Introduction

Human immunodeficiency virus-associated neurocognitive disorder (HAND) encompasses a spectrum of neurocognitive impairments resulting from HIV infection, characterized by a progressive and persistent decline in previously established cognitive functions1,2. The primary clinical manifestations of HAND include deficits in attention processing, memory dysfunction, and psychomotor slowing3. The Frascati criteria, introduced in 2007, offer a standardized framework for diagnosing HAND, emphasizing cognitive deficits and functional impairments in daily activities as core diagnostic elements4,5. According to the updated criteria, HAND encompasses three potential research diagnoses—asymptomatic neurocognitive impairment, HIV-associated mild neurocognitive disorder, and HIV-associated dementia—and is characterized by substantial impairments in daily functioning, severely limiting social and occupational engagement, reducing adherence to ART, increasing susceptibility to comorbid conditions, and ultimately diminishing overall quality of life6,7.

Human immunodeficiency virus (HIV) remains a significant global public health concern, affecting approximately 40 million individuals worldwide, with 26.6 million residing in sub-Saharan Africa8,9. A multicounty study conducted in resource-limited settings reported that HIV-associated neurocognitive disorders (HAND) affect an estimated 20% to 37% of individuals living in sub-Saharan Africa10. According to a United States Agency for International Development (USAID) report, HAND prevalence ranges from 20 to 50% among individuals with HIV, potentially impacting up to 35 million people globally, despite the widespread implementation of effective antiretroviral therapy (ART)11.

Neurocognitive assessment is crucial for determining the extent of cognitive impairments associated with HIV/AIDS; however, its administration is seldom feasible as a first-line approach in low-resource settings due to the time-consuming nature of the tests and the requirement for specialized expertise and training to accurately interpret early-stage impairments12,13,14. The gold standard for diagnosing HIV-associated neurocognitive disorder (HAND) entails the use of comprehensive neuropsychological assessments; however, these assessments are frequently inaccessible in high-volume clinical environments due to limitations in resources, time constraints, and the necessity for specialized expertise in their administration and interpretation15,16. However, instruments such as the Montreal Cognitive Assessment, the Mini-Mental State Examination, and the International HIV Dementia Scale have demonstrated utility as practical screening tools for detecting neurocognitive disorders in HIV-positive individuals17,18. The Montreal Cognitive Assessment (MoCA) is an effective alternative to the International HIV Dementia Scale (IHDS) for assessing cognitive performance in HIV-positive patients, but concerns about screening for asymptomatic neurocognitive impairment persist due to the lack of highly sensitive, specific methods and uncertainty over treatment strategies19,20.

The International HIV Dementia Scale (IHDS) and the Mini-Mental State Examination (MMSE) are both commonly used cognitive screening instruments; however, they differ significantly in design, target cognitive domains, and diagnostic sensitivity for HIV-associated neurocognitive disorder (HAND)21,22,23. The MMSE is a general cognitive assessment tool originally developed to detect cortical dementias such as Alzheimer’s disease24,25. It evaluates domains including orientation, attention, memory, language, and the ability to follow simple commands24,26. While the MMSE is valued for its simplicity, widespread use, and broad applicability, its utility in detecting the subcortical cognitive impairments typical of HAND is limited27,28. In contrast, the International HIV Dementia Scale (IHDS) is a validated and widely utilized screening instrument specifically developed to detect HIV-associated neurocognitive disorder (HAND), particularly in resource-limited settings29,30. Unlike general cognitive assessments, the IHDS focuses on cognitive domains typically affected by subcortical dysfunction—a key feature of HAND31,32. It consists of three concise subtests: timed finger tapping to evaluate motor speed, an alternating hand sequence task to assess psychomotor coordination, and delayed recall of four words to measure memory33,34. By targeting subcortical functions, the IHDS is capable of identifying subtle neurocognitive impairments that may not be captured by tools such as the Mini-Mental State Examination (MMSE), which predominantly assess cortical functions22,35,36,37. However, the wide range of available cognitive screening tools, each varying in suitability depending on population, clinical context, and resources, has led to a lack of consensus on the most reliable method for detecting cognitive impairment38,39.

