evidence as medicine ever produces.
• Treatment interruptions are harmful. Periods off treatment allow viral rebound, immune activation, and CNS injury.

So: which regimen? For the overwhelming majority of people, the answer is "the one that works, that you can tolerate, and that you will actually take." Modern first-line regimens — typically an integrase inhibitor (dolutegravir or bictegravir) with two nucleoside analogues — are potent, well tolerated, one pill once a day, and have a high barrier to resistance. Adherence beats theoretical pharmacology.

Part 2: The CNS penetration (CPE) question — what the evidence actually shows

You will encounter this idea online and possibly from a clinician: antiretroviral drugs differ in how well they cross the blood-brain barrier; therefore, if you have cognitive symptoms, you should switch to drugs that get into the brain better.

The idea is intuitive and biologically reasonable. A ranking system exists — the CNS Penetration-Effectiveness (CPE) score, developed by Letendre and colleagues and published in Archives of Neurology in 2008 — which assigns each antiretroviral a score based on its chemistry, CSF concentrations, and observed effect on CSF viral load. Add up the scores for a regimen and you get a total. Higher is supposed to be better for the brain.

It was tested. It did not work.

• The CIT trial (Ellis and colleagues, Clinical Infectious Diseases 2014;58:1015–1022; NCT00624195) randomized people with HIV-associated neurocognitive disorder to a CNS-targeted ART strategy versus a non-CNS-targeted one. The conclusion, stated plainly by the authors: no evidence of neurocognitive benefit for the CNS-targeted strategy. The trial was stopped early. A small benefit in a subgroup could not be excluded, but the primary result was negative.
• The A5324 / InMIND trial (Letendre and colleagues, Clinical Infectious Diseases 2023;77:866–874; NCT02519777) took a different approach: rather than switching, it added drugs with good CNS distribution — dolutegravir with or without maraviroc — on top of an existing suppressive regimen, versus placebo, in 191 people with cognitive impairment and an undetectable viral load. It ran for 96 weeks and it was double-blind and placebo-controlled. Result: cognitive performance, depressive symptoms, and daily functioning all improved over time — equally in every arm, including placebo. Intensification added nothing. Notably, the fact that everyone improved is itself informative: it tells you how much of measured "improvement" in open-label studies is practice effect, regression to the mean, and the benefit of simply being in a study with good care and attention.
• Observational data are inconsistent. Some cohort studies find associations between higher-CPE regimens and better outcomes; others find no association; at least one large prospective analysis found a negative association with HIV dementia risk. Observational data on this question are heavily confounded — sicker people get switched to different regimens.

Part 3: When a regimen change IS appropriate

There are legitimate reasons to change antiretrovirals in someone with cognitive symptoms. They are specific:

1. CSF viral escape. If a lumbar puncture shows HIV replicating in the CSF despite plasma suppression, the regimen should be changed — guided by resistance genotyping of the CSF virus, which may show different mutations than the blood virus. BHIVA and EACS guidance suggests avoiding two-drug regimens in this situation, generally including a dual nucleoside backbone, and considering twice-daily dolutegravir. This is one of the few situations where a targeted regimen change can produce genuine neurological recovery. It requires a specialist.
2. Suspected drug neurotoxicity. If your symptoms began or clearly worsened after starting a specific drug — classically efavirenz, sometimes an integrase inhibitor — switching to a different agent is entirely reasonable and often helps. Efavirenz in particular is now largely avoided in new regimens for exactly this reason. This is not the same as CPE-guided switching; it is de-prescribing a suspected offender.
3. Virological failure. If your viral load is not suppressed, the regimen needs to change — for your whole body, brain included.
4. Intolerable side effects affecting adherence. A drug you don't take doesn't work.

What is not a good reason to switch: a CPE score, a magazine article, or an internet forum.

Part 4: Vascular and metabolic risk factor control

This is where a large and growing share of the achievable benefit now lies, and it is systematically under-emphasized in HIV clinics that are focused on virology.

• Blood pressure. Get it to target. Discuss what your target should be. This is likely the highest-yield non-ART intervention available.
• Lipids and statins. Cardiovascular prevention in people with HIV has been an active area of research, and practice has shifted toward broader statin use in this population. Ask your provider specifically whether a statin is indicated for you.
• Diabetes. Control it.
• Smoking cessation. The highest-value intervention for most smokers with HIV, full stop.
• Weight and physical activity. Aerobic exercise is the best-evidenced lifestyle intervention for cognition in general. Even modest amounts help.

Part 5: Treating the confounders you found

• Depression: treat it properly, with medication, therapy, or both. Expect 6–8 weeks before judging the effect. Choose antidepressants with attention to interactions with your ART regimen — your HIV pharmacist is the right person to ask.
• Sleep apnea: CPAP or an alternative. Actually use it.
• Insomnia: cognitive behavioral therapy for insomnia (CBT-I) first. Sedatives worsen cognition.
• Substance use: evidence-based treatment. Methamphetamine and alcohol cessation both produce measurable cognitive recovery over months.
• Thyroid, B12, syphilis, hepatitis C: treat what you find.
• Deprescribing: remove anticholinergics, minimize benzodiazepines, review everything.

