Neuropsychological Assessment of Older Adults in Nepal for Population-Based Dementia Ascertainment: Needs,Challenges,and Opportunities

Prevalence of dementia and its risk factors in Nepal

There are no existing population-based data on the cognitive health of older adults, the prevalence and determinants of dementia, or the needs of individuals with dementia and their support system in Nepal [6]. Existing, limited data suggest dramatically different prevalence rates of cognitive impairment in older adults, likely driven by differences in sampling and assessment methods. For example, one study of 260 Nepali adults aged 60 and older who sought care in a psychiatry outpatient department in a tertiary-level hospital reported a dementia diagnosis among 11.4% of the sample over a 1-year period [6]. For this study, clinical diagnoses of dementia were made using ICD-10 criteria and included assessment with the Mini-Mental State Examination (MMSE). In contrast, the Ministry of Health and Population of Nepal reported only 117 diagnosed cases of AD in the country [7], which almost certainly reflects substantial underdiagnosis of dementia. A study of 115 community-dwelling older Nepali reported a mild cognitive impairment (MCI) prevalence of 93% [8]. This study utilized the Montreal Cognitive Assessment (MoCA) for MCI assessment. The authors did not describe an approach to MoCA translation and adaptation and reported that 81.7% of their sample had no formal education. They used the original publisher’s cut-offs for classification of impairment (MoCA < 26) and did not differentiate between the classification of MCI and dementia. These limitations likely contribute to a substantial overestimate of MCI in this sample. These findings also underscore current challenges in implementing existing cognitive screening tools in Nepal, given the absence of tools that have been adapted for the Nepalese cultural and linguistic context, the lack of tools that have been validated for individuals with no formal education in Nepal, and the lack of existing population-specific normative data from which to interpret test scores. Taken together, the prevalence of mild cognitive impairment and dementia in community-dwelling older adults in Nepal remains unknown.

There are limited data on the physical and mental health characteristics of older Nepalis, some of which may be important risk factors for dementia. Studies that exist are limited by small or non-representative samples and inconsistent use of measurement tools [9]. These previous studies suggest that diabetes, hypertension, gastritis, and musculoskeletal health conditions are the most common physical health conditions in older adults and often coexist [10, 11]. Diabetes and hypertension, in particular, represent important modifiable risk factors for dementia [12]. Similarly, depression (49.2%) and loneliness (55.6%) appear to affect a sizeable percentage of Nepali older adults [13] and each have been shown to be associated with risk for cognitive decline and dementia [12, 14]. There have been several significant stressors impacting the Nepali older adult population, including a decade-long armed conflict (1996-2006), a devastating earthquake (2015), and the COVID-19 pandemic. Exposure to these chronic stressors may increase dementia risk through their impact on mental health (e.g., post-traumatic stress disorder; depression) and physical health (e.g., allostatic load) [15]. While studies have reported a higher risk of mental health diagnoses among those exposed to the decade-long armed conflict in Nepal [16], earthquake [17], and COVID-19 [18], no existing studies have linked these stressors to the cognitive health of older adults in Nepal. Nepal thus provides a unique context to understand the association between these shared, multiple environmental exposures and cognitive aging in LMICs.

Existing neuropsychological assessment tools in Nepal

A reliable and valid method of dementia assessment is a critical foundation to accurately determine the prevalence of dementia in Nepal. Currently, there are no comprehensive dementia assessment instruments that have been developed and validated for use in Nepal. Table 1 provides a summary of the methodology used in existing studies of dementia and/or MCI in Nepal. As seen in Table 1, most studies have used screening instruments such as the MMSE, although there is not yet an evidence base for the translation, adaptation, and validation of this instrument for the Nepali older adult population. One group has undertaken the translation and adaptation of the Rowland Universal Dementia Assessment Scale [19] in Nepal. This work is a significant step forward in the development of culturally appropriate cognitive assessment tools in Nepal, although validation of this instrument in older adults with ADRD is currently lacking.

There have been very few cognitive assessment instruments that have been utilized in other adult populations (e.g., other neurologically impaired populations) in Nepal, leaving no substantial evidence base to draw from in developing cognitive assessment instruments for ADRD assessment. This represents a challenge for the adaptation of ADRD neuropsychological assessment instruments in Nepal. As such, a very careful translation, adaptation, and validation approach is urgently needed to ensure that new cognitive assessment instruments are appropriate for the Nepali population.