Given Ethiopia’s low coverage and access to education, coupled with high levels of illiteracy the country’s cultural and linguistic diversity, it is essential to assess the applicability and validity of cognitive screening tools like the International HIV Dementia Scale (IHDS) in this context40. The validation of such tools is critical for ensuring their effectiveness in detecting HIV-associated neurocognitive disorders in Ethiopian populations, enabling evidence-based decision-making. While the gold standard for evaluating cognitive disorders remains the administration of an extensive neuropsychological assessment, the cost, time, and resource constraints make this impractical in many settings. Therefore, validating the IHDS and MMSE in Ethiopia will provide a reliable, accessible, and culturally relevant tool for early detection and intervention in HIV/AIDS patients, improving clinical outcomes in the country.

Methods and materials

A hospital-based cross-sectional study was conducted from February 10 to March 2, 2023, at three selected primary hospitals in the South Gondar Zone of the Amhara Region, Ethiopia. The South Gondar Zone, comprising 13 districts (woredas) and seven town administrations, had an estimated population of 2,609,823 in 2023 and is supported by a healthcare system that includes nine hospitals—eight primary hospitals and one comprehensive specialized hospital—along with 109 health centers and 402 health posts. For this study, Addis Zemen Primary Hospital, Nifas Mewicha Primary Hospital, and Ebnat Primary Hospital were selected using a lottery-based simple random sampling technique from among the eight primary hospitals that provide general medical and HIV care services. These government-owned institutions are located in different districts and serve predominantly rural and semi-urban populations. Each hospital delivers a broad range of healthcare services, including outpatient and inpatient care, maternal and child health services, emergency care, and treatment for communicable diseases such as tuberculosis and malaria. They also operate dedicated HIV clinics that offer antiretroviral therapy (ART), voluntary counseling and testing, clinical monitoring, and management of HIV-related opportunistic infections. As primary referral centers for nearby health centers and health posts, these facilities play a critical role in the delivery of essential healthcare services across the zone.

Participants

The study population comprised HIV-positive adults aged 18 to 64 years who were receiving care at the antiretroviral therapy (ART) clinics of Addis Zemen, Nifas Mewicha, and Ebnat Primary Hospitals during the study period. Inclusion criteria required participants to be actively engaged in routine HIV care and able to provide informed consent. Participants were excluded if they were antiretroviral therapy (ART)-naïve, had a newly diagnosed HIV infection, or exhibited severe medical or psychiatric conditions—such as acute psychosis or delirium—that could interfere with effective communication or the reliability of cognitive assessments. Additional exclusion criteria included a history of cerebrovascular accidents (stroke), significant visual or speech impairments, and educational attainment below the fifth-grade level, as low literacy could compromise the validity of cognitive assessments.

The sample size was determined based on psychometric validation guidelines, which recommend 5 to 20 participants per test item, with 50–100 participants generally considered adequate for assessing reliability and validity41,42,43,44. Given the three-item structure of the International HIV Dementia Scale (IHDS), a sample of 60 participants was deemed sufficient for robust statistical analysis, with proportionate stratified sampling employed to enhance representativeness and minimize selection bias. Participants were systematically allocated across the three study sites based on the monthly ART clinic attendance rates at each hospital. Ethical approval for this study was obtained from the Institutional Review Board (IRB) of the University of Gondar. Before enrollment, all participants received comprehensive information regarding the study’s objectives, procedures, potential risks, and anticipated benefits. The research was conducted in accordance with the ethical principles outlined in the Declaration of Helsinki, specifically its 2013 revision45. Written informed consent was obtained from each participant prior to their involvement in the study. Throughout the research process, strict ethical standards were maintained, including the assurance of confidentiality, the voluntary nature of participation, and the participants’ right to withdraw from the study at any point without any negative consequences.