Cancer and Other Comorbidities: What Else Can Affect the Brain

This section exists because cognitive symptoms in a person with HIV can have causes that have nothing to do with the HIV itself — and some of them are urgent.

Cancer-related causes

• Primary CNS lymphoma. An AIDS-defining cancer, strongly associated with Epstein-Barr virus and with advanced immunosuppression (usually CD4 under 50–100). It has become much less common in the ART era but has not disappeared. It presents with progressive cognitive decline, personality change, focal neurological deficits, or seizures over weeks. This is why an MRI matters when symptoms are progressive. It is treatable, and restoring immune function with ART is a central part of treatment.
• Systemic lymphoma with CNS involvement.
• Metastatic cancer — people with HIV have elevated rates of several non-AIDS-defining cancers, particularly lung cancer (driven substantially by smoking rates).
• "Chemo brain." If you are receiving or have received chemotherapy for any cancer, cancer-related cognitive impairment is a real, well-described phenomenon and is a completely different problem from HIV-associated impairment. It should be recognized as such rather than folded into an HIV diagnosis.

Opportunistic infections of the brain

These are now uncommon in people on treatment with good CD4 counts, but they are the reason nobody should get a HAND diagnosis without imaging when symptoms are progressive.

• Cryptococcal meningitis — headache, fever, confusion, sometimes vision changes. A leading cause of death in advanced HIV globally.
• Cerebral toxoplasmosis — focal deficits, seizures, headache; characteristic ring-enhancing lesions on MRI.
• Progressive multifocal leukoencephalopathy (PML) — caused by JC virus; progressive focal neurological deficits and cognitive decline; treatment is immune restoration with ART.
• CMV encephalitis, tuberculous meningitis, neurosyphilis.

Immune reconstitution inflammatory syndrome (IRIS)

An important and counterintuitive one. When someone with a very low CD4 count starts ART, the recovering immune system can mount a vigorous inflammatory response against organisms it previously ignored. Paradoxically, the person can get sicker in the weeks after starting effective treatment. When this happens in the brain, it can look like acute cognitive worsening, headache, seizures, or focal deficits.

This does not mean ART is failing or should be stopped. It usually means it is working. But it requires urgent specialist evaluation, and treatment may include corticosteroids alongside continued ART. If you or someone you care for gets neurologically worse within the first weeks to months of starting or restarting ART from a low CD4 count, seek care promptly and specifically mention the timing.

The aging picture

People with HIV are now living into old age in large numbers. This means:

• Alzheimer's disease will occur in people with HIV at the ages it occurs in everyone. Whether HIV increases the risk is unresolved. If the pattern looks amnestic, it should be evaluated on its own terms.
• Vascular cognitive impairment is likely a growing contributor, given the accelerated vascular disease seen in HIV.
• Mixed pathology is probably the norm, not the exception, in older people with HIV and cognitive symptoms. Someone can have legacy HIV-related injury and small-vessel disease and early Alzheimer's changes and untreated depression, all at once. Insisting on a single unifying diagnosis is often the wrong instinct.
• Polypharmacy rises with age, and with it the risk of medication-induced cognitive impairment.

Symptom and Daily-Function Management

Everything above is about causes. This section is about living with the symptoms that remain after the causes have been addressed — which, for many people, is the part that actually determines quality of life.

The evidence base here is thinner than we would like, and honesty requires saying so. But the principles are drawn from cognitive rehabilitation generally, and they are safe, cheap, and frequently effective.

The core principle: externalize

The most reliable strategy in cognitive rehabilitation is not "train your brain to remember better." It is: stop asking your brain to do things that a piece of paper or a phone can do for it. Working memory is the resource under strain. Anything you can move out of your head and into the environment frees capacity for the things that actually need thinking.

• One calendar. One list. One place. Multiple systems fail. Pick a single calendar (phone or paper, whichever you'll actually use) and put everything in it.
• Write it down at the moment of the thought. Not later.
• A landing zone. One dish, hook, or bowl by the door for keys, wallet, phone, glasses. Nothing else goes there. This single change eliminates an enormous amount of daily frustration.
• Alarms for everything time-based. Medications, appointments, when to leave the house.
• Labels. On cupboards, drawers, and containers if needed. There is no shame in this.
• Automate the recurring. Automatic bill payment. Automatic pharmacy refills. Automatic transfers. Every recurring task you automate is one fewer thing to remember.

Managing mental fatigue and processing speed

• Do the hardest thing when you're sharpest. Most people have a window — often mid-morning. Protect it.
• One thing at a time. Multitasking is precisely the ability that is impaired. Doing tasks sequentially is not a failure; it is an adaptation.
• Reduce the noise. Background TV, open-plan offices, and busy restaurants disproportionately degrade performance when attention is impaired. Ask for a quiet table. Use noise-cancelling headphones.
• Build in rest. Cognitive fatigue is real and it accumulates. Short scheduled breaks work better than pushing through and crashing.
• Slow down deliberately. Accuracy usually survives when speed is sacrificed. Many people can do everything they used to do — just not as fast. Adjusting expectations rather than making errors is a legitimate strategy.

