A systematic review of the prevalence of mental and neurocognitive disorders amongst older adults’ populace in Egypt

Mental and neurocognitive disorders are the leading cause of disabilities amongst the older adult populace worldwide. The population of the older adult in Egypt is fast growing. The national census in 2017 revealed a 2.56% increase in the older adult population from the 2006 census, and these figures are expected to double by year 2031. This study aims to review current evidence about the prevalence of mental and neurocognitive disorders amongst older adult population in Egypt. A systematic review was carried out, and 16 studies met the inclusion criteria outlined in this study. Four main mental and neurocognitive disorders were identified: depression, anxiety, cognitive impairment and dementia. The reported prevalence of depression, anxiety, dementia and cognitive impairment are 23.7–74.5%, 14.2–72%, 3.66–39.2% and 1.74–51.4%, respectively. Anxiety and depression were positively correlated with the female gender, increasing age and lower educational status. Also, cognitive impairment and dementia were positively correlated with age, illiteracy or low education. However, there appears to be inconsistencies in the diagnostic tools used. Egypt aging population is growing, and this research brings to forefront the scale of mental and neurocognitive disorders amongst the older adult population. This may help ensure evidence-based initiatives are put in place and priority is given to resource allocation for geriatric mental and neurocognitive disorders in Egypt. PROSPERO CRD42018114831

the future [10]. The national census in 2017 revealed a 2.56% increase from the 2006 census, and it is expected that the older adult population will double by year 2031, with a projected increase of 11.5% forecasted [11].
Internationally, research which focuses on examining trends of mental and neurocognitive disorders amongst older adults' group is limited. Furthermore, uncertainties exist about prevalence and correlates of mental and neurocognitive disorders amongst older adults especially in the Middle East region and Egypt in particular. Therefore, this study aims to review current evidence about the prevalence of mental and neurocognitive disorders amongst older adult population in Egypt. Furthermore, sociodemographic correlates of these conditions will be investigated.

Research questions
-What is the prevalence of mental and neurocognitive disorders amongst older adult populace in Egypt? -What are the sociodemographic correlates of mental and neurocognitive disorders amongst older adult populace in Egypt? -What are the current gaps in research about mental and neurocognitive disorders amongst older adult populace in Egypt?

Methods
This systematic review was developed according to the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) 2009 checklist [12] and the documented in a protocol (PROSPERO registration number: CRD42018114831) [13].

Search strategy
Electronic search was conducted on November 1, 2018, on five key databases (MEDLINE, PsychINFO, EMBASE, AMED and PUBMED). All databases were searched from their date of inception. In addition, we carried out key journal searching, citation searching, reference list scanning and relevant internet resources to ensure this systematic review contains all possible studies and also ascertain that studies not indexed in the five databases listed above were included in this review.
Prior to commencing the literature search, the Centre for Reviews and Dissemination as well as PROSPERO was accessed to identify if there had been a previous or ongoing systematic review carried out in Egypt. We found one study by Elshahidi et al. [3] that had carried out a systematic review on the prevalence of dementia in Egypt. However, our study is focused around mental and neurocognitive disorders using the International Classification of Disease (ICD) code 10, chapter 5 (F code) for mental and behavioural disorder, of which dementia is a type of neurocognitive disorders. Thus, our study covers a wider range of mental and neurocognitive disorders. Moreover, our inclusion and exclusion criteria are quite different from those recorded in the study by Elshahidi et al. [6].
An updated search was carried out on the 28th of December, 2018; no additional paper was found. Searches included MeSH and text word terms, with combination AND, OR, ADJ Boolean operator and truncation when necessary. The search terms used in this study is detailed in Fig. 1:

Inclusion and exclusion criteria
Studies were included in this review, if they were published in English, cross-sectional prevalence studies, and carried out on adult Egyptians aged 60 and above. In addition, mental and neurocognitive disorders had to be ascertained using any validated diagnostic tools. Validation of diagnostic tools is based on self-reported declaration of the original authors of included papers. We included studies conducted in hospitals (both inpatients and outpatients), residential homes, household surveys and other settings within Egypt. Studies which did not meet any of the criteria detailed above were excluded.