Measures and procedures

Sociodemographic data, including age, marital status, and educational attainment, were collected using semi-structured questionnaires. All participants underwent cognitive evaluation utilizing the International HIV Dementia Scale (IHDS) and the Mini-Mental State Examination(MMSE) both of which were delivered in the participants’ native language, Amharic, by trained research personnel. The International HIV Dementia Scale (IHDS) is a brief screening tool designed to assess cognitive impairment associated with HIV33. The IHDS comprises three components that assess motor speed, psychomotor speed, and memory function. These are evaluated through three subtests: timed finger tapping, timed alternating hand sequence, and delayed recall of four items after a 2-min interval. The test takes approximately 10 min to administer. Each subtest is scored independently, with a maximum of four points per component, yielding a total score ranging from 0 to 12. A score of ≤ 10 is commonly used as a threshold to indicate potential neurocognitive impairment46,47. The Mini-Mental State Examination (MMSE) is a brief cognitive screening tool consisting of 11 items that assess five key domains: orientation, registration, attention and calculation, recall, language, and visuospatial abilities. The test takes approximately 5–10 min to administer and yields a total score out of 30, with higher scores indicating better cognitive function. A score of 25 or above is generally considered normal, while a score 24 and below is commonly used as a cut-off to suggest possible neurocognitive impairment48,49,50. The International HIV Dementia Scale (IHDS) and the Mini-Mental State Examination (MMSE) were administered as index tests to determine their diagnostic accuracy against the Frascati-based classification (ANI, MND, HAD). Because the MMSE is influenced by literacy and formal education, its limitations in low-education, low-income settings (e.g., Ethiopia) were taken into account during interpretation.

Gold standard assessment

The reference standard was the Frascati criteria, based on a comprehensive neuropsychological test battery administered by trained professionals. Participants were evaluated across seven cognitive domains: attention and working memory, learning and memory, executive function, processing speed, language/verbal fluency, and sensory-perceptual/motor skills. The selection and allocation of tests to these domains were determined a priori, guided by established neuropsychological practice, psychometric robustness, cognitive theory, and evidence from prior research on HIV-related cognitive impairment. The Frascati framework classifies HIV-associated neurocognitive disorders (HAND) into three categories: asymptomatic neurocognitive impairment (ANI), mild neurocognitive disorder (MND), and HIV-associated dementia (HAD). ANI and MND are defined by performance at least one standard deviation below normative means in two or more domains, with ANI showing no functional impairment and MND associated with mild functional decline. HAD, the most severe form, is diagnosed when deficits fall at least two standard deviations below the mean in two or more domains and are accompanied by marked impairment in daily functioning51. In this study, the Frascati criteria were used as the definitive reference to assess the diagnostic accuracy of the International HIV Dementia Scale (IHDS) and the Mini-Mental State Examination (MMSE) among Ethiopian individuals living with HIV52,53.

Data collection and analysis

Data were collected through semi-structured face-to-face interviews conducted by three trained psychiatry professionals from the ART clinics of selected primary hospitals. The data collection process was overseen by master’s-level psychiatry professional to ensure methodological rigor and strict adherence to the study protocol. Prior to data collection, a comprehensive training session was provided to both data collectors and supervisors. The training covered the study’s objectives, standardized data collection procedures, ethical considerations, and a detailed walkthrough of the questionnaire to ensure clarity, uniform administration, and test–retest reliability. To enhance cultural appropriateness and ensure linguistic accuracy, the questionnaire underwent a rigorous validation process. This included forward and backward translation into Amharic, verified by two independent bilingual experts proficient in both English and Amharic. Continuous supervision and quality control measures were maintained throughout the data collection period. The principal investigator, along with the supervisors, conducted regular monitoring visits, provided on-site guidance, and reviewed completed questionnaires daily to assess accuracy, completeness, and internal consistency. Immediate corrective feedback was provided to data collectors as needed to maintain data integrity and minimize errors.

