Dementia is a broad term for neurodegenerative conditions that affect memory, thinking, language, other cognitive abilities and behaviour. 1 It significantly interferes with an individual’s ability to perform daily activities, causing substantial distress to both patients and their families. Reference Livingston, Huntley, Sommerlad, Ames, Ballard and Banerjee2 Recent estimates indicate that over 55 million people worldwide are living with dementia, Reference Schwarzinger and Dufouil3 with projections suggesting a rise to 139 million by 2050. Reference Lai, Jeon, McKenzie and Withall4 As global life expectancy increases, dementia is becoming a significant public health challenge, necessitating timely diagnosis and management. Reference Nichols, Steinmetz, Vollset, Fukutaki, Chalek and Abd-Allah5 Recent estimates place the prevalence of dementia in the United Arab Emirates (UAE) at 1.33% while that across the Arab world is approximately 0.7%, with a substantial associated economic burden. Reference Qassem, Itani, Nasr, Al-Ayyat, Javaid and Al-Sinawi6 The World Health Organization predicts that, in the Middle East and North Africa (MENA) region, the number of individuals living with dementia will reach 2.59 million by 2030, rising to over 6 million by 2050. 7 In the UAE, demographic transitions and an increasing elderly population further emphasise the need for improved dementia care.
A major challenge in diagnosing dementia is the prolonged delay between symptom onset and formal medical recognition. Reference Bamford, Lamont, Eccles, Robinson, May and Bond8,Reference Paterson and Pond9 Delayed diagnosis remains a persistent issue, influenced by various demographic, cultural and systemic factors. Reference Prince, Ali, Guerchet, Prina, Albanese and Wu10 Studies indicate that delayed diagnosis is associated with poorer prognostic outcomes, increased caregiver burden and higher healthcare costs. Reference Wilson, Cookson, Bosch, Zetterberg, Holtzman and Dewachter11 Early detection enables better management, access to support services and potential interventions that may slow disease progression. Reference Parker, Barlow, Hoe and Aitken12
Literature suggests that the time from symptom onset to diagnosis varies widely, depending on factors such as healthcare accessibility, cultural perceptions and health literacy. Reference Bernstein Sideman, Al-Rousan, Tsoy, Piña Escudero, Pintado-Caipa and Kanjanapong13 A significant contributor to delayed dementia diagnosis is the lack of structured screening programmes. Reference Helvik, Engedal, Šaltytė Benth and Selbæk14 Despite concerns regarding the accuracy of dementia screening tools and potential false positives, research consistently advocates for early screening due to the increasing prevalence of dementia and anticipated advancements in therapeutic interventions. Reference Milne15 Undiagnosed cognitive impairment can complicate the management of comorbidities, because patients with dementia frequently have serious coexisting conditions and take multiple medications. Reference Browne, Edwards, Rhodes, Brimicombe and Payne16 Without proper cognitive assessment, these conditions may be mismanaged, affecting treatment adherence and overall patient outcomes. Reference Milne15
In the UAE and countries with similar cultural backgrounds, the stigma associated with cognitive decline remains a major obstacle. Reference Javaid, Al-Zahmi and Abbas17,Reference El-Metwally, Toivola, Al-Rashidi, Nooruddin, Jawed and AlKanhal18 In societies where mental health issues are not openly discussed, families often underestimate early symptoms, attributing these to normal ageing rather than recognising them as signs of neurodegeneration. Reference Aminzadeh, Molnar, Dalziel and Ayotte19 Additionally, limited awareness among healthcare providers regarding mild cognitive impairment, lack of confidence in diagnosing patients with dementia and the absence of standardised diagnostic protocols further exacerbate delays. Reference Aminzadeh, Molnar, Dalziel and Ayotte19
To our knowledge, this is the first large-scale study to examine diagnostic delay in dementia within the MENA region. There is limited evidence available in this regard, primarily from countries with well-developed dementia care settings. Reference Helvik, Engedal, Šaltytė Benth and Selbæk14,Reference Zhao, Lv, Tuerxun, He, Luo and Chen20–Reference Fiske, Gatz, Aadnøy and Pedersen22 The UAE’s demographic profile, characterised by a rapidly ageing population and a large expatriate community, provides valuable insights not only nationally but also across other Gulf Cooperation Council (GCC) countries with similar healthcare challenges. This study seeks to analyse the average duration of cognitive and behavioural symptoms before a formal dementia diagnosis within a UAE-specific population. We also report the symptom profile of dementia patients in the UAE, highlighting common presenting issues and their impact on the duration of undiagnosed dementia. By investigating demographic and systemic contributors, including age, gender and presenting symptoms, this research provides critical insights for shaping future dementia care policies in the UAE. Furthermore, these findings contribute to the global discourse on dementia diagnosis by addressing the unique challenges faced in the UAE healthcare system, and by providing region-specific strategies to enhance early detection and intervention efforts for dementia.