Sociocultural features of the aging population in Nepal that are relevant for neuropsychological assessment

When considering the need to appropriately develop, translate, and/or adapt neuropsychological assessment tools for a population that does not yet have an evidence base in neuropsychology, it is critical to consider the population’s cultural, linguistic, educational, and social context. Nepal is an ethnically and linguistically diverse country. There are a total of 142 caste/ethnic groups, the largest of which is Chettri (16.45%) followed by Brahmin (11.29%) [20]. Although Nepali is the official and most commonly spoken language in Nepal (by 44.6% of the population as the primary language and by 46% of the population as the second language [20]), 123 local dialects are spoken nationwide [21]. The lifestyle of the majority of the Nepali population is embedded in the Hindu religion. According to the Census data from Nepal of 2011, more than 80% of people aged 60 + are Hindus, and 10% are Buddhist. Although Nepal has made significant strides in increasing literacy rates in recent years, many older adults in Nepal have no to limited formal education and are not able to read or write. In 2001, only 14.1% of adults aged 60 + were literate, which increased to 22.9% in 2011 [22]. The vast majority of older adults reside in rural areas: only 14.3% of older adults resided in urban areas in 2011 [22].

In terms of socioeconomic status, about half of the older adult population is economically active and the vast majority work in the agriculture sector. In 2001, 50.7% of older adults participated in the labor force, while this decreased to 46.1% in 2011. According to the 2011 Census, 6.4% of older adults (60+) reported disability, an increase from 1% in 2001 [22]. Since the vast majority of older adults do not receive pensions from retirement, the allowance received from the government is an important source of income for these age groups, although age eligibility varies by geographic location, ethnicity, and marital status. In the absence of retirement pensions, this allowance is often not adequate for older adults who are not supported by their relatives [23].

Family relationships are highly valued in Nepali culture. Older adults often have strong relationships with their children and extended family members [10]. A small fraction of older adults live alone (4% in 2011). The average family size in households with older adults was 5.7 in 2011 [22]. However, there is a trend for increasing widowhood in Nepal. Whereas in 2001, 81.3% of 60 + were married or in union and 16.2% were widowed, the 2011 numbers were 77.0% and 21.4%, respectively [22]. In addition, older adults are increasingly likely to live alone and manage their own health due to emigration of younger men and due to younger women and children moving out of villages to access better access to education and health opportunities [24]. Until recently, older adults would be cared for by family members, and only those without family members or those neglected by them would be sent to Old Age Homes (Briddhashrams). Since establishing an Old Age Home is a cumbersome process requiring multiple layers of approval from municipalities, the District Administrative Office and at times from the Social Welfare Council, there are very few (141) Old Age Homes in Nepal, of which 75 are in Kathmandu and the rest elsewhere in the country [25]. Many of these Old Age Homes also lack adequate amenities for proper care, support and basic needs [25].

Taken together, the unique social, cultural, linguistic, and educational characteristics of Nepal have a number of implications and challenges for neuropsychological assessment. Given the high prevalence older Nepali adults with no formal education, neuropsychological assessment tools need to be accessible for individuals with little to no literacy and numeracy. This reflects a critical and urgently needed challenge for the field of neuropsychology, as most neuropsychological assessment instruments have been developed and validated within North American populations with high levels of education. In addition, formal education systems provide exposure to the experience of testing; as such, a systematic approach to orientation to the neuropsychological testing experience is also needed. It is critically important to consider linguistic diversity when developing neuropsychological assessment tools for this population. Receiving a cognitive assessment in one’s non-dominant language may interfere with assessment validity and sensitivity of measurement. In addition, all test stimuli need to be scrutinized to evaluate their familiarity and cultural appropriateness in the Nepali population.

There are several cultural factors that may affect the assessment of functional impairment due to dementia in this population. Family members are likely to attribute milder memory loss as a part of the normal aging process and more severe memory loss as mental illness or psychiatric disorders [26]. Additionally, there is not a specific Nepali term for dementia [26, 27]. The manifestation of dementia symptoms is typically described as pagalpan (people are referred as “pagal”), which closely translates to madness in the English language [27]. Relating dementia to pagalpan may demotivate both older Nepali and their family members to acknowledge and disclose ADRD symptoms [28] due to the country’s deeply-rooted stigma associated with pagalpan and other mental illnesses. This likely results in the failure to recognize symptoms of dementia as a condition that requires medical assessment and intervention. Due to cultural values and stigma, close family members may also be hesitant to disclose concerns regarding changes in an older adult’s cognition and functional independence, which may interfere with attempts to identify cognitive disorders such as dementia in the general population.