Search outcome
The literature search from the five databases revealed 343 papers; this was reduced to 296 after duplicates were removed. Of the 296 papers, 268 were excluded after screening the title and abstract resulting in 29 papers being retrieved for more details. We also carried out citation scanning, reference list scanning and internet source search and identified 9 more studies that were potentially relevant. Thus, a total of 38 full articles were accessed, and after matching it with the eligibility criteria, 16 studies were included in this study (see Appendix 1 for list of included studies). Figure 2 provides details of the search outcome. Papers were excluded due to age (< 60 years or unspecified) and unspecified diagnostic criteria and focusing more on chronic illnesses which co-morbid with mental and neurocognitive disorders. We identified two potentially relevant papers based on the title and abstract; however, both were later excluded when the full article could not be accessed.

Quality assessment
The 16 studies which met the inclusion criteria were assessed using the Herzog et al. [14]  assesses three main domains which are selection, comparability and outcomes. Each study could attain a maximum of 10 points. In this study, we took out question 4 about exposure risk as it was not relevant to this systematic review. Thus, quality assessment was based on 8 points. We also modified the grading as we realised that using the old grading will wrongly categorise those studies. The new grade scale was 0-2 point, unsatisfactory; 3-4 point, satisfactory; 5-6 point, good; and 7-8 point, very good. See Appendix 2 for the quality assessment of included studies.
Two reviewers assessed the risk of bias, and disagreements were resolved by discussion. Ten of the studies were classed as good, and five were very good with only one classed as unsatisfactory.

Data extraction
Two reviewers performed data extraction, and any disagreements were resolved by discussion. The following information was extracted: first author's name; year of publication; city, settings and duration of the study; demographics; number of participants; method of recruitment; diagnostic test used; mental and neurocognitive disorder studied; prevalence of mental and neurocognitive disorders; existence of co-mobility, current medication and smoking status.

Demographic information
All studies gathered demographic information such as age and gender. All but one study (study 10) (93.75%) reported information about the educational status of the participants. Marital and employment status were reported by more than half of the papers (56.25%), and less than half (37.5%) examined the living arrangement of participants. Furthermore, less than half of the studies (43.75%) indicated whether participants resided in urban or rural region. Table 1 contains the demographic details of all included studies.

Sample and population
The included studies were conducted in various cities in Egypt of which more than a quarter (37.5%) were from Cairo (studies 1, 2, 4, 11, 14 and 15). The studies were conducted in five major settings: residential homes, community surveys, inpatients, outpatients and community clubs. Of the 16 studies, a quarter (25%) was community surveys (studies 3, 7, 8 and 10). One study (study 5) was a mix of three settings (residential homes, community surveys and inpatient). It was difficult to specify the settings where two studies (studies 11 and 13) were conducted. All studies were cross-sectional prevalence studies, but two (studies 1 and 11) did not specify data collection period or duration.
Random sampling technique (56.25%) was more commonly used amongst the included papers. All community surveys used this method and provided in-depth information of how randomisation occurred. Three studies (studies 2, 6 and 13) claiming to use this technique were vague in their explanation. For instance, one study (study 2) claimed to have randomly selected the study setting but did not randomly select the participants. Five studies (studies 1, 9, 11, 12 and 14) making 31.25% did not specify the sampling technique, while two studies (studies 4 and 15) making 12% used convenience sampling.
Sample size varied widely with some studies using a small number and others using a large number. The community surveys tend to use a larger sample size. More studies (56.25%) provided details justifying their sample size. One study's (study 5) justification of sample size is unclear. The study made comparison between 3 settings (residential homes, inpatient wards and community survey in a slum area) and claimed 100 participants was the ideal sample size. Table 2 contains details about the sample and population of the included paper

Mental and neurocognitive disorders
This systematic review identified mental disorders, such as anxiety and depression, as well as neurocognitive disorders, such as cognitive impairment and dementia. Majority (81.25%) of the studies were conducted solely to estimate the prevalence of these illnesses; however, three studies (studies 1, 3 and 11) (18.75%) were conducted to estimate the association of these illnesses with other chronic health conditions. Two studies (studies 4 and 14) were aimed to validate Arabic version of diagnostic tools. For clarity, findings about mental disorders (depression and anxiety) will be presented first; then, neurocognitive disorders (cognitive impairment and dementia) will be discussed later.