Data were double-checked for completeness before being entered into EpiData version 4.6.02 and subsequently exported to SPSS version 21 for statistical analysis. Descriptive statistics, including means, standard deviations, frequencies, and percentages, were computed. Intraclass correlation coefficients (ICCs) were calculated for the IHDS and MMSE using a subsample of 30 HIV-positive participants, with test–retest reliability assessed over a 2–3 h interval. Receiver Operating Characteristic (ROC) curve analysis was employed to evaluate the diagnostic accuracy and discriminatory capacity of the IHDS and MMSE, using cognitive impairment classified by the Frascati criteria as the reference standard. Sensitivity, specificity, positive and negative predictive values (PPV and NPV), and likelihood ratios were computed across a range of cutoff points. Optimal cutoff thresholds were identified using Youden’s J index value, with acceptable diagnostic performance defined as sensitivity ≥ 0.8 and specificity ≥ 0.654. Furthermore, multivariable linear regression analysis was employed to identify independent predictors of IHDS scores, following verification of all underlying model assumptions.

Result

Socio-demographic characteristics of the participants

Of the 60 HIV-positive individuals invited to participate, 59 completed the questionnaire. The majority of participants were female (57.7%). The mean age of the study population was 36.93 years (± 13.08), with an age range of 18 to 64 years. The majority of participants (88.1%) resided in urban areas. In terms of educational attainment, 45.8% had completed secondary education. Similarly, 45.8% of participants were married, while 30.5% were employed in the government sector (Table 1).

Table 1 Sociodemographic characteristics of HIV-positive patients attending in South Gondar primary hospitals.
Full size table

Content validity

Following the selection of neuropsychological assessment tools, a panel of experts—including two psychiatrists, one psychiatry resident, and three master’s-level psychiatric professionals—convened to evaluate the content validity of the instruments. The panel reached full consensus on all components, resulting in a Scale Content Validity Index (S-CVI) of 100%55,56. One psychiatrist remarked, “The questionnaires can effectively measure neurocognitive disorders.” However, concerns were raised regarding specific test items. For example, one psychiatrist noted that the Amharic words for “red” and “beans” sounded phonetically similar, potentially affecting recall due to their resemblance to a metallic sound. To address this, the term “beans” was replaced with “lentils” (misir, ) to enhance word differentiation. Further concerns focused on the IHDS motor tasks, such as finger tapping and hand clenching within specified timeframes (5 and 10 s), which some participants found difficult due to variations in comprehension and motor coordination. To mitigate this, data collectors demonstrated the tasks before participant attempts to ensure accurate performance. While the panel ultimately deemed the IHDS items clear and appropriate for assessing neurocognitive disorders in people living with HIV/AIDS, challenges remained. Notably, the MMSE posed difficulties for individuals with limited formal education, leading to the exclusion of those below grade five (Table 2).

Table 2 Content validity index of raters for international HIV dementia scale.
Full size table

Pearson correlation of international HIV dementia scale scores with sub-items

Correlations between the total IHDS score, subtest scores on the non-dominant hand, and each neuropsychological test were assessed. Significant correlations were observed between the IHDS total score and subtests such as finger tapping, hand sequencing, and memory recall, all of which demonstrated strong associations with the IHDS components. The International HIV Dementia Scale showed a high correlation with these subtests, while memory recall exhibited a moderately lower correlation. The final item of the scale, which required participants to recall four words to assess memory, was accompanied by prompts containing semantic cues, potentially contributing to the lower correlation coefficient observed for this specific subtest (Table 3).