Method
This study was conducted at the Behavioural Sciences Institute (BSI) of Al Ain Hospital, Al Ain City, UAE, one of the largest governmental hospitals in the eastern region of Abu Dhabi (the largest of the seven emirates of UAE). Its psychiatric services are well developed and serve the population of Al Ain City and surrounding towns, which numbers over 600 000. Reference Abdel Aziz, Sajwani, Galadari, Al-Ammari, AlHassani and Al-Nuaimi23 Ethical approval for this study was obtained from the Al Ain Hospital Research Ethics Governance Committee (approval no. AAHEC-11-20-030). Consent to participate was not applicable in this retrospective data-based study.
Study design and participants
Data were extracted retrospectively from patient electronic medical records for subjects diagnosed with dementia and presenting to the BSI over a 10-year period (2010–2019). Patient confidentiality was safeguarded by extracting only de-identified data from the electronic medical records. No personal identifiers were collected, and analyses were conducted on anonymised data-sets in accordance with ethics committee approval. The investigators then identified and extracted the relevant data needed to conduct their study.
We included male and female subjects over the age of 18 years, of all nationalities, with a principal documented diagnosis of dementia and who were registered with BSI. Subjects were excluded if they had a diagnosis of intellectual disability, traumatic brain injury or cerebrovascular event, or a concurrent mental disorder, as a principal diagnosis. Due to the retrospective nature of the study, the exact number of excluded cases could not be reliably quantified (see Supplementary Fig. 1, Additional file 1). From the electronic medical records, data were obtained for the following: demographics (age at the time of diagnosis, gender and nationality); presenting symptoms; principal diagnosis; and the duration of undiagnosed symptoms of dementia (DUS), i.e. the number of months between the participant’s first cognitive/behavioural symptoms and a formal diagnosis of dementia. Symptom onset was determined from caregiver-reported information, as documented in the electronic medical records at the time of clinical assessment. Documentation of symptom onset was narrative and not standardised, because caregiver accounts were recorded in routine clinical notes rather than in structured forms. Cognitive test scores, including the Mini-Mental State Examination (MMSE) and Montreal Cognitive Assessment (MoCA), were extracted from electronic medical records when available; however, documentation was inconsistent and the data were too sparse for inclusion in the analyses.
Measures
For the diagnosis of dementia, ICD-10 criteria were used. 24 ICD-10 describes dementia as a syndrome resulting from a brain disease, typically chronic or progressive, characterised by disturbances in various higher cortical functions such as memory, thinking, orientation, comprehension, calculation, learning ability, language and judgement, without affecting consciousness. 24 Dementia cases were coded as ‘unspecified’ when documentation did not permit a reliable subtype classification – for example, due to the absence of biomarker confirmation, lack of advanced neuroimaging or insufficient clinical detail in the records. Neuropsychiatric symptoms were evaluated as described in the Neuropsychiatric Inventory Questionnaire (NPI-Q). Reference Kaufer, Cummings, Ketchel, Smith, MacMillan and Shelley25 This covers the following symptoms: delusion, hallucination, elation/euphoria, agitation/aggression, disinhibition, irritability/lability, depression/dysphoria, anxiety, apathy/indifference, appetite/eating, night-time behaviour and aberrant motor behaviour.