Mental disorders
Eight of the 16 included studies reported the prevalence of mental disorders. Amongst these eight studies, two (studies 6 and 15) investigated the prevalence of both anxiety and/or depression with cognitive impairment. Hence, these studies are also reported in the cognitive impairment and dementia section later. Table 3 illustrates the prevalence of mental disorders reported in this study. Reported prevalence of depression ranged from 23.7 to 74.5% and anxiety from 14.2 to 72%, with mixed anxiety and depression disorder stated as 30% by one study (study 1). The manner two studies (studies 5 and 16) reported prevalence rate was cumbersome, and we will suggest that readers access these papers for further details. The prevalence of anxiety and depression reported varied with the data collection setting. It was noted that residential homes and community clubs had a lower prevalence compared with inpatient and household community survey.
Generally, depression was detected using the Geriatric Depression Scale (GDS). One study (study 16) detected depression from the participants by conducting psychiatric assessment using Structured Clinical Diagnosis (SCID-1) and arriving at a diagnosis based on the Diagnostic and Statistical Manual of Mental Disorders (DSM-IV). There was inconsistency in the GDS scale format used, more studies utilised the short form version, and only two studies (studies 3 and 15) used the long version. The GDS scale could sometimes be the Arabic version which had been previously validated or one simply translated to Arabic for the purpose of the study. The cutoff point for both the short and long versions of the GDS also varied amongst studies, with some using 5 or 6 points for the short and 10 or 11 for the long version, respectively. Anxiety was reported in three studies (studies 2, 15 and 16), and all three used different kinds of diagnostic tool.
Amongst the eight studies, all except one (study 15) indicated some type of correlation with sociodemographic status. See Table 4 for the relationship between depression and/or anxiety with sociodemographic status. It was mostly stated that depression and anxiety were positively correlated with the female gender. Furthermore, increasing age and lower educational status was reported as positively correlated with depression and anxiety. With regards to marital status, one study (study 2) reported a positive correlation of anxiety with married elderly. Others indicated a positive relationship with those who are divorced, widowed or separated. Four (studies 3, 6, 10 and 16) of the eight studies reported the residency status (urban or rural) of the study. All four studies reported more participants residing in rural than urban areas. Only one study (study 10) reported a significant increase in the prevalence of depression amongst the rural resident. Generally, it was indicated that both depression and anxiety was significantly associated with presence of co-morbid chronic diseases, poor perceived health status, having more than one chronic illness and loneliness. Table 5 presents the prevalence of dementia and cognitive impairment. In total, 10 studies reported findings, and this was a mix of cognitive impairment alone (studies 1, 4 and 13), dementia alone (studies 7 and 9), both cognitive impairment and dementia (studies 8, 11 and 14) and both anxiety and/or depression with cognitive impairment (studies 6 and 15).  The reported prevalence of dementia was 3.66-39.2%. All but one study (study 11) indicated a low prevalence rate of dementia. It is not particularly clear why the rate described in this study is particularly high (total dementia rate of 39.2%); perhaps it could be due to the data collection settings unclear. It was stated that participants in this study were recruited from the hospital without indicating if they are outpatient or inpatient. Although, this particular study (study 11) did indicate they excluded individuals with possible health conditions that may affect cognitive function. However, the description of the sample population is unclear and this may impact on prevalence rate of dementia reported. There was a wide range in the reported prevalence of cognitive impairment from 1.74 to 51.4%. Converse to findings about anxiety and depression, household community surveys reported a very low rate (1.74%) compared to all other settings.  Table 6 presents studies which show significant positive correlation between cognitive impairment and/or dementia with sociodemographic status. All but one study (study 15) of the 10 studies did not  report any significant correlation between dementia and cognitive impairment with sociodemographic status, respectively. Most of the studies indicated a positive correlation with age, illiteracy and low education with cognitive impairment and/or dementia. Two studies (studies 1 and 4) reported a positive correlation with female gender. Also, being unemployed and a manual worker was strongly associated with dementia. There was no report of a significant correlation between marital status and living arrangement. One study (studies 8) reported a positive relationship between living in urban and industrialised areas with dementia. Furthermore, three studies (studies 1, 13 and 14) stated a positive relationship between cognitive impairment and dementia with depression.