Table 3 Pearson correlation of the International HIV Dementia Scale total with sub-items.
Full size table

Psychometric properties of international HIV dementia scale and Mini-mental state examination

The International HIV Dementia Scale (IHDS) achieved a maximum Youden’s J-index of 0.61 at an optimal cutoff of ≤ 10, with a sensitivity of 84.2%, specificity of 77.5%, positive and negative likelihood ratios of 3.38 and 0.2, respectively, a PPV of 64%, and an NPV of 91.2%. In comparison, the Mini-Mental State Examination (MMSE) showed a maximum Youden’s J-index of 0.41 at an optimal cutoff of ≤ 27, yielding a sensitivity of 83.9%, specificity of 57.1%, positive and negative likelihood ratios of 1.95 and 0.28, respectively, with a PPV of 68.4% and NPV of 76.2% (Tables 4 and 5).

Table 4 Psychometric properties of IHDS for detection of neurocognitive disorder.
Full size table
Table 5 Psychometric properties of mini-mental state examination for detection of neurocognitive disorder.
Full size table

ROC curve analysis of international HIV dementia scale and Mini-mental state examination

Receiver operating characteristic (ROC) analyses were conducted to assess the discriminatory capacity of the International HIV Dementia Scale (IHDS) in identifying varying degrees of neurocognitive impairment. The area under the ROC curve (AUC) for the IHDS was 0.81 (95% CI 0.68–0.93, p-value < 0.001), indicating good diagnostic accuracy. The optimal cutoff score, determined by the highest Youden’s J-index (0.61), was ≤ 10. At this threshold, the IHDS achieved a sensitivity of 84.2% and a specificity of 77.5%. When the cutoff was increased to ≤ 11, sensitivity slightly improved to 84.8%, whereas specificity declined to 53.8%. In comparison, the Mini-Mental State Examination (MMSE) yielded an AUC of 0.71, suggesting moderate diagnostic accuracy. The optimal cutoff score for the MMSE, corresponding to the highest Youden’s J-index (0.41), was ≤ 27, resulting in a sensitivity of 83.9% and a specificity of 57.1%. Using a stricter cutoff of ≤ 26 led to reduced sensitivity and specificity, at 69.6% and 44.4%, respectively (Figs. 1 and 2).

Fig. 1
figure 1

Receiver Operating Characteristic (ROC) curve of the International HIV Dementia Scale (IHDS) for detecting HIV-Associated Neurocognitive Disorders (HAND), demonstrating good diagnostic accuracy with an area under the curve (AUC) of 0.81 (95% CI 0.68–0.93, p-value < 0.001). The optimal cutoff score of ≤ 10 yielded a sensitivity of 84.2% and a specificity of 77.5%. The green diagonal line represents the reference line (line of no discrimination), and the blue curve indicates the IHDS performance based on the Frascati criteria as the state variable.

Full size image
Fig. 2
figure 2

Receiver Operating Characteristic (ROC) curve of the Mini-Mental State Examination (MMSE) for detecting HIV-Associated Neurocognitive Disorders (HAND), demonstrating moderate diagnostic accuracy with an area under the curve (AUC) of 0.71. An optimal cutoff score of ≤ 27 yielded a sensitivity of 83.9% and specificity of 57.1%. The green diagonal line represents the reference line (no discrimination), and the blue curve depicts MMSE performance using the Frascati criteria as the gold standard.

Full size image

Reliability

The IHDS demonstrated good test–retest reliability with an intra-class correlation coefficient of 0.73 (95% CI 0.54–0.83, p < 0.001) and acceptable internal consistency, indicated by a Cronbach’s alpha of 0.73. Deletion of individual items did not result in any increase in the alpha value, indicating that all items contributed meaningfully to the overall consistency of the tool. In comparison, the Mini-Mental State Examination (MMSE) showed moderate test–retest reliability with an intra-class correlation coefficient of 0.51 (95% CI 0.51–0.71) and internal consistency, as measured by Cronbach’s alpha, also at 0.51.

Factors associated with International HIV Dementia Scale scores

A multiple linear regression analysis indicated that several factors significantly affected neurocognitive performance, as measured by IHDS scores, among HIV-positive patients in South Gondar primary hospitals. Specifically, being married (β = 0.6, p-value = 0.02) and young adults (β = 0.52, p-value = 0.04) were significantly associated with higher IHDS scores among HIV patients in South Gondar primary hospitals, whereas sex, educational status, occupation, and living conditions were not significantly associated with IHDS scores (P-value > 0.05) (Table 6).