CVFS-SCAN and the HCAP International Network

CVFS-SCAN will join the growing collaboration of the Harmonized Cognitive Assessment Protocol (HCAP) International Partner Studies. The HCAP International Network represents a collaborative effort to develop a cross-nationally comparable cognitive assessment instrument, informant report, and dementia ascertainment approach across several countries across the world. The HCAP was developed within the framework of the Health and Retirement Study (HRS) in the United States and its international network of partner studies. The HCAP has been implemented in 18 countries, with new data collection planned for several additional countries [36]. The cognitive assessment and informant interview are comprised of several neuropsychological instruments that are commonly used for the assessment of ADRD and were selected to optimize cross-national comparability [37]. The items in the battery are adapted by each study to optimize the balance between within-study appropriateness (cultural, linguistic, educational) and cross-study comparability. For the HCAP informant interview, an informant is recruited to provide collateral information with a structured interview. The informant is typically a close family member or friend that knows the respondent well. The informant interview is composed of several sections that are designed to capture the informant’s observations of the respondent’s cognition and functioning.

The HCAP international network offers an optimal foundation from which to develop a population-based cognitive assessment and dementia ascertainment protocol for older adults in Nepal. Each study within the HCAP network has collected cognitive data with adaptation to each study population’s context and includes studies with populations with educational (e.g., India, Mexico) and linguistic (e.g., India) diversity. Harmonized HCAP data across several studies have been made publicly available (http://www.g2aging.org). Initial work with the HCAP across national contexts has shown a similar factor structure across several countries [38–41] and evidence of comparable measurement [38, 42] across national contexts after cultural neuropsychology-informed harmonization procedures [43].

CVFS-SCAN will develop an HCAP assessment aimed at arriving at the appropriate classification of MCI and dementia among older Nepali adults. It will build on the HCAP assessment that is used in the context of the Longitudinal Aging Study in India (LASI) study, a nationally representative study of more than 73,000 adults in India aged 45 and older across all 29 states in India. Of all the international network of HCAP partner studies, LASI takes place in a population that is geographically closest, and most socially, educationally, and culturally similar to the Nepali older adult population. LASI implemented the HCAP assessment with its Diagnostic Assessment of Dementia sub-study (LASI-DAD). The LASI-DAD study administered the HCAP to a subsample of 4,096 LASI participants aged 60 and older across 18 states in India, with oversampling of participants with low cognitive test scores in the core LASI survey. This sample is linguistically diverse (12 assessment languages, with the highest proportion (31.6%) tested in Hindi) and 56% of the sample has no formal education [39, 44]. The LASI-DAD study team rigorously translated, adapted, and piloted the HCAP instrument in India to ensure its appropriateness for their study population and performed a feasibility study in a small sample of older adults with dementia [44, 45]. They have published the factor structure of the HCAP instrument in India, which showed adequate model fit, with a general cognitive performance (GCP) factor and five broad cognitive domains [39]. The study has excellent documentation of their instrument’s administration, scoring, and data (http://www.lasi-dad.org) and has demonstrated comparable measurement with the HRS-HCAP instrument [42].

Adaptation of the HCAP for the Nepal context

Given the relative sociocultural, educational, and linguistic similarities of the study populations and the strong methodological and empirical foundation of the LASI-DAD study, CVFS-SCAN will use the LASI-DAD HCAP (Table 2) as the base for translation and adaptation in Nepal. Despite LASI-DAD HCAP assessment’s strong foundation, comprehensive methods are necessary to ensure that the HCAP instrument is appropriate to measure cognitive health and dementia in the Nepal context. The Chitwan Valley Family Study research team has extensive experience in the translation and culturally appropriate validation of several health instruments for the Nepal context [46] using established methodology from cross-cultural survey guidelines.

To develop a translation and adaptation of the HCAP, our team is drawing upon guidance on best practices recommended through the International Test Commission Guidelines for Translating and Adapting Tests [47] and the Cross-Cultural Survey Guidelines [48]. Our work has been approved by the University of Michigan institutional review board (IRB-MED) and the Nepal Health Research Council and is completed in accordance with Helsinki Declaration. These methods (Fig. 1) include an iterative process. We have assembled a local multidisciplinary expert team to conduct an initial translation of the HCAP instrument. Next, a multidisciplinary local Nepali expert panel including a practicing neurologist, geriatrician, psychiatrist, sociologist, and anthropologist reviewed the initial translation and provided input regarding the appropriateness of the translations and the items in the local context. A focus group was then conducted with this expert panel for group discussion. The study neuropsychologist collaborated with other study team members to review this feedback and consider items in need of further refinement. Next steps will be to conduct cognitive interviewing with cognitively healthy older adults residing in the Chitwan Valley community, with representation of the full spectrum of education, language, ethnicity, and sex/gender that exists in the larger Chitwan Valley community. Cognitive interviewing [49] is a qualitative methodology that aims to interrogate how understandable and familiar test instructions and stimuli are and to confirm that the intended construct is being measured by a given test item. We will integrate this feedback with focus group feedback to further refine the HCAP instrument. After additional expert review, we will then perform pre-testing of the instrument with a diverse and representative group of older adult community members and further identify any items in need of refinement. We will then perform preliminary clinical validation to ensure that clear performance differences are observed between healthy older adults and patients with a local clinical diagnosis of dementia. We will then further refine the instrument and repeat pre-testing and validation as needed until the team is in consensus that it is ready for the full baseline data collection.

