Reliability and Validity of the Turkish Version of the Mini Depression Status Test (MDST-TR)
Abstract
Abstract:Objectives: Depression is a frequent mental disorder. Easy and fast tools for screening depression are of high relevance. Common screenings require sufficient abilities to understand the information on the questionnaires. For patients with a migration background, these factors can be problematic. Method: The present cross-sectional study was conducted with 329 participants (293 healthy and 36 clinical participants). We analyzed the internal consistency reliability using Cronbach’s α coefficient and the test-retest reliability using Spearman’s rank correlation coefficient. We determined the cut-off scores using receiver operating characteristics (ROC). Results: MDST shows high construct validity (rs = −.55, p < .001), satisfying internal consistency reliability (Cronbach’s α for overall sample = .76) and good reliability in terms of test-retest reliability (rs = .55) among Turkish migrants. Discussion: The MDST is a valid screening instrument for depressive symptoms in clinical and healthy Turkish migrants living in Germany.
Depression is a common mental disorder affecting an estimated 322 million people worldwide (World Health Organization, 2017). The COVID-19 pandemic proved to be a risk factor for increasing the prevalence of depressive disorders (Santomauro et al., 2021). Furthermore, depression is one of the five leading causes of years lived with disability (YLD; Vos et al., 2017). Because the prevalence is generally constant between cultures, the diagnosis and treatment of depression are equally important in all populations (World Health Organization, 2017). Approximately 4 million people currently suffer from depression in Germany (World Health Organization, 2017). Notably, many elderly people show signs of clinical late-life depression (Butters et al., 2004; Kang et al., 2014). Furthermore, depression also plays a critical role poststroke, affecting approximately 33% of all stroke survivors (Ladwig et al., 2018; Towfighi et al., 2017) as well as patients with neurological diseases, where the prevalence of depression is particularly high, such as dementia (Zhao et al., 2016), Parkinson’s disease (Reijnders et al., 2008), and multiple sclerosis (Boeschoten et al., 2017).
Data indicate a higher prevalence of depression in first-generation migrants aged 50 years or older living in Europe (Aichberger et al., 2010). More specifically, there is a high lifetime prevalence of mental disorders among Turkish individuals with a migration background living in Germany (Dingoyan et al., 2017). Notably, a recent systematic review by Igde et al. (2019) revealed higher prevalence rates of mental illnesses in Turkish migrants because of various factors, including low acculturation levels (Ünlü Ince et al., 2014), low socioeconomic status (Knesebeck et al., 2017; Schlax et al., 2019), discrimination (Morawa & Erim, 2014), and inadequate mental health literacy (Ganahl et al., 2016). In line with these data, migration background and living in a foreign country are associated with stress, which has different possible consequences, such as depression (Foo et al., 2018).
A diagnosis of depression is mainly based on key clinical symptoms, including depressed mood, loss of interest, and reduced energy (American Psychiatric Association, 2013; World Health Organization, 2019). Clinical diagnoses may be based on screening instruments and structured or semistructured interviews (Smith et al., 2013). In this context, simple and quick tools for the initial screening of depression are highly relevant. In practice, questionnaires such as the Beck Depression Inventory-II (BDI-II; Beck et al., 1996), the DESC (Rasch-based Depression Screening; Forkmann et al., 2009), the HADS (Hospital Anxiety and Depression Scale; Zigmond & Snaith, 1983), and the DASS (Depression Anxiety Stress Scales; Lovibond & Lovibond, 1995) are often used to detect depressive symptoms. However, these screening instruments require sufficient ability to differentiate and understand the information on the questionnaires. This requirement is especially problematic for patients with migration backgrounds, language disorders, mild cognitive impairment, or dementia. Our study also found evidence that education is an important factor in cognitive test performance (Anapa et al., 2021). It is well known that migrants face barriers in accessing health services, mainly because of language deficits and social disadvantages (Arias-Uriona & Guillén, 2020; Klein & von dem Knesebeck, 2018; Steinbach, 2018). Furthermore, health literacy plays an important role in the utilization of health services. Health literacy can be defined as the ability to find, understand, and implement health-relevant information (Sørensen et al., 2012). Migrants generally have lower health literacy, stemming particularly from their lower levels of education (Berens et al., 2016). From the healthcare professional’s perspective, there are necessary steps to improve healthcare for migrants (Baumeister et al., 2021), such as ensuring the availability of a nonverbal tool for screening depression in patients with a migration background or cognitive impairment.