Table 6 Multiple linear regression model results for factors associated with international HIV dementia scale scores among HIV patients in south Gondar primary hospitals.
Full size table

Discussion

Multiple studies suggest that due to the lack of an internationally standardized and endorsed screening tool for neurocognitive disorder among HIV positive patients, local assessment of screening tools is necessary. Our study is the first attempt to validate and see the psychometric properties of screening tools with respect to cART usage in Amharic language speakers. In the current study, we attempted to determine the psychometric properties of the IHDS. In this cross-sectional study of HIV+ patients receiving routine follow-up care in an ART clinic, by using an international HIV dementia scale screening tool with a cut-off point less than or equal to ten, 27 (45.8%) participants were positive for neurocognitive disorder, while 32 (54.2%) participants were negative for HIV-associated neurocognitive disorder.

These findings also indicated the good reliability of the tool in consistently measuring cognitive function and the ability of the tool to provide consistent measurements when used with test–retest reliability. The study also showed that the tool had great specificity, sensitivity, and predictive values at an ideal cutoff value of 10, with an area under the curve (AUC) value of 0.81. Over the years, a number of studies have been carried out to investigate the validity and reliability of IHDS in various languages, contexts, and demographic groups. The majority of this research has provided varying reports on specificity, sensitivity, and reliability metrics.

The initial study was conducted by Ned Sacktor’s (MD) and his colleagues in the USA and Uganda, with cohort studies showing sensitivity and specificity of 80% and 57% and 80% and 55%, respectively30, and a systematic review done in China with a cut-off point of 10 or less showed pooled sensitivity and specificity for IHDS values of 85.5 and 92.2%, respectively, with an overall AUC value of 0.7346. These consistent reports indicate that the tool has good reliability in measuring the cognitive functioning of individuals in different settings. Another study was carried out in Germany among a population with a multicultural and language background, which found that a cut-off value of 10 increased both sensitivity (94%) and specificity (86%), respectively11. All corroborated with the current study’s findings regarding the effectiveness and validity of IHDS in multicultural and linguistically diverse populations, including Africa.

Conversely, in a study conducted in South Africa, the AUC was 0.64 and the sensitivity and specificity were 45% and 79%, respectively57; a meta-analysis conducted in Romania on the five continents with a cut-off point of 10 or less, a specificity of 65.4%, and a sensitivity of 74%3; in China, it was also reported that Cronbach’s alpha was 0.58 and the sensitivity and specificity were 73.7% and 71.3%58; and in Italy, a cutoff value of 11, with a sensitivity of 0.70 and a specificity of 0.8259. Those results were far from the current study. The reason for the low detection might be because of the study participants, for whom in South Africa the participants age range was between [18–40] years. With respect to age, early adulthood and middle adulthood are less risky and may have a shorter duration of being infected with HIV as compared with late adulthood. The pathophysiological pathways of ageing and HIV may have independent additive effects that cause higher cognitive and neurological deficits in older adults with HIV/AIDS60,61. Another reason might be a study carried out in Italy, where they enrolled patients with neurological disorders with sub cortical features like multiple sclerosis and sub cortical ischemic vascular disease62,63. Patients with multiple sclerosis may exhibit behavioral symptoms such as depressed mood, euphoria, and pathological laughing or crying, but psychosis is less common compared to HIV-positive individuals, which may influence their IHDS scores64,65.

