https://orcid.org/0000-0001-6918-5689 
Figures
Abstract
Background
The Hamilton Depression Rating Scale (HDRS or HAMD) is widely used scale for depression assessment. A shortened version of the HDRS, composed of 7 items, was implemented. The latter is timesaving compared to the original version, while still providing similar precision. Our objective in this study was to assess the psychometric properties of the Arabic HAMD-7 scale among non-clinical and clinical samples of Lebanese adults.
Methods
In study 1, 443 Lebanese citizens enrolled in this cross-sectional study (June-September 2021). The total sample in study 1 was divided into two subsamples to conduct the exploratory-to-confirmatory factor analysis (EFA-to-CFA). Another cross-sectional study was conducted in September 2022 on another sample of Lebanese patients (independent from the sample of study 1) and included 150 patients attending two psychology clinics. The Montgomery–Asberg Depression Rating Scale (MADRS), Lebanese Depression Scale (LDS), Hamilton Anxiety Scale (HAM-A) and Lebanese Anxiety Scale (LAS) were used to assess the validity of the HAMD-7 scale.
Results
The results of the EFA (subsample 1; study 1) showed that the HAM-D-7 items converged into a one-factor solution (McDonald’s ω = .78). The CFA (subsample 2; study 1) confirmed the one-factor solution obtained in the EFA (ω = .79). CFA indicated that fit of the one-factor model of the HAM-D-7 was acceptable: χ2/df = 27.88/14 = 1.99, RMSEA = .066 (90% CI = .028, .102), SRMR = .043, CFI = .960, TLI = .939. All indices suggested that configural, metric, and scalar invariance was supported across gender. The HAMD-7 scale score positively correlated with the MADRS (r = 0.809; p<0.001), LDS (r = 0.872; p<0.001), HAM-A (r = 0.645; p<0.001) and LAS (r = 0.651; p<0.001) scales scores. The optimal cutoff score between healthy individuals and depressive patients for the HAMD-7 was at a value = 5.50 (Se = 82.8% and Sp = 62.4%). The positive and negative predicted values for the HAMD-7 were 25.1% and 96.0%, respectively. The positive and negative likelihood ratios were 2.20 and 0.28 respectively. No significant difference was found between the non-clinical total sample (study 1) and the clinical sample (study 2) in terms of HAM-D-7 scores (5.24 ± 4.43 vs 4.54 ± 5.06; t(589) = 1.609; p = .108).
Conclusion
Psychometric properties of the Arabic HAMD-7 scale are satisfactory, warranting its use clinically and in research. This scale seems highly efficient for ruling out depression; however, those with positive scores need a referral to a mental health professional for deeper evaluation. HAMD-7 might be self-administered by non-clinical subjects. Future studies are recommended to additionally confirm our results.
Citation: Obeid S, Azzi V, Hallit S (2023) Validation and psychometric properties of the Arabic version of Hamilton Depression Rating Scale 7 items (HAMD-7) among non-clinical and clinical samples of Lebanese adults. PLoS ONE 18(5): e0285665. https://doi.org/10.1371/journal.pone.0285665
Editor: Boshra Ismael Ahmed Arnout, King Khalid University, EGYPT
Received: November 4, 2022; Accepted: April 27, 2023; Published: May 18, 2023
Copyright: © 2023 Obeid et al. This is an open access article distributed under the terms of the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original author and source are credited.
Data Availability: All data generated or analyzed during this study are not publicly available due the restrictions from the ethics committee (data are owned by a third-party organization). The dataset supporting the conclusions is available upon request to member of the ethics committee at the Psychiatric Hospital of the Cross, Me Rana Nader (rnader@naderlawoffice.com).
Funding: The author(s) received no specific funding for this work.
Competing interests: The authors have declared that no competing interests exist.
Background
Depression is an emotional disorder causing dysfunction in the capacities and functions of daily life. While considered a public health concern, depression is classified among the disorders that ponder most heavily on individuals, families, and society [1]. Depression, as a mental disorder, portrays the presence of anhedonia (loss of pleasure), physical fatigue with reduced energy (anergy), a feeling of culpability, lack of self-assurance, disrupted sleep and appetite, with cognitive disorders, in particular, an attenuation of concentration and memory problems [2]. Likewise, depression often begins in early life, recurring at different stages, hence correlated with various negative consequences, encompassing higher morbidity and mortality rates and reduced quality of life [3]. Consequently, depression is the foremost source of incapacity globally regarding the number of disoriented years attributable to infirmity [4].
Previously conducted studies revealed a prevalence of depression of 4.4% before the COVID-19 pandemic [5], while this number increased by 27.6% due to that pandemic [6]. In Lebanon, a recent study showed even higher rates (59.7%) [7]. Across all psychiatric disorders, depression and anxiety usually co-occur [8]. A worldwide study stated that 45.7% of subjects with lifetime major depressive disorder had at least one anxiety disorder episode [9]; in fact, one or more anxiety disorders were present in 41.6% of people with 12-month severe depression.
The assessment of depressive symptoms depends on the use of standardized tools, given the lack of clinically relevant biomarkers for depression. Self-report scales, a rapid and reliable choice in the initial assessment of depression, are widely applied clinically and in the research field. These scales include but are not limited to the Beck Depression Inventory (BDI) [10], Montgomery-Asberg Depression Rating Scale (MADRS) [11], Lebanese Depression Scale (LDS-19) [12], Patient Health Questionnaire-9 (PHQ-9) [13], Center for Epidemiological Studies Depression Scale (CES-D) [14] and the Zung Self-Report Depression Scale [15].
The Hamilton Depression Rating Scale (HDRS or HAMD), published in 1960 [16], is the furthermost commonly used scale to assess depression [17]. The original scale comprises 21 items, with the last four (diurnal variation