Data tools
The study involved extraction of patient data from medical records by members of the research team, using a nine-item questionnaire (see Additional file 2), for each patient record that matched the aforementioned inclusion criteria. The questionnaire collected demographic information, including age at the time of diagnosis, gender and nationality. It also collected data about the diagnosis, presenting symptoms and their onset.
Statistical analysis
A priori power analysis was conducted using G*Power 3.1.9.4 software for Windows (Heinrich Heine University Düsseldorf, Düsseldorf, Germany; https://www.psychologie.hhu.de/arbeitsgruppen/allgemeine-psychologie-und-arbeitspsychologie/gpower). Reference Faul, Erdfelder, Buchner and Lang26 For analysis of variance (ANOVA), we specified 13 groups (representing the 13 symptoms), a medium effect size, a desired power of 0.8 and an alpha level of 0.05. The analysis indicated that a total sample size of 288 participants would be sufficient to detect a medium effect with adequate power. Our sample size (N = 825) far exceeds this requirement, thereby ensuring robust statistical validity. Additionally, we performed a power analysis for a two-tailed t-test with a medium effect size, alpha of 0.05 and power of 0.8, which yielded a required sample size of 128. Again, with 825 participants, our sample size was more than adequate for this analysis too.
Values of the data analysed are expressed as means ± s.d. Where relevant, t-tests were carried out and P-values equal to or less than 0.05 were considered statistically significant. The assumption of homogeneity of variance for a two-tailed independent t-test was assessed using Levene’s test. In the case where the assumption of homogeneity was not met after conducting Levene’s test, Welch’s test was performed. A post hoc Tukey’s test was conducted to determine the statistical significance of the differences in DUS between initial symptoms.
A multiple linear regression analysis was conducted to examine the association between demographic factors, dementia types and initial symptoms with the dependent outcome (DUS score). The model included predictors such as age at diagnosis, gender, nationality (Emirati versus non-Emirati), dementia subtype and initial symptom clusters (e.g. agitation, hallucinations, depression, motor disturbance). Standardised estimates, standard errors, confidence intervals and significance levels were calculated, and overall model fit was evaluated using R, R 2, F-statistics and P-values.
Results
Demographic and clinical characteristics
Eight hundred and twenty-five (825) patients were identified using the inclusion criteria. The mean age of diagnosis was 78 years (s.d. = 11.1). The range of afflictions spanned from the youngest patient, at 39 years, to the oldest, at 117 years. The demographic characteristics of the sample are presented in Table 1.
Table 1 Demographic characteristics of the study sample (N = 825)
The most frequent type of dementia diagnosed was unspecified dementia (42.1%), followed by Alzheimer’s disease (40.6%). Table 2 depicts the distribution of different types of dementia among our participants.
Table 2 Types of dementia in the study sample (N = 825)
DUS
The mean DUS was 2.89 years (34.7 months; s.d. = 28.8), with a median of 24 months (interquartile range (IQR) 12–48). A small number of extreme outliers were identified (>121 months, >3 s.d. above the mean), but sensitivity analyses excluding these cases (n = 822) yielded unchanged medians (24 months, IQR 12–48) and minimal changes in means (34.2 v. 34.6), confirming the robustness of the findings. These findings are presented in Supplementary Table 1 (Additional file 3).
The mean DUS among female study participants was 35 months (s.d. = 28) compared with 34 months (s.d. = 29.6) in male participants. Both males and females had a median DUS of 24 months (IQR 12–48). A two-tailed independent t-test was conducted to assess the effect of gender on DUS. No significant difference in DUS was found between males and females (t(823) = −0.41, P = 0.69, Cohen’s d = −0.03), as depicted in Fig. 1. Given the large sample size (N = 825), the normality assumption was not necessary. The homogeneity of variance assumption was met following application of Levene’s test (F(1, 823) = 0.01, P = 0.95).
Fig. 1 Gender comparison regarding the duration of undiagnosed symptoms (DUS) of dementia.