This study provides data for a nonverbal depression screening instrument for the specific group of older Turkish migrants in Germany with and without cognitive impairment. For this purpose, we validated a recently developed screening instrument, the Mini Depression Status Test (MDST), which shows good psychometric properties in patients with various neurological diseases (Christen et al., 2019), including reliability (Cronbach’s α = .84) and validity (GDS: r = −.47). The MDST was developed over approximately 5 years in the Clinic and Polyclinic for Neurology at the University Hospital of Cologne. Criteria such as performance, comprehensibility, and scaling were tested on patients and healthy subjects. The observations and results from this developmental phase were used to modify the instrument and develop a final version. Figure 1 shows the final version of the MDST for females.
The main aims of the study were to examine the psychometric properties of the MDST adapted for the Turkish population and to determine cut-off scores for depressive symptoms for a sample of Turkish migrants in Germany.
Materials and Methods
We conducted the present cross-sectional study with healthy senior Turkish migrants living in Germany (n = 293) and patients of the University Hospital Cologne’s Department of Neurology with a Turkish migration background and cognitive impairment (n = 36) in North Rhine-Westphalia. Recruitment took place from 2017 to 2019 after the study was approved by the University Hospital Cologne’s Ethical Committee (16-249). The neurological patients were recruited and tested at the Department of Neurology, University Hospital Cologne, and healthy participants were recruited at different institutions in North-Rhine Westphalia. All participants gave written informed consent and completed questionnaires assessing sociodemographic details. Testing was conducted in the participants’ mother tongue by bilingual and bicultural German-Turkish psychologists (authors GA and ÜSS) and trained bilingual students of medicine or psychology, supervised by authors GA and ÜSS.
Inclusion and Exclusion Criteria
Inclusion criteria for all participants were residency in Germany; native Turkish with a migration background; aged 50 years or older; normal, only slightly restricted, or corrected-to-normal vision and hearing; and the provision of written consent to participate in the study. Additionally, exclusion criteria for healthy seniors were self-reported cognitive impairment or other past or current diagnosed neurological or psychiatric illnesses and cognitive disorders.
Assessment of Depression and Cognition
The MDST is a new screening tool for depression specifically developed for neurological patients and patients with low levels of education or limited reading skills. A more detailed description of the MDST and its development is described elsewhere (Christen et al., 2019). It is available in two versions (for males and females) and consists of 3 items assessing the key symptoms of depression (mood disturbance, loss of interest, and loss of motivation). Each item has a short heading and is visualized by an image. Items evaluating the patient’s state over the previous 2 weeks are answered on a bipolar rating scale from 1 (negative) to 5 (positive). A total score is determined to evaluate the MDST, which can assume a value from 3 to 15. All items of the MDST are equally weighted and thus contribute equally to the overall score. Lower values are associated with depressive mood and higher values with a normal affective state. The MDST takes about 2 minutes to complete (the MDST can be requested from the authors GA and ÜSS).
Furthermore, participants received the Turkish version of the Geriatric Depression Scale short form (GDS-15; Yesavage et al., 1983), a validation instrument consisting of 15 yes/no items. Each answer can be counted as 1 or 0 points, with a score of 6 to 10 points indicating mild to moderate and 11 to 15 points severe depressive symptoms.
The MoCA (Nasreddine et al., 2005) was used in the Turkish version (Selekler et al., 2010) to assess cognitive state. A maximum of 30 points can be achieved, and participants are classified as cognitively impaired if they score less than 21. The cognitive domains tested by the MoCA are visuospatial/executive functions, naming, verbal short and long-term memory, attention, language, and abstraction (Nasreddine et al., 2005).
Statistics
We performed the statistical analyses using IBM SPSS Statistics 28 for Windows. Normal distribution was tested using the Kolmogorov-Smirnov test. General participant characteristics were expressed using means and ± SD for continuous variables and frequencies (%) for categorical variables. Possible differences in demographic data (age, sex, education), cognitive functioning, and depressive symptoms between the two groups (healthy and clinical) and within groups were analyzed using t-tests or Chi-square tests where appropriate, each with a significance level of α = .05. Spearman’s rank correlation coefficients were calculated between MDST and screening tools for depression to assess convergent validity and between depressive symptoms (MDST) and cognitive functioning (MoCA) to assess the association between cognitive impairment and depression (Ismail et al., 2017). We analyzed internal consistency reliability using Cronbach’s α coefficient and test-retest reliability using Spearman’s rank correlation coefficient. Furthermore, cut-off scores were determined using receiver operating characteristics (ROC) analyses. Effect sizes were determined and evaluated following Cohen (Cohen, 1992).