Conversation and social strategies

• It is fine to say "give me a second" or "say that again."
• It is fine to ask people to slow down or to write things down for you.
• Repeat back important information to check you got it right.
• Take notes during medical appointments, or bring someone who does. Or ask if you can record it — most clinicians will agree.
• If you are struggling in group settings, consider whether it is actually a hearing problem. Difficulty following conversation in noisy rooms is the classic presentation of hearing loss, and it masquerades as cognitive decline constantly.

Exercise, sleep, and diet

• Aerobic exercise has the strongest evidence of any lifestyle factor for cognition. Small studies in people with HIV are encouraging, though not definitive. Aim for whatever you can sustain — walking counts. The dose that helps is the dose you'll do.
• Resistance training supports mobility, mood, and metabolic health.
• Sleep is non-negotiable. Fix apnea. Fix insomnia. Protect a regular schedule.
• Diet: a Mediterranean-style pattern has the best (though still imperfect) evidence for brain health. There is no HIV-specific brain diet, and anyone selling you one is selling you something.
• Alcohol: less is better. For someone with cognitive symptoms, a trial period of complete abstinence for 4–8 weeks is one of the most informative experiments available — and costs nothing.

Work, driving, and legal capacity

• Work. Many people with mild impairment work successfully with accommodations: written instructions rather than verbal, a quieter workspace, more time for complex tasks, fewer simultaneous projects. In the United States, HIV is a protected disability under the Americans with Disabilities Act, and reasonable accommodations can be requested. You are not required to disclose your HIV status to request accommodations for a cognitive condition. Talk to an employment lawyer or an HIV legal services organization before disclosing anything.
• Driving. This is emotionally loaded and needs to be handled honestly. Mild impairment does not automatically mean you cannot drive. Significant impairment of attention, reaction time, and judgment does. If there is real concern, a formal driving assessment (often available through occupational therapy or a rehabilitation center) is far better than a family argument. It gives an objective answer, and it protects both the person's autonomy and other people's safety.
• Financial and legal planning. Do this early, while capacity is unquestioned. Advance directives, healthcare proxy, durable power of attorney. This is not a concession that things will get worse — it is basic adult planning that everyone should do, and doing it early means it is done on your terms.

Surveillance and Monitoring: What to Track Over Time

A single cognitive test result is a snapshot. It tells you where you are, not where you are going. The trajectory is what matters, and the only way to know your trajectory is to measure more than once.

This is one of the most under-used tools in the field. If a person is told they have "mild impairment" and never retested, nobody — including them — learns anything. If they are retested in a year and are unchanged, that is enormously reassuring. If they are worse, that triggers a fresh search for causes.

What should be monitored, and roughly how often

What to track yourself

You do not need an app or a device. A notebook works. Track:

• Missed doses — count them honestly, and note why
• Specific incidents — "forgot appointment on the 12th"; "got lost driving to my sister's"
• Good days and bad days — and what was different about them (sleep, alcohol, stress). Fluctuation is diagnostic information: purely structural brain injury does not fluctuate day to day; mood, sleep, and substances do.
• Sleep hours and quality
• Alcohol and other substances
• Mood — even a simple 1–10 daily rating
• Anything that changed — a new medication, a new stressor, a bereavement

Clinical Trials: What's Being Studied and How to Find One

Let's set expectations honestly before anything else. The pipeline for HIV-associated cognitive impairment is thin. It is not like Alzheimer's disease, where there are dozens of large late-stage trials running at any moment. Most of the drug trials in this field are small, early-phase, or mechanistic. Several of the biggest, best-designed studies have been negative.

That said — trials matter, participating is often a way to get an unusually thorough evaluation, and the field is not standing still.

Landmark trials you should know about (mostly because of what they found)

What is being studied now

Current research directions, described in prose because trial identifiers change and specific studies open and close:

• Anti-inflammatory strategies. The leading hypothesis for why mild impairment persists despite viral suppression is ongoing low-grade neuroinflammation. Agents targeting this — including repurposed drugs with anti-inflammatory or antioxidant properties — remain of interest. So far, none has succeeded.
• Statins. Beyond cardiovascular benefit, statins have anti-inflammatory effects and have been studied for cognitive endpoints in HIV (a trial registered as NCT01600170 has been cited in the literature for this purpose). Given the recent expansion of statin use in people with HIV for cardiovascular prevention, any cognitive benefit would be a bonus rather than the primary reason to take one.
• Cognitive rehabilitation and computerized cognitive training. Active area. The consistent finding across cognitive training research generally is that you get better at the trained task; whether that generalizes is the hard part.
• Aerobic exercise interventions. Promising, safe, and cheap. Several small studies; larger ones needed.
• Non-invasive brain stimulation (tDCS, transcranial magnetic stimulation). Early, small, exploratory.
• Biomarker and imaging studies. A large fraction of current NeuroHIV research is not testing treatments at all — it is trying to find reliable markers of who has active brain injury versus legacy damage versus nothing. This is arguably the most important work in the field, because until we can identify the right people, we cannot design a trial that would show a benefit even if the drug worked.
• Alzheimer's biomarker studies in people with HIV. Cross-cohort work comparing amyloid and tau markers in people with and without HIV, to answer whether HIV increases Alzheimer's risk and to distinguish the two conditions in older people. Genuinely important, genuinely unresolved.
• HIV reservoir and cure research. If the CNS reservoir is what drives persistent inflammation, then cure strategies would in principle address the root cause. This is a long horizon.
• Long-acting injectable ART and its effects on adherence and, indirectly, on cognition in people who struggle with pills.