The mean DUS among Emirati participants was 35.6 months (s.d. = 29.7), compared with 33 months (s.d. = 27) in non-Emiratis. The median DUS (24 months, IQR 12–48) was similar between Emirati and non-Emirati participants. A two-tailed independent t-test was conducted to assess the effect of the nationality group on DUS; there was no significant difference between Emiratis and non-Emiratis in DUS (t(823) = 1.33, P = 0.19, Cohen’s d = 0.10), as shown in Fig. 2. Given the large sample size (N = 825), the normality assumption was not necessary. The homogeneity of variance assumption was met following application of Levene’s test (F(1, 823) = 1.13, P = 0.29).
Fig. 2 Nationality group comparison in the duration of undiagnosed symptoms (DUS) of dementia between Emirati and non-Emirati study participants.
The cohort was divided into two groups: those under 65 years old and those 65 years or older at the time of diagnosis, in line with the traditional cut-off used in literature. Reference Van De Veen, Bakker, Peetoom, Pijnenburg and Papma27 The mean DUS among study participants under 65 years old at the time of diagnosis was 24.5 months (s.d. = 21.4), compared with 35.6 months (s.d. = 29.2) among those older than 65. Those under 65 years had a shorter median DUS of 18 months (IQR 12–30), while those aged 65 and above had a median of 24 months (IQR 12–48). The assumption of homogeneity of variance for a two-tailed independent t-test to assess the effect of age group (younger than 65 years, 65 or older) was not met following the application of Levene’s test (F(1, 823) = 16.8, P < 0.001; consequently, Welch’s test was conducted to assess the effect of age group on DUS. There was a significant effect on age group (t(92.9) = 3.98, P < 0.001), with an effect size of d = 0.43, where individuals 65 years or older had longer DUS, as shown in Fig. 3.
Fig. 3 Age group comparison of the duration of undiagnosed symptoms (DUS) of dementia between participants aged 65 years and older and those under 65 years.
One-way ANOVA was conducted to assess the effect of initial symptoms (forgetfulness, agitation/aggression, delusions, depression/dysphoria, irritability/lability, motor disturbance, appetite/eating, hallucinations, apathy/indifference, anxiety, night-time behaviours, elation/euphoria and disinhibition) on DUS. Forgetfulness (60.2%) was the most common presenting symptom. Table 3 depicts DUS according to symptom group in our sample.
Table 3 Duration of undiagnosed symptoms (DUS) according to initial symptoms (N = 825)
IQR, interquartile range.
High standard deviations in some categories reflect small subgroup size (e.g. apathy/indifference, n = 8) and the potential influence of outliers.
There was a significant effect of initial symptoms on DUS (F(11, 69.2) = 22, P < 0.001, η 2 = 0.14). A post hoc Tukey’s test was conducted to determine the statistical significance of the differences between each pair of initial symptoms. There was a statistically significant mean difference in DUS between forgetfulness and agitation/aggression (19.7, P < 0.001), between forgetfulness and irritability/lability (16.24, P < 0.003), between forgetfulness and motor disturbance (22.97, P < 0.001), between forgetfulness and appetite/eating (26.53, P = 0.002), between forgetfulness and hallucinations (32.65, P < 0.001) and between forgetfulness and disinhibition (34.52, P < 0.017).
The overall regression model was statistically significant (R = 0.509, R 2 = 0.259, F(21, 803) = 13.4, P < 0.001), explaining about 25.9% of variance in the dependent outcome. Age at diagnosis was a significant positive predictor (P < 0.001). Several initial symptom presentations showed negative associations when compared with the reference category: forgetfulness, including agitation/aggression; delusions; irritability/lability; motor disturbance; appetite/eating changes; hallucinations; anxiety, night-time behaviours; disinhibition; and depression/dysphoria (all P < 0.05). In contrast, apathy/indifference was not significant. Regarding gender, females did not differ significantly from males. Similarly, Emirati status showed no effect when comparing non-Emirati individuals with Emiratis. For dementia subtypes (reference, Alzheimer’s disease), Parkinson’s disease dementia demonstrated a significant negative association (P = 0.045). Other subtypes, including vascular dementia, Lewy body dementia, frontotemporal dementia, mixed dementia and posterior cortical atrophy, did not differ significantly from Alzheimer’s disease in this model. Full regression results are provided in Supplementary Table 2 (Additional file 4).