Results
Sample Characteristics
329 participants out of an initial group of 358 participants were included in the final analysis. Data from 29 participants were excluded because of missing data on GDS and MDST. The final sample included 36 patients, 19 (52.8%) males and 17 (47.2%) females, with a mean age of 70.92 ± 9.93, and 293 healthy seniors, 102 (34.8%) males and 191 (65.2%) females, with a mean age of 56.32 ± 7.33 years. Both groups’ demographic characteristics and MoCA, GDS, and MDST results are shown in Table 1. Participants in the clinical sample were significantly older, less educated, scored lower in the MoCA, and were more depressed than participants in the healthy sample (all p < . 001). 90 in the healthy group and 21 in the clinical group were depressed as defined by the GDS (five points or more).
Construct Validity and Divergent Validity
Highly significant correlations between MDST and GDS scores with large effect sizes were shown for the clinical (rs = −.68, p < .001, n = 36) and healthy (rs = −.52, p < .001, n = 293) groups as well as for the overall sample (rs = −.55, p < .001, n = 329). MDST and MoCA scores were also significantly correlated in the healthy group (rs = .29, p < .001, n = 293) with a small effect size. No significant correlation between MDST and MoCA scores was observed in the clinical subgroup (rs = .18, p < .30, n = 36); the effect size was also small.
Internal Consistency
Cronbach’s α coefficient was calculated for the MDST and its items to assess internal consistency. The internal consistency of the MDST in the clinical sample was good, with Cronbach’s α = .79. The item selectivity on all three scales lay between .60 and .67 and can therefore be rated as sufficient (Table 2). Cronbach’s α for the MDST in the healthy sample was satisfactory at .75. The item selectivity on all three scales lay between .50 and .63 and can therefore be rated as sufficient (Table 2). Cronbach’s α for the overall sample was also satisfactory at .76.
Test-Retest Reliability
Test-retest reliability for the MDST was assessed for the healthy sample with Spearman’s rank correlation; the retest was conducted after 2 weeks. It was highly significant, with a large effect size (rs = .55, p = .01, n = 36).
Determination of Cut-off Scores
For the ROC analyses, we used the GDS to classify whether or not an individual had depression. All subjects in the clinical and healthy samples who reached a GDS value of six or higher were classified as possibly depressed (Gauggel & Birkner, 1999). The healthy group included 292 subjects who scored an average of 3.64 (SD = 3.53) on the GDS and 10.78 (SD = 2.55) on the MDST. In the clinical group (n = 36), the average GDS score was 5.83 (SD = 3.71) and 9.31 (SD = 3.23) for the MDST. ROC analyses were conducted for both groups separately. For the clinical sample, ROC analysis determined a cut-off score of 10 for the MDST, with a sensitivity of .76 and a specificity of .67. The area under the curve (AUC) was .77 (see Figure 2). For the healthy sample, a cut-off score of 11, with a sensitivity of .73 and a specificity of .67, was determined; the AUC was .78 (see Figure 3). In both groups, the AUC value indicated a good separation of individuals with depression as classified by the GDS.
Discussion
In this cross-sectional study, we validated the MDST for Turkish migrants aged 50 years or older living in Germany. The main results were that (1) the MDST is a valid screening instrument for depressive symptoms in clinical and healthy Turkish migrants living in Germany, as demonstrated by its high construct validity operationalized with the GDS. Furthermore, (2) the test-retest reliability and internal consistency of the MDST in Turkish migrants (healthy and clinical samples) were good or satisfactory.
Sample
The results show that the two samples differed in age, education, cognitive functioning, and depression. Specifically, the patients were older, less educated, more limited in cognitive functioning, and more depressed than our healthy group. However, this was an expected finding, as patients who visit neurological services because of diseases and deficiencies that usually occur in old age. In contrast, the healthy sample was recruited from other facilities.