This study found that younger adult age was significantly associated with higher IHDS scores (β = 0.52, p-value = 0.04), suggesting better cognitive function among younger participants. Cognitive aging is a gradual and highly individual process in which abilities like vocabulary often remain stable, while functions such as memory, executive control, and processing speed typically decline from midlife onward, making it challenging to distinguish normal aging from pathological conditions like dementia66,67,68. While cognitive abilities like vocabulary and general knowledge often remain stable, functions such as memory, executive functioning, and processing speed typically decline from midlife onward along a continuum, blurring the line between normal aging and pathological conditions like dementia69,70. Early-life cognitive ability also predicts later-life outcomes, emphasizing the role of lifelong influences71. According to the common cause theory, biological processes like oxidative stress and immune system aging contribute to both cognitive and physical decline72,73. Younger adults may benefit from fewer comorbidities and better overall health, supporting stronger neurocognitive outcomes.

This study found that being married was significantly associated with higher IHDS scores (β = 0.6, p-value = 0.02), suggesting that marital status may play a protective role in cognitive functioning. Being married may serve as a protective factor against cognitive impairment in older adults. Spousal relationships often provide ongoing emotional support, social engagement, and cognitive stimulation, which are all associated with better cognitive health outcomes74,75. This aligns with the brain reserve hypothesis, suggesting that the mental and social enrichment within marriage may buffer against age-related cognitive decline76,77. Longitudinal research also indicates that cognitive changes in one spouse can influence the other, further supporting the idea that marital dynamics contribute to cognitive resilience78,79. Additionally, evidence from longitudinal National Health and Aging Trends Study reveals that never-married and widowed individuals are at higher risk of dementia, reinforcing the potential cognitive benefits of marriage in later life80. These findings highlight the importance of considering marital status as a key factor in the prevention and management of cognitive impairment.

Overall, the results of this study showed the tool’s usefulness and efficiency in detecting cognitive disorders and screening for neurocognitive diseases. The tool has proven to be a valid, trustworthy, and useful instrument for usage in Ethiopia due to its demonstrated ability to evaluate people’s cognitive status and screen for cognitive diseases of diverse etiologies in the current study and other investigations. The tool was also demonstrated to be appropriate for the frequently hectic outpatient settings and to be usable by workers without training in psychiatry or mental health. This is the first study to date that has evaluated and reported on the validity of international HIV dementia scale in Ethiopia.

Limitation

Our study assessed test–retest reliability; it did not evaluate inter-rater reliability, which is essential for determining whether different examiners can administer the IHDS consistently in clinical practice. Moreover, the short 2–3 h interval between test administrations raises concerns about potential practice effects, which may have led to inflated reliability estimates. The relatively small sample size, drawn exclusively from a single geographic region, limits the generalizability of our findings to the broader Ethiopian population or other resource-constrained settings. Additionally, excluding individuals with less than a fifth-grade education due to MMSE-related difficulties may have introduced selection bias, thereby reducing the applicability of our results to populations with lower educational attainment. Although cultural adaptations were made, certain components of the IHDS—particularly motor tasks—remained difficult for some participants to understand, potentially affecting the accuracy of their performance. Finally, the cross-sectional design of the study precludes an assessment of cognitive changes over time, and the absence of longitudinal follow-up limits the evaluation of the tool’s predictive validity and sensitivity to clinical progression.

Conclusion

Both the International HIV Dementia Scale (IHDS) and the Mini-Mental State Examination (MMSE) proved valuable for screening HIV-associated neurocognitive disorder (HAND) among HIV-positive Ethiopian adults, with the IHDS showing superior diagnostic performance, including higher sensitivity and a greater area under the curve (AUC). The IHDS emerged as a reliable, valid, low-cost, and time-efficient screening tool suitable for use in resource-limited settings. This study established local normative data and optimal cutoff scores specifically tailored for Amharic-speaking populations, enhancing the tool’s relevance and applicability within Ethiopian clinical practice. The findings support the integration of the IHDS into routine HIV care for early detection of HAND. However, further research is warranted to refine the IHDS, assess its inter-rater reliability, and validate its use across diverse languages, settings, and demographic groups. Such efforts will help strengthen its role as a brief and practical screening tool in both primary care and specialized clinical environments.