Risk factors of mental and neurocognitive disorders
Some studies reported some risk factors that can affect mental and neurocognitive disorders. Table 7 presents risk factors identified in this systematic review. Risk factors such as the use of medication, smoking and chronic health condition were discussed in the included studies. No study reported the alcohol status of the participants.
Only one study (study 5) (6.25%) indicated if the sampled population were on medications or not at the time of data collection. This study indicated that there was a positive correlation between increased number of medications and depression. Less than half of the studies (31.25%) stated the smoking status; however, there was no report of smoking having a positive association with anxiety, depression, cognitive impairments and dementia. Five studies (studies 4, 6. 10, 11 and 14) did not report if the sample populations had co-existing chronic health condition. Just one study (study 9) reported a significant association between chronic health condition (hypertension) and dementia.

Discussion
This systematic review reports current evidence about mental and neurocognitive disorders amongst the older adult population in Egypt. Two mental disorders (anxiety and depression) and neurocognitive disorders (cognitive impairments and dementia) were identified in this review. To the best of our knowledge, this is the first systematic review that considers the prevalence of both mental and neurocognitive disorders amongst the older adult populace in Egypt. Moreover, this study investigates sociodemographic factors that could potentially influence these conditions.
It is interesting to note that no single study in this review investigated the wider range of mental disorders such as dissociative disorders, misidentification disorder, obsessive-compulsive disorder, posttraumatic stress disorder, late-onset psychosis and other psychotic features that may be found amongst the older adult population. The strong cultural and religious influence towards mental disorders in Egypt could be attributed to studies not reporting data on substance and alcohol misuse in order to avoid stigmatisation and discrimination [15,16]. This opinion is supported by Ghanem et al.'s [17] study that examined mental disorders amongst individuals aged 18-64 years. These authors indicated that the low prevalence rate of alcohol and drug dependence reported may be inaccurate due to the fear of being ostracised by family members. Depression and anxiety are classed as 'common mental disorders' whose prevalence continues to rise globally and known as the largest contributor to global disability (WHO, 2017). This study found higher rates of depression and anxiety in females, increasing age and lower educational status. These findings are homogenous in the Middle East North Africa (MENA) countries and also globally [5,18,19]. Eloul et al. [18] indicated that the higher depression rate amongst females in MENA countries can be attributed to postpartum depression and the altering roles of females in these countries.
The estimated prevalence of cognitive impairment and dementia in this study were higher than those reported in previous studies [6,20]. This may be attributed to this review looking at studies conducted in various settings. For instance, Elshahidi et al. [6] study only looked at studies conducted via household surveys. Besides, a higher prevalence of dementia is often noted in institutionalised settings in comparison with other settings [20]. As expected, dementia and cognitive impairment rose with increasing age and illiteracy. This is consistent with previous studies [1,4,6,20,21].
The inconsistencies in the diagnostic tools may impact on the prevalence reported in some studies. This is because there might be variations in the way mental and neurocognitive disorders are defined and categorised [22]. Moreover, presence of risk factors such as smoking, alcohol, medication and co-existing chronic illnesses may also influence reported prevalence categorised [20][21][22].

Conclusion
Egypt's aging population is growing, and there is a need to prioritise research and interventions about mental and neurocognitive disorders amongst this population group. Egypt's strong cultural and religious influence also calls for public health interventions that educate and help to disabuse myths and misconceptions about mental and neurocognitive disorders. Recommendations for further research is for more studies that explore the wider range of mental disorders like psychosis, alcohol and drug dependence amongst the older adult population. Besides, prevalence studies that categorises cognitive impairment should be conducted. Also, studies should provide an extensive report of the influence of sociodemographic and other risk factors for mental and neurocognitive disorders.
The limitation of this study is that we might have excluded some potentially insightful studies because it did not meet the inclusion criteria. Nevertheless, this systematic review is unique as it contributes knowledge about the current evidence of the prevalence of mental and neurocognitive disorders amongst older adult population in Egypt. It also shed insights into sociodemographic correlates and possible risk factors of mental and neurocognitive disorders amongst the older adult population in Egypt. Moreover, a detailed description of the inclusion and exclusion criteria also will be valuable for other researchers who might want to explore mental and neurocognitive disorders in Egypt and other MENA countries.