Psychometric Criteria and Cut-off Values
The MDST showed high construct validity as a graphically based instrument, comparable to the GDS as a verbally based instrument in both the healthy and patient groups. These results correspond with those from the validation study of the MDST in a German sample, where the MDST significantly correlated with established depression screening tools, with large effect sizes (Christen et al., 2019). Our analyses only showed a small and nonsignificant correlation between the MDST and MoCA in the clinical subgroup. However, our sample scored low in the MoCA overall (15.31 ± 6.71), and it is well known that patients with depression are more affected by cognitive deficits (Rock et al., 2014). Furthermore, in a recent systematic review and meta-analysis, Ismail et al. (2017) found a high prevalence of depression in patients with mild cognitive impairment. However, we observed only small effect sizes in our healthy group (p < .001) and clinical group (p < .30) for the correlation between MDST and MoCA. One reason for these conflicting results could be our very heterogeneous sample – patients present to our clinic with the entire spectrum of neurological diseases. Another reason could lie in the use of the MoCA as a screening instrument for cognitive functioning and, therefore, insufficient assessment of all cognitive domains as in the case of elaborated neuropsychological testing. Mohn and Rund (2016) investigated the neurocognitive profile of major depression and found that patients with depression showed significantly lower scores in all cognitive domains. Additionally, depression severity was significantly associated with cognitive functioning. It should also be noted that there was impairment in all cognitive domains, with speed of processing and reasoning and problem-solving being most affected. These domains are not adequately assessed by the MoCA, which measures visuospatial, executive functions, naming, verbal short- and long-term memory, attention, language, and abstraction (Nasreddine et al., 2005). However, cognitive deficits in depression can express themselves differently, and the comparison with a global score can lead to bias.
Taken together, the MDST has been shown to have good psychometric properties, implementation, evaluation, and interpretation criteria, as instructions are standardized and evaluation and interpretation schemes have been defined. It has good construct and divergent validity and reliability in terms of Cronbach’s α and test-retest reliability.
In contrast to the German sample, where the MDST cut-off score for depression was < 11 (Christen et al., 2019), our ROC analyses revealed a cut-off score of < 10 for the clinical group (AUC .77) and < 11 for the healthy group, with an AUC value of .78, which indicates a good separation between subjects with and without depression, with GDS results as an external criterion. As mentioned previously, Turkish migrants are affected more by psychiatric diseases (Aichberger et al., 2010; Igde et al., 2019), and migrants, in general, are affected more by depression (Arias-Uriona & Guillén, 2020). Turkish healthy subjects are classified as possibly depressed when scoring 11, and Turkish patients are classified as possibly depressed when scoring 10 points in the MDST.
Limitations and Strengths
Some limitations of this study should be considered. The department in which our study was conducted was not a psychiatric but a neurological clinic; therefore, a detailed psychiatric anamnesis was not carried out. Additional data from psychiatric clinics and patients should be obtained for improved generalisability. Our cut-off values were based purely on data from the GDS as another screening instrument, so they should be re-validated in a study in which an elaborate diagnosis is provided. Because our patient population was heterogeneous regarding underlying pathology, further analysis should focus on specific neurological diseases (such as dementia and Parkinson’s disease) for more specific cut-off scores. In contrast, our study showed applicability for heterogeneous neurological diseases. Furthermore, health-based information in the healthy group was assessed by self-reports. As this could lead to a bias, we cannot be certain that this subgroup did not include subjects with additional health problems.
The GDS is not the most appropriate choice as a gold standard for depression because of the disadvantages of self-report screening scales. The cognitive decline of our subjects could have led to their misunderstanding the verbal questions, so that the results of the GDS may not be valid. We did not exclude patients with severe cognitive decline because the MDST was specifically developed for the clinical patient population and is intended to be used in this context. During development, we also wanted to consider patients who suffer from a severe cognitive disorder.
The main strength of our study is that it provides data for a nonverbal depression screening tool which has the potential to fill an important gap in the clinical diagnosis of (Turkish) patients with a migration background and language barriers. Another strength is our large sample size, in both the healthy group and clinical sample, although recruiting Turkish participants was challenging. High compliance could be achieved by recruiting native Turkish speakers and conducting tests in their native language.
Conclusion
This study shows the MDST to have good psychometric properties for detecting affective symptoms in neurological patients and healthy subjects with a Turkish migration background. Therefore, we believe it is a suitable alternative to linguistic and written-based screenings in this context. Because of its low dependence on language skills, the MDST has the potential to circumvent language barriers, making it an unbiased screening tool for affective status in persons with limited German proficiency. Therefore, the MDST has the potential to fill an important gap in the diagnosis of depression in neurological patients with a migration background status.
The authors wish to thank all participants of the study.
References
(2010). Depression in middle-aged and older first generation migrants in Europe: Results from the Survey of Health, Ageing and Retirement in Europe (SHARE). European Psychiatry, 25(8), 468–475. https://doi.org/10.1016/j.eurpsy.2009.11.009
. (2013). Diagnostic and statistical manual of mental disorders: DSM-5 (5th ed.). APA.