Discussion
This study investigated the average duration of cognitive, behavioural and neuropsychiatric symptoms prior to participants receiving a formal dementia diagnosis. Additionally, we report the initial symptom profile of dementia patients in the UAE and its association with DUS.
Across the sample, the mean duration between symptom onset and confirmed diagnosis of dementia was 2.89 years, or 34.7 months, indicating a substantial delay in diagnosis and considerable variability among individuals. The sensitivity analysis excluding extreme outliers confirmed the robustness of these findings, showing that the overall conclusions are not driven by a small number of atypical cases. Several rigorously conducted studies have reported the mean duration from symptom onset to diagnosis of dementia. Reference Helvik, Engedal, Šaltytė Benth and Selbæk14,Reference Zhao, Lv, Tuerxun, He, Luo and Chen20–Reference Fiske, Gatz, Aadnøy and Pedersen22 The mean diagnostic delay of nearly 3 years in our cohort is notably longer than that reported in high-income countries, such as Norway, where mean delay was approximately 20 months. Reference Helvik, Engedal, Šaltytė Benth and Selbæk14 This disparity highlights the impact of health system and sociocultural factors in the UAE and the wider region, underscoring the need for tailored strategies to narrow the diagnostic gap.
A high proportion of cases (42.1%) were classified as ‘unspecified dementia’. This reflects challenges in subtype classification in routine clinical practice in the region, where biomarker testing and advanced imaging are not always available. Similar findings have been reported by El-Metwally and colleagues in Middle Eastern cohorts, underscoring the need for improved diagnostic infrastructure and more consistent application of subtype criteria. Reference El-Metwally, Toivola, Al-Rashidi, Nooruddin, Jawed and AlKanhal18 Our findings highlight the under-recognition of dementia in its early stages and the variability of diagnostic pathways, underscoring the need for greater public awareness and primary care screening. Practical region-specific strategies include piloting of brief cognitive screening tools, embedding dementia messages into national health campaigns and providing targeted training for general practitioners and community health workers. These represent feasible interventions within existing health system structures in the UAE and GCC region to reduce diagnostic delays.
Delays in identifying and diagnosing dementia in the UAE and the broader Middle Eastern context are influenced by a confluence of cultural, systemic and clinical factors. Reference Aminzadeh, Molnar, Dalziel and Ayotte19 Cultural attitudes towards ageing and cognitive decline play a pivotal role in postponing dementia diagnoses. In Middle Eastern societies, public knowledge gaps and stigma, including misattribution of early cognitive symptoms to ‘normal ageing’, are well documented. Reference Alhazzani, Alqahtani, Alqahtani, Alahmari and Zarbah28,Reference Hakami, Madkhali, Saleh, Ayoub, Moafa and Moafa29 This perception leads to the normalisation of early dementia symptoms, resulting in postponed medical consultations. Reference Alhazzani, Alqahtani, Alqahtani, Alahmari and Zarbah28 Furthermore, the stigma associated with mental health conditions can discourage families from seeking timely medical advice due to fears of social repercussions. While mental health and dementia issues are stigmatised at the global level, Reference Benbow and Jolley30 unique cultural challenges within Middle Eastern countries, such as the UAE, should also be considered in relation to the results of this current study. These include pervasive and negative culturally held views surrounding mental illness, with a strong belief that such experiences are the result of God’s punishment, evil spirits or black magic. Reference Zolezzi, Alamri, Shaar and Rainkie31 Stigma and cultural context may differentially influence help-seeking among Emirati and expatriate families. For expatriate communities, language barriers and difficulty in navigating services can hinder early presentation. Furthermore, the lack of standardised, culturally adapted diagnostic tools in Arabic, coupled with high illiteracy rates among older adults, hinder accurate and timely detection. Reference Phung, Chaaya, Waldemar, Atweh, Asmar and Ghusn32 These findings suggest that sociocultural norms, language and health literacy shape diagnostic pathways differently across groups in the UAE and wider GCC, helping to explain the longer delays associated with memory-led presentations.