How to search for trials yourself

Failed and De-Adopted Approaches: What Has NOT Worked

This section exists because you will encounter these claims — in forums, in old articles, from well-meaning friends, and occasionally from clinicians working from outdated information. You deserve to know what has actually been tested and what happened.

Drugs that were tested and failed

De-adopted strategies

1. CPE-guided regimen switching

Covered in detail in the Treatment section. Briefly: tested in a randomized trial (CIT, NCT00624195), negative. Tested again as intensification (A5324, NCT02519777), negative. Recent expert reviews state that CPE scores should no longer be used to guide regimen selection. This was a reasonable idea that did not survive contact with evidence. If a clinician proposes switching your working regimen purely on CPE grounds, ask them to explain the CIT and A5324 results.

2. Universal cognitive screening of asymptomatic people

Once recommended more broadly; now scaled back in most guidance. The reason: screening asymptomatic people generates a very large number of false positives, causes real anxiety and stigma, and produces no actionable management change — because there is no treatment beyond what you would already be doing. Symptom-triggered assessment is now favored.

3. The "ANI" label itself, arguably

The 2023 international consensus effectively argues against routinely diagnosing asymptomatic people with a disorder on the basis of test scores alone. This is a de-adoption in progress, and not universally accepted — there is a real counter-argument that ANI predicts later decline and should be captured. But the direction of travel is clear.

Things sold to patients that have no evidence base

Said plainly, because you will encounter them:

• "Brain-boosting" supplement stacks marketed to people with HIV. No supplement has been shown in a randomized trial to improve cognition in HIV. Several carry real interaction risks with antiretrovirals (see the Complementary Approaches section).
• Chelation therapy. No rationale, no evidence, real harm.
• Hyperbaric oxygen therapy. No evidence for this indication.
• Stem cell "clinics" offering unproven infusions. Expensive, unregulated, dangerous. Not the same thing as the legitimate stem cell transplant research that has produced a handful of HIV cures in people who needed transplants for cancer.
• Anything claiming to "detox" the brain.
• Anything that requires you to stop your antiretrovirals. This is the brightest line in this entire guide. Any practitioner who suggests stopping ART is endangering your life and your brain. Leave.

Devices and Practical Tools

There is no FDA-cleared device that treats HIV-associated cognitive impairment. What exists is a set of assistive technologies — some medical, most not — that reduce the burden on a strained cognitive system. Nearly all of them are cheap or free.

Medication management

• Weekly/monthly pill organizers with AM/PM compartments. The cheapest, most effective assistive device in existence.
• Blister packaging / bubble packs from your pharmacy — pre-sorted by day and time. Ask; many pharmacies do this free, and it dramatically reduces error.
• Automatic pill dispensers that lock, alarm, and release only the correct dose at the correct time. Some alert a caregiver if a dose is missed. Useful when impairment is more significant.
• Medication reminder apps with escalating alerts. Free options are fine.
• Automatic pharmacy refills and mail delivery — removes one entire failure point.
• Long-acting injectable ART — not a device, but the most powerful "adherence technology" available. Ask about it.

Memory, planning, and orientation

• Smartphone calendar with alerts set for the day before and an hour before.
• Voice assistants ("remind me to take my medication at 8 pm"). Hands-free capture of a thought at the moment it occurs is genuinely powerful.
• Voice memo apps — faster than writing.
• A large whiteboard in the kitchen with the week laid out. Low-tech, high-yield.
• GPS navigation, used even on familiar routes. There is no prize for navigating from memory.
• Bluetooth trackers for keys, wallet, and phone.
• Photo contacts on the phone.
• Automatic bill payment.

Sensory support — do not skip this

• Hearing aids. Untreated hearing loss is one of the most significant modifiable risk factors for cognitive decline in the general population. Over-the-counter hearing aids are now available in the US at a fraction of previous costs. If you strain to follow conversation in a restaurant, get your hearing tested. This may be the highest-yield "device" in this entire section.
• Up-to-date glasses. Straining to see consumes cognitive resources.
• Good lighting. Genuinely reduces error and fall risk.

Safety

• Stove auto-shutoff devices if leaving the stove on has become a concern.
• Personal emergency response systems (wearable alert buttons) if falls or getting lost are a risk.
• Location-sharing with a trusted person — by consent, not surveillance.
• Home safety modifications — grab bars, removing trip hazards, night lights. An occupational therapist can do a home assessment.