(2021). Effects of sociodemographic variables and depressive symptoms on MoCA test performance in native Germans and Turkish migrants in Germany. International Journal of Environmental Research and Public Health, 18(12). https://doi.org/10.3390/ijerph18126335
(2020). Inequalities in depressive symptoms between natives and immigrants in Europe: the mediating role of social exclusion. Cadernos De Saude Publica, 36(11), Article
e00124319 . https://doi.org/10.1590/0102-311X00124319(2021). “The system has to be health literate, too”: Perspectives among healthcare professionals on health literacy in transcultural treatment settings. BMC Health Services Research, 21(1), Article
716 . https://doi.org/10.1186/s12913-021-06614-x(1996). BDI-II, Beck depression inventory: Manual (2nd ed.). Psychological Corp./Harcourt Brace.
(2016). Health literacy among different age groups in Germany: Results of a cross-sectional survey. BMC Public Health, 16(1), Article
1151 . https://doi.org/10.1186/s12889-016-3810-6(2017). Prevalence of depression and anxiety in multiple sclerosis: A systematic review and meta-analysis. Journal of the Neurological Sciences, 372, 331–341. https://doi.org/10.1016/j.jns.2016.11.067
(2004). The nature and determinants of neuropsychological functioning in late-life depression. Archives of General Psychiatry, 61(6), 587–595. https://doi.org/10.1001/archpsyc.61.6.587
(2019). Mini-Depressions-Status-Test (MDST): Entwicklung eines grafisch basierten Screenings depressiver Verstimmungen.
[Mini-Depression-Status-Test (MDST): The development of a screening tool for depression] Zeitschrift für Neuropsychologie, 30(1), 9–17. https://doi.org/10.1024/1016-264X/a000242(1992). A power primer. Psychological Bulletin, 112(1), 155–159. https://doi.org/10.1037/0033-2909.112.1.155
(2017). Lifetime prevalence of mental disorders among first and second generation individuals with Turkish migration backgrounds in Germany. BMC Psychiatry, 17(1). https://doi.org/10.1186/s12888-017-1333-z
(2018). Prevalence of depression among migrants: A systematic review and meta-analysis. International Journal of Environmental Research and Public Health, 15(9). https://doi.org/10.3390/ijerph15091986
(2009). Development and validation of the Rasch-based Depression Screening (DESC) using Rasch analysis and structural equation modelling. Journal of Behavior Therapy and Experimental Psychiatry, 40(3), 468–478. https://doi.org/10.1016/j.jbtep.2009.06.003
(2016). Gesundheitskompetenz bei Personen mit Migrationshintergrund aus der Türkei und Ex-Jugoslawien in Österreich. Ergebnisse einer quantitativen und qualitativen Studie (BIHPR Forschungsbericht)
[Health literacy among persons with migration background from Turkey and former Yugoslavia in Austria. Results of a quantitative and qualitative study (BIHPR Research Report)] .(1999). Validität und Reliabilität einer deutschen Version der Geriatrischen Depressionsskala (GDS).
[Validity and reliability of a German version of the Geriatric Depression Scale (GDS)] Zeitschrift für Klinische Psychologie und Psychotherapie, 28(1), 18–27. https://doi.org/10.1026//0084-5345.28.1.18(2019). Depressive und somatoforme Störungen bei türkeistämmigen Personen in Deutschland.
[Depressive and somatoform disorders in persons with a Turkish migration background in Germany] Der Nervenarzt, 90(1), 25–34. https://doi.org/10.1007/s00115-018-0602-5(2017). Prevalence of depression in patients with mild cognitive impairment: A systematic review and meta-analysis. JAMA Psychiatry, 74(1), 58–67. https://doi.org/10.1001/jamapsychiatry.2016.3162
(2014). Pseudo-dementia: A neuropsychological review. Annals of Indian Academy of Neurology, 17(2), 147–154. https://doi.org/10.4103/0972-2327.132613
(2018). Inequalities in health care utilization among migrants and non-migrants in Germany: A systematic review. International Journal for Equity in Health, 17(1), Article
160 . https://doi.org/10.1186/s12939-018-0876-z(2017). Differences in depression stigma towards ethnic and socioeconomic groups in Germany: Exploring the hypothesis of double stigma. Journal of Affective Disorders, 208, 82–86. https://doi.org/10.1016/j.jad.2016.08.071
(2018). Comparison of treatment rates of depression after stroke versus myocardial infarction: A systematic review and meta-analysis of observational data. Psychosomatic Medicine, 80(8), 754–763. https://doi.org/10.1097/PSY.0000000000000632
(1995). The structure of negative emotional states: Comparison of the Depression Anxiety Stress Scales (DASS) with the Beck Depression and Anxiety Inventories. Behaviour Research and Therapy, 33(3), 335–343. https://doi.org/10.1016/0005-7967(94)00075-u
(2016). V Neurocognitive profile in major depressive disorders: Relationship to symptom level and subjective memory complaints. BMC Psychiatry, 16, Article
108 . https://doi.org/10.1186/s12888-016-0815-8(2014). Zusammenhang von wahrgenommener Diskriminierung mit Depressivität und gesundheitsbezogener Lebensqualität bei türkisch- und polnischstämmigen Migranten.