In this study, individuals over the age of 65 years were statistically more likely to experience a longer DUS compared with those under 65. Age-related biases and misconceptions significantly impact the timeliness of dementia diagnoses. Reference Lai, Jeon, McKenzie and Withall4 Older adults may encounter delays due to the normalisation of cognitive decline with ageing, while younger individuals face challenges stemming from the atypical nature of their symptoms. Reference van Vliet, de Vugt, Bakker, Pijnenburg, Vernooij-Dassen and Koopmans33 A deficit in health literacy concerning conditions such as dementia remains a significant challenge in the UAE. Reference Nair, Sreedharan, Satish and Ibrahim34 However, considering that there was no statistically significant difference between Emirati and non-Emirati participants regarding DUS in this study, it can be inferred that the issue of stigma and limited health literacy is not confined to the UAE but is a global concern across cultures, indicating the universal and indiscriminate nature of dementia as a neurodegenerative condition. Reference Cova, Markova, Campini, Grande, Mariani and Pomati35
Our study revealed no statistically significant difference between males and females in regard to DUS. This is intriguing, because gender is a significant determinant in accessing mental health services, characterised by a multidimensional disparity involving social norms, help-seeking behaviours, healthcare biases and systemic inequalities. Reference Bener and Ghuloum36 Previous research has identified female gender as being associated with delays in the referral and diagnosis of dementia, reflecting an implicit bias that links cognitive decline more strongly with male behavioural presentation. Reference Lee, Lee and Kim37
The observed lack of statistically significant differences in DUS between nationalities and genders may indicate underlying issues in the UAE health services when addressing dementia, given that the common denominator across the sample was the Emirati health service accessed by the study participants. The knowledge of clinicians has been noted in other literature as being vital for ensuring timely referrals. Reference Arsenault-Lapierre, Henein, Rojas-Rozo, Sourial, Bergman and Couturier38 There is a perceived lack of understanding within UAE health services regarding the symptoms and challenges of living with dementia. Reference El-Metwally, Toivola, Al-Rashidi, Nooruddin, Jawed and AlKanhal18 It can be argued that, alongside the general population, frontline health professionals may also need increased awareness about dementia. Public health bodies in the UAE should also aim to promote campaigns that destigmatise dementia and highlight the importance of early help-seeking behaviours.
Concerning presenting symptoms, the sample was categorised into various groups (forgetfulness, agitation/aggression, delusions, anxiety, etc.). The results revealed significant mean differences in DUS between forgetfulness and a variety of other presenting symptoms, including aggression, lability, motor disturbance, appetite issues, hallucinations and disinhibition. The most frequent clinical presentation of dementia is progressive loss of memory. Reference Arvanitakis, Shah and Bennett39 While most individuals can recognise forgetfulness as a core symptom of dementia, other symptoms may be more likely to elicit a response from key stakeholders. These secondary presenting symptoms may diverge significantly from common perceptions of dementia, prompting more urgent reactions from both family members and clinicians. For instance, hallucinations and motor disturbances are crucial components of Lewy body dementia Reference Walker, Possin, Boeve and Aarsland40 yet they can have a far greater impact on daily life, particularly in the early stages, than forgetfulness. Reference Outeiro, Koss, Erskine, Walker, Kurzawa-Akanbi and Burn41 As previously mentioned, this may closely correlate with the level of dementia literacy among the public and professionals alike.
To further explore these associations, we conducted a multivariable regression adjusting for age, gender, nationality, dementia subtype and presenting symptoms. In this adjusted model, older age and memory-related presentations remained significant independent predictors of longer diagnostic delays. Gender and nationality were not significant predictors once other factors were accounted for, consistent with the univariate results. Dementia subtype was not independently associated with DUS, although the large proportion of unspecified cases may have limited the statistical power. These findings reinforce the observation that age and symptom profile, rather than demographic factors, exert the strongest influence on diagnostic delay.