Complementary and Alternative Approaches: A Tiered Evidence Assessment

Tier 1: Supported by reasonable evidence, safe, recommended

Tier 2: Plausible, low-risk, but not proven for this condition

Tier 3: Insufficient evidence, and/or real safety concerns

Specialty Center Directory

Mountain West and Utah

A candid note for Utah residents. Utah does not have a dedicated NeuroHIV research program of the kind found at UCSD, Johns Hopkins, or Mount Sinai. What it does have is a strong, long-established academic HIV clinic with the ability to refer to university neurology and neuropsychology. For most people, that is sufficient — the workup described in this guide does not require a specialized center. If you have a complex or refractory problem (suspected CSF escape, progressive decline despite full workup), it is reasonable to ask about a second opinion or referral to one of the national centers below, some of which offer remote consultation.

US National centers with NeuroHIV expertise

Veterans

The Department of Veterans Affairs is the largest single provider of HIV care in the United States, and it has genuine strengths here: integrated care, an excellent pharmacy benefit, established HIV clinics at most VA medical centers, and a large research infrastructure (the Veterans Aging Cohort Study has produced a great deal of the evidence on aging with HIV, including cognitive and cardiovascular outcomes).

Service connection and disability — things veterans should know:

• HIV can be service-connected, and cognitive impairment can be claimed as secondary to service-connected HIV. This matters financially and it matters for access to care.
• Secondary service connection also potentially applies to depression, sleep apnea, and other conditions linked to a service-connected primary condition.
• Documentation drives outcomes. Formal neuropsychological testing, documented functional impairment, and a clear medical opinion linking the cognitive problem to HIV are what a claim needs. Vague notes about "memory complaints" will not do it.
• Veterans Service Organizations (VFW, DAV, American Legion, and state veterans affairs offices) provide free help filing claims. Use them. Do not pay a "claim consultant."
• If you are being evaluated for a cognitive claim, ask specifically whether the examiner has experience with HIV-associated cognitive impairment, which does not look like Alzheimer's disease and can be missed by an examiner expecting it to.

Canada

Drug coverage note for Canada. Antiretroviral coverage varies substantially by province and territory — some provinces cover ART fully through public programs, others require private insurance or have deductibles. British Columbia distributes antiretrovirals centrally at no cost to the patient through the BC Centre for Excellence. If you move between provinces, confirm your coverage before you move, because a gap in ART supply is exactly the kind of treatment interruption this guide is trying to help you avoid. CATIE can advise.

International

International Access and the Regulatory Landscape

This is unusual. For most conditions covered in guides like this, the interesting international content is about which drug is approved where. Here, there is no such content, because the drug does not exist. What varies internationally is something more consequential: access to the things that actually work.

What actually varies by country

The global burden picture — stated honestly

Most of the reassuring statistics in this guide come from well-resourced cohorts in North America, Europe, and Australia. They do not transfer automatically.

• Sub-Saharan Africa carries the large majority of the world's HIV burden. Systematic reviews of neurocognitive assessment in this region consistently report higher rates of impairment, and more severe impairment, than in Western cohorts. Contributors include later diagnosis, more advanced disease at presentation, treatment interruptions from supply chain problems, higher rates of CNS opportunistic infections and tuberculosis, co-morbid malnutrition, and — importantly — inadequate normative data, which inflates apparent impairment rates. Disentangling these is genuinely difficult and is an active research priority.
• South and Southeast Asia, Latin America, and Eastern Europe each have distinct patterns shaped by their epidemics (injection drug use is a much larger driver in Eastern Europe and parts of Asia, bringing hepatitis C co-infection and substance-related cognitive effects with it).
• The ACTG A5199 study (NCT00096824), which enrolled 860 people across seven resource-limited countries, is one of the most important datasets here. At baseline, before ART, 25% had mild, 17% moderate, and 3% severe neurocognitive impairment. After starting ART, impairment fell substantially — and moderate and severe impairment fell the most. That is the single most encouraging finding in global NeuroHIV research: getting people on treatment works, everywhere.

Decision Triggers: When to Act, and What to Do

This section is a set of if-then rules. Use it to convert worry into action.

Top Priorities: What Matters Most, In Order

If you could only do a limited number of things, do them in this order.

1. Take your antiretrovirals every single day and keep your viral load undetectable. If this is hard, fix that first, before anything else on this list. Everything else is built on this.
2. Get depression identified and treated. It is the most common treatable cause of the exact symptoms you are worried about, and treating it improves adherence too, which loops back to priority one.
3. Get your blood pressure to target. The highest-yield non-HIV intervention for your brain.
4. Fix your sleep. Get tested for sleep apnea if there is any suggestion of it. Treat insomnia without sedatives.
5. Stop smoking. Nothing else you can do buys as many healthy years.
6. Cut back or stop alcohol, and address stimulant use. A trial period of abstinence is one of the most informative things you can do — it tells you how much of your problem is reversible.
7. Move your body regularly. Aerobic exercise has the best evidence of any lifestyle intervention for cognition.
8. Insist on a complete confounder workup — thyroid, B12, syphilis, hepatitis C, medication review — before accepting that HIV is the cause.
9. Get your hearing tested. Genuinely.
10. Externalize your memory. Calendar, lists, alarms, a landing zone for your keys. Free, immediate, effective.
11. Get retested over time. Trajectory matters more than a single score.
12. Stay socially connected. Isolation harms cognition, and HIV stigma actively drives isolation. Fight it.
13. Do your legal and financial planning early, while there is no question about your capacity.
14. Do not spend money on unproven supplements or treatments, and check every product with your HIV pharmacist for interactions.