OPEN IN VIEWER

Fig. 1

Summary of CVFS-SCAN approach to HCAP translation and adaptation. Figure 1 displays a summary of the procedures to be used for the translation and adaptation of the Harmonized Cognitive Assessment Protocol (HCAP) in the Chitwan Valley Family Study-Study of Aging and Cognition in Nepal (CVFS-SCAN). LASI-DAD, Longitudinal Aging Study in India - Diagnostic Assessment of Dementia. *Indicates informal back-translation at the time of HCAP revision.

We anticipate the need to address a number of challenges in the adaptation and validation of the HCAP in the CVFS context. Given the very limited evidence base on cognitive assessment in Nepal, our multidisciplinary translation and adaptation team will need to make decisions about cognitive test adaptation without a robust evidence base and will thus need to anticipate possible threats to cognitive test validity across our diverse study population. In addition, although we anticipate that Nepali will be an accessible cognitive assessment language for our study population, we need to scrutinize this assumption through our preliminary adaptation work. We may need to translate the HCAP into other local languages, which would then require additional preliminary validation work. We may also need to develop procedures for determining the most appropriate assessment language for multilingual individuals [50]. We may encounter challenges in creating appropriate informant ratings of cognitive decline, given concerns about stigma and the limited use of the concept of dementia in the local community. We will address each of these concerns with careful attention and community-engaged strategies.

Use of the HCAP instrument for dementia ascertainment in Nepal

The HCAP assessment in HRS and several of its international network of partner studies use an algorithmic approach to the classification of MCI and dementia [51, 52]. The critical components of this dementia classification algorithm include the classification of HCAP cognitive domain impairment using a local robust normative sample and the identification of informant-rated cognitive and functional decline. Efforts are currently underway in the HCAP international network to develop and optimize harmonized approaches to the classification of dementia with these algorithm components. For the Nepal context, the identification of a local robust normative sample will require careful consideration and collection of the specific factors in Nepal that may influence cognitive test performance. This work will require collaboration with local Nepali experts and with other HCAP network colleagues.

CVFS-SCAN aims to administer the HCAP to a sample of 4,000 community-dwelling older adults who have previously enrolled in the Chitwan Valley Family Study and have become age-eligible (≥50 years) during the study period. We selected the lower age limit of 50 years old for eligibility to account for the extended prodromal period that has been observed before clinical diagnosis of ADRD [53, 54]. This will allow us to link disease incidence in future waves with risk factor data from before disease onset with greater certainty. We will repeat HCAP assessments for all enrolled participants at 2-year follow-up. We will collaborate with our HCAP international network colleagues and our local clinical experts to determine the most appropriate identification of our robust normative sample, including brain health-relevant medical conditions in need of exclusion and sociocultural and demographic factors in need of adjustment. Longitudinal HCAP data will also be used when available for the identification of the robust normative sample.

Cross-national harmonization with other studies in the HCAP network

Our prospective collaboration with other studies within the HCAP network to optimize the comparability of our assessment and other survey data collection procedures will facilitate future cross-cultural cognitive data harmonization efforts. Data from CVFS-SCAN will be added to the growing list of the international network of HCAP studies. We will have the opportunity to co-calibrate our cognitive assessment with the other studies in the HCAP network. The statistical harmonization approach used for the HCAP allows for co-calibration of cognitive domains using both common and unique cognitive test data. This approach is optimal for cross-cultural cognitive aging studies as it allows each study to adapt the instrument according to its own needs and to balance the goals of within-study validity with cross-study harmonization. This will allow for powerful investigation of measurement equivalence and validity across national contexts, and critically, across diverse educational contexts, which offers the opportunity for tremendous advancement in cross-cultural neuropsychological assessment and in neuropsychological measurement more broadly. This work will also allow for substantive investigation of the complex determinants and impacts of cognitive impairment in older adults across national contexts. This work will also inform national policies needed to support the aging population for the early detection and treatment of ADRD. For example, systematic data on the prevalence of dementia in Nepal may reveal informative demographic patterns in dementia risk and highlight critical gaps in the clinical resources available for the identification and treatment of ADRD in older Nepali adults. In addition, we anticipate our data also to provide evidence for policymakers to develop comprehensive dementia care policies, which may focus on strategies to prevent cognitive impairment and ADRD in the general population, on increased clinical resources including more specialized training for health care providers and greater availability of clinical care of patients with ADRD, and programs to assist families who provide care for older Nepali adults with ADRD.