Importantly, the persistence of symptom-related effects in the adjusted model highlights how clinical presentation shapes pathways to diagnosis. Forgetfulness, the most common presenting feature, was independently associated with prolonged delays whereas behavioural and psychiatric symptoms (e.g. hallucinations, agitation) prompted earlier clinical recognition. This suggests that both families and clinicians may normalise memory loss as part of ageing, while disruptive behavioural features trigger more urgent help-seeking and referral.
This is the first large-scale study in the region to quantify delays in dementia diagnosis and explore symptom profiles. The large sample size (N = 825) and the use of real-world clinical data enhanced the ecological validity and relevance of the findings for healthcare planning. However, limitations exist; the retrospective design restricts symptom onset accuracy, relying on caregiver-reported data in medical records. This may introduce recall bias, because caregivers could confuse early symptoms with normal ageing or report them inaccurately. Variability in recording caregiver symptom reports, due to non-standardised documentation, may have impacted accuracy. Prospective studies with standardised symptom logs could yield more precise diagnostic delay estimates. Although we examined demographic and symptom-related predictors, unmeasured confounders such as education, comorbidities and socioeconomic factors were not consistently documented in the medical records and could not be included, thereby limiting our ability to account fully for factors contributing to delays in dementia diagnosis. Cognitive test scores (MMSE, MoCA) were inconsistently documented in the retrospective data-set and could not be analysed, limiting assessment of the link between cognitive severity and diagnostic delay, and these should be addressed in future prospective studies. Although our study focuses on patient factors, system-level barriers may also contribute to delayed dementia diagnosis in the UAE. Reference Al-Ramlawi42 Despite a large sample size, this study relied on a single specialist centre, which may have limited its generalisability to other healthcare settings. The UAE’s mixed healthcare system, comprising both public and private providers, features distinct referral pathways, resources and costs. Research in the Arab region highlights financial, administrative and organisational barriers to accessing specialist or high-quality care. Reference Kronfol43 These structural differences were not captured and are a further potential limitation. The lack of qualitative data constrained insight into caregivers’ perspectives and lived experiences behind the delayed diagnosis, because symptom classification using NPI-Q might have excluded subtle or culturally framed symptoms not captured in inventories. Future mixed-methods research, including qualitative interviews with families and clinicians, will be important in exploring perceptions of ‘forgetfulness’ and help-seeking norms.
This study reveals a delay of nearly 3 years between symptom onset and dementia diagnosis among older UAE adults. Gender and nationality did not affect delays, but older age and symptoms did. It highlights gaps such as low dementia literacy, stigma and lack of structured screening pathways in the UAE. This study underscores the importance of examining global issues through a local lens to identify potential improvements. Reducing diagnostic delays can also lessen the health economic burden of dementia in the UAE and the Arab region. Future research should incorporate qualitative methods to explore personal, cultural and healthcare system-level barriers from the perspectives of both families and clinicians.
Supplementary material
The supplementary material for this article is available online at https://doi.org/10.1192/bji.2025.10077.
Data availability
The data that support the findings of this study are available from the corresponding author, S.F.J., on reasonable request.
Author contributions
S.F.J.: conceptualisation, methodology, writing (original draft preparation), supervision; S.A.: data curation, methodology; N.I.: methodology, data analysis; F.A.: data curation and analysis; S.A.H.: data curation and analysis; F.A.M.: writing (reviewing and editing), data curation, methodology; K.A.A.: conceptualisation, supervision; G.A.: methodology, data analysis, writing (original draft preparation); Z.S.: conceptualisation, writing (reviewing and editing), supervision.
Funding
This research received no specific grant from any funding agency, commercial or not-for-profit sectors.
Declaration of interest
K.A.A. is a member of the BJPsych International editorial board. He did not take part in the review or decision-making process of this paper. All other authors have no conflicts of interest to declare.
Ethical standards
Ethical approval for this study was obtained from the Al Ain Hospital Research Ethics Governance Committee (approval no. AAHEC-11-20-030). Consent to participate is not applicable because this is a retrospective data-based study. Patient confidentiality was maintained throughout: all data were de-identified prior to extraction, no personal identifiers were collected and analyses were conducted on anonymised data-sets in accordance with ethical standards.
Open access
eLetters
No eLetters have been published for this article.