What We Don't Know

Any guide that projects total confidence is lying to you. Here is what is genuinely unresolved — stated so that you can recognize overconfident claims when you meet them.

• How common is real HIV-caused cognitive impairment? The 40–50% figures are almost certainly inflated by criteria that over-diagnose. But the true figure is not known, and it will not be known until better criteria are applied prospectively. Anyone who tells you a precise number is overstating what the field knows.
• Does mild impairment (ANI) actually progress? Some evidence says yes, some says most people remain stable. The honest answer is that it increases risk on average, while most individuals do fine.
• What is causing the persistent mild impairment in well-treated people? The leading hypothesis is low-grade CNS inflammation driven by a persistent brain reservoir. But it could equally be legacy damage that will never resolve, or accumulating vascular disease, or aging, or depression, or testing artifacts, or all of these in different people. We do not know the proportions.
• Are people with HIV at increased risk of Alzheimer's disease? Truly unresolved. Cross-cohort biomarker studies are underway. Do not believe confident claims in either direction.
• Is there any point at which "legacy" damage becomes reversible? Unknown. Some improvement clearly occurs with ART, particularly in more severe cases. How much is recoverable in mild cases, and for how long, is unclear.
• Do integrase inhibitors have meaningful long-term cognitive effects? Data are mixed. Most people do fine. A minority report neuropsychiatric effects. The signal, if real, is small.
• Would an effective anti-inflammatory drug work if we could identify the right patients? We do not know, because we have never been able to identify the right patients. This may be the central failure of the last two decades of trials.
• Does cognitive rehabilitation transfer to real-world function? Unproven, in HIV and in general.
• What are the right cognitive test norms for the global HIV population? A solvable problem that has not been solved, and one with real consequences.
• Will HIV cure strategies improve cognition? Plausible, unproven, and a long way off.
• Which of the two competing diagnostic frameworks is right? The 2023 consensus versus the Frascati defenders is an active, unresolved, and rather sharp scientific argument. Both sides have real points. Time and data will decide it, not this guide.

Living Well

It is worth stopping to say something that gets lost in guides like this one, which by their nature dwell on what can go wrong.

The overwhelming majority of people with HIV who worry about their memory are going to be fine. They are going to work, raise children, travel, argue about politics, fall in love, get annoyed at their neighbors, and grow old. Some of them will be a bit slower than they used to be. Most of them will not develop dementia. Many of them will discover that the problem was sleep, or grief, or a medication, or the fact that they are fifty-eight and tired — and will feel considerably better once it is addressed.

Living well with this involves a small number of things:

Give yourself an accurate story

Fear thrives on ambiguity. "I might be getting dementia" is far more corrosive than "I have mild difficulty with processing speed, my viral load is undetectable, my depression is being treated, and I am being retested in a year." Get the actual facts of your situation, in writing, and stop carrying an imagined version around.

Adjust expectations without surrendering

There is a middle path between denial and collapse. If tasks take longer, allow longer. If noise is hard, choose quieter places. If you can't multitask, don't. These are adaptations, not defeats. People make adaptations for reading glasses and nobody considers it a tragedy.

Keep working, if you can and want to

Work is a source of structure, income, purpose, and social contact — all of which protect cognition. Accommodations exist. Leaving work prematurely because of a test score is often a mistake.

Fight isolation deliberately

HIV stigma pushes people toward isolation, and isolation is bad for the brain. This means that connecting with other people — a support group, a peer navigator, an old friend, a community organization — is not a soft, optional extra. It is part of the treatment plan.

Take the survivor question seriously

Many long-term survivors of HIV are carrying grief and trauma of a scale that most clinicians do not appreciate: they watched an entire generation of friends die, they expected to die themselves, they made no plans for old age because they did not expect to have one, and now they are here. "AIDS survivor syndrome" is a real and increasingly recognized phenomenon. If that describes you, the cognitive symptoms you are experiencing may be sitting on top of untreated grief, PTSD, and depression that has never been addressed. Addressing it is not a diversion from the medical problem. It may be the medical problem.

Ask for help earlier than feels comfortable

Case managers, social workers, peer navigators, benefits counselors, HIV legal services, food assistance, transport assistance — these exist, they are usually free, and people who use them do better. The barrier is almost always pride, not eligibility.

Caregiver Support

Caregiving in this context has features that make it unusually hard, and they deserve to be named.