[The interrelation between perceived discrimination, depressiveness, and health-related quality of life in immigrants of Turkish and Polish origin] Psychiatrische Praxis, 41(4), 200–207. https://doi.org/10.1055/s-0033-1343221(2005). The Montreal Cognitive Assessment, MoCA: A brief screening tool for mild cognitive impairment. Journal of the American Geriatrics Society, 53(4), 695–699. https://doi.org/10.1111/j.1532-5415.2005.53221.x
(2008). A systematic review of prevalence studies of depression in Parkinson’s disease. Movement Disorders: Official Journal of the Movement Disorder Society, 23(2), 183–189, quiz 313. https://doi.org/10.1002/mds.21803
(2014). Cognitive impairment in depression: A systematic review and meta-analysis. Psychological Medicine, 44(10), 2029–2040. https://doi.org/10.1017/S0033291713002535
(2021). Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 because of the COVID-19 pandemic. The Lancet, 398(10312), 1700–1712. https://doi.org/10.1016/S0140-6736(21)02143-7
(2019). Income and education predict elevated depressive symptoms in the general population: Results from the Gutenberg health study. BMC Public Health, 19(1), Article
430 . https://doi.org/10.1186/s12889-019-6730-4(2010). Power of discrimination of Montreal Cognitive Assessment (MOCA) scale in Turkish patients with mild cognitive impairment and Alzheimer’s disease. Türk Geriatri Dergisi, 13(3), 166–171.
(2013). The diagnosis of depression: Current and emerging methods. Comprehensive Psychiatry, 54(1), 1–6. https://doi.org/10.1016/j.comppsych.2012.06.006
. (2012). Health literacy and public health: A systematic review and integration of definitions and models. BMC Public Health, 12, Article
80 . https://doi.org/10.1186/1471-2458-12-80(2018). Older migrants in Germany. Population Ageing, 11, 285–306. https://doi.org/10.1007/s12062-017-9183-5
. (2017). Poststroke depression: A scientific statement for healthcare professionals from the American Heart Association/American Stroke Association. Stroke, 48(2), e30–e43. https://doi.org/10.1161/STR.0000000000000113
(2014). The relationship between acculturation strategies and depressive and anxiety disorders in Turkish migrants in the Netherlands. BMC Psychiatry, 14(1). https://doi.org/10.1186/s12888-014-0252-5
(2017). Global, regional, and national incidence, prevalence, and years lived with disability for 328 diseases and injuries for 195 countries, 1990–2016: A systematic analysis for the Global Burden of Disease Study 2016. The Lancet, 390(10100), 1211–1259. https://doi.org/10.1016/S0140-6736(17)32154-2
. (2017). Depression and other common mental disorders (Global Health Estimates, p. 24). https://apps.who.int/iris/bitstream/handle/10665/254610/WHO-MSD-MER-2017.2-eng.pdf?sequence=1
. (2019). International statistical classification of diseases and related health problems (11th ed.). WHO https://icd.who.int/
(1983). Development and validation of a Geriatric Depression Screening Scale: A preliminary report. Journal of Psychiatric Research, 17(1), 37–49.
(2016). The prevalence of neuropsychiatric symptoms in Alzheimer’s disease: Systematic review and meta-analysis. Journal of Affective Disorders, 190, 264–271. https://doi.org/10.1016/j.jad.2015.09.069
(1983). The hospital anxiety and depression scale. Acta Psychiatrica Scandinavica, 67(6), 361–370. https://doi.org/10.1111/j.1600-0447.1983.tb09716.x