What makes this different from other caregiving

• Stigma cuts you off from the ordinary supports. A person caring for a parent with Alzheimer's can tell their colleagues, their church, their friends. A person caring for a partner with HIV-related cognitive impairment often cannot — because disclosing the cognitive problem risks disclosing the HIV status, which is not theirs to disclose. This produces an isolation that other caregivers do not face, and it is one of the most damaging aspects of the whole situation.
• The person may be younger than typical. Caring for a partner in their forties or fifties carries different grief, different financial pressure, and different social invisibility than caring for an elderly parent.
• Uncertainty is prolonged. Unlike a progressive dementia with a predictable course, HIV-associated impairment often plateaus, sometimes improves, and frequently turns out to be partly something else. This is good news — but it makes it very hard to know how to plan, and it can make caregivers feel foolish for having grieved.
• Adherence puts you in an impossible position. You are trying to preserve someone's autonomy while also making sure they take a medication that keeps them alive and protects their brain. There is no clean answer to this. Aim for systems, not supervision.
• You may share the diagnosis. Many caregivers in this space are themselves living with HIV. Your own health cannot be sacrificed to someone else's.

Practical caregiver checklist

Glossary

Key References and Sources

These are the primary sources this guide draws on. Where a PMID or NCT number appears, it was verified at the time of writing. Where an identifier could not be confirmed, the study is described in prose and marked accordingly rather than guessed at.

Diagnostic criteria and the nomenclature debate

• Antinori A, Arendt G, Becker JT, et al. "Updated research nosology for HIV-associated neurocognitive disorders." Neurology 2007;69(18):1789–1799. — The Frascati criteria. The foundational document defining ANI, MND, and HAD.
• Nightingale S, Ances B, Cinque P, et al. "Cognitive impairment in people living with HIV: consensus recommendations for a new approach." Nature Reviews Neurology 2023;19(7):424–433. — The 2023 international consensus. Six recommendations from the International HIV-Cognition Working Group; introduces HIV-associated brain injury (HABI). Adopted by EACS in 2023.
• Nightingale S, Dreyer AJ, Saylor D, Gisslén M, Winston A, Joska JA. "Moving on from HAND: why we need new criteria for cognitive impairment in persons living with HIV and a proposed way forward." Clinical Infectious Diseases 2021;73(6):1113–1118.
• Gisslén M, Price RW, Nilsson S. "The definition of HIV-associated neurocognitive disorders: are we overestimating the real prevalence?" BMC Infectious Diseases 2011;11:356. — The over-diagnosis critique.
• Meyer AC, Boscardin WJ, Kwasa JK, Price RW. "Is it time to rethink how neuropsychological tests are used to diagnose mild forms of HIV-associated neurocognitive disorders? Impact of false-positive rates on prevalence and power." Neuroepidemiology 2013;41:208–216.
• Cysique LA, Brew BJ, et al. "Cognitive criteria in HIV: greater consensus is needed." Nature Reviews Neurology 2024. — The formal counter-argument to the 2023 consensus, with the authors' reply published alongside it. Read both if you want to understand the live disagreement.

Epidemiology and cohort studies

• Heaton RK, Clifford DB, Franklin DR Jr, et al. "HIV-associated neurocognitive disorders persist in the era of potent antiretroviral therapy: CHARTER Study." Neurology 2010;75(23):2087–2096. — The most-cited prevalence study; the source of the "about half" figure.
• Heaton RK, Franklin DR Jr, Deutsch R, et al. "Neurocognitive change in the era of HIV combination antiretroviral therapy: the longitudinal CHARTER study." Clinical Infectious Diseases 2015;60:473–480.
• Grant I, Franklin DR Jr, Deutsch R, et al. "Asymptomatic HIV-associated neurocognitive impairment increases risk for symptomatic decline." Neurology 2014;82(23):2055–2062. — The main argument for taking ANI seriously.
• Robertson K, Jiang H, et al. (ACTG A5199 / International Neurological Study). Neurocognitive outcomes after ART initiation across seven resource-limited countries. ClinicalTrials.gov: NCT00096824. — Key finding: odds of impairment fell ~12% per 24 weeks on ART.
• Nyamayaro P, Chibanda D, Robbins RN, Hakim J, Gouse H. "Assessment of neurocognitive deficits in people living with HIV in sub-Saharan Africa: a systematic review." Clinical Neuropsychologist 2019;33:1–26.
• Global prevalence meta-analysis published in Neurology (2020), reporting an overall HAND prevalence of roughly 42–43%. (Verify exact citation and PMID before formal use.)

Treatment trials — including the important negative ones

• Letendre SL, Chen H, McKhann A, et al. (A5324 Study Team). "Antiretroviral therapy intensification for neurocognitive impairment in human immunodeficiency virus." Clinical Infectious Diseases 2023;77(6):866–874. ClinicalTrials.gov: NCT02519777. — Negative. 191 participants; dolutegravir ± maraviroc vs. placebo; everyone improved equally over 96 weeks.
• Ellis RJ, Letendre S, Vaida F, et al. "Randomized trial of central nervous system-targeted antiretrovirals for HIV-associated neurocognitive disorder." Clinical Infectious Diseases 2014;58(7):1015–1022. ClinicalTrials.gov: NCT00624195. — Negative. The definitive test of the CPE hypothesis.
• Letendre S, Marquie-Beck J, Capparelli E, et al. "Validation of the CNS penetration-effectiveness rank for quantifying antiretroviral penetration into the central nervous system." Archives of Neurology 2008;65:65–70. — The origin of the CPE score.
• Sacktor N, Miyahara S, Deng L, et al. "Minocycline treatment for HIV-associated cognitive impairment: results from a randomized trial." Neurology 2011;77(12):1135–1142. — Negative.
• Schifitto G, Navia BA, Yiannoutsos CT, et al. "Memantine and HIV-associated cognitive impairment: a neuropsychological and proton magnetic resonance spectroscopy study." AIDS 2007;21:1877–1886. — Negative for clinical cognition.
• Schifitto G, Zhong J, Gill D, et al. "Lithium therapy for HIV-1 associated cognitive impairment." Journal of NeuroVirology 2009;15:176–186.
• Evans SR, Yeh TM, Sacktor N, et al. "Selegiline transdermal system (STS) for HIV-associated cognitive impairment: open-label report of ACTG 5090." HIV Clinical Trials 2007;8(6):437–446.
• Sacktor N, Skolasky RL, Moxley R, et al. "Paroxetine and fluconazole therapy for HIV-associated neurocognitive impairment: results from a double-blind, placebo-controlled trial." Journal of NeuroVirology 2018;24:16–27. — Negative.
• Cenicriviroc pilot for HAND (University of Hawaii): ClinicalTrials.gov NCT02128828.
• Intranasal insulin for HIV-associated cognitive impairment: ClinicalTrials.gov NCT03277222.
• Computerized cognitive training in adults aging with HAND: ClinicalTrials.gov NCT02758093.
• Computer-delivered cognitive training with tDCS in HAND: ClinicalTrials.gov NCT03440840.
• Statin trial cited in the HAND literature: ClinicalTrials.gov NCT01600170. (Verify current status and published results.)

CSF viral escape

• Canestri A, Lescure FX, Jaureguiberry S, et al. and subsequent case series describing neuro-symptomatic CSF escape with progressive neurological dysfunction despite plasma suppression. (Verify exact citation.)
• Multi-center retrospective case series of neuro-symptomatic CSF escape across four US and European centers — median CSF HIV RNA 3,900 copies/mL against median plasma 62 copies/mL; ART optimization guided by resistance testing produced improvement. (Verify exact citation and PMID.)
• "HIV Cerebrospinal Fluid Escape: Interventions for the Management, Current Evidence and Future Perspectives." Tropical Medicine and Infectious Disease 2025;10(2):45. — Recent review; states that CPE scores are no longer recommended, and summarizes BHIVA/EACS guidance on CSF sampling and regimen optimization.

Mechanisms and reviews

• Saylor D, Dickens AM, Sacktor N, et al. "HIV-associated neurocognitive disorder — pathogenesis and prospects for treatment." Nature Reviews Neurology 2016;12(4):234–248.
• "Mechanisms underlying HIV-associated cognitive impairment and emerging therapies for its management." Nature Reviews Neurology 2023;19:668–687. — Comprehensive current review of mechanisms and the (limited) therapeutic pipeline. Notes explicitly that no treatment for HIV-associated neurocognitive impairment has been approved by the FDA or EMA.
• Angelovich TA, et al. "Regional analysis of intact and defective HIV proviruses in the brain of viremic and virally suppressed people with HIV." Annals of Neurology 2023;94:798–802.
• HIV-Associated Neurocognitive Disorder. StatPearls (NCBI Bookshelf) — freely available, regularly updated clinical summary.

Guidelines and organizations

• European AIDS Clinical Society (EACS) Guidelines, current version — free online. Includes cognitive impairment screening and CSF escape management. Adopted the 2023 consensus approach.
• British HIV Association (BHIVA) guidelines — free online; guidance on CSF sampling and regimen choice in CSF escape.
• US Department of Health and Human Services (HHS) HIV treatment guidelines — clinicalinfo.hiv.gov.
• American Academy of Neurology (AAN) — neurological practice guidance.
• WHO HIV treatment guidelines — the basis for global first-line ART, including the dolutegravir transition.

Patient-facing resources

• HIV.gov (US) — treatment, services, and locator tools.
• HRSA Ryan White HIV/AIDS Program — findhivcare.hrsa.gov, to locate funded care near you.
• CATIE (Canada) — 1-800-263-1638. Excellent plain-language information.
• NAM aidsmap (UK/international) — reliable, readable summaries of HIV research.
• HIV i-Base (UK) — treatment information written for people with HIV; published a clear summary of the 2023 consensus guidelines.
• POZ — community magazine and provider directory.
• 988 Suicide & Crisis Lifeline (US) — call or text 988. Veterans: 988 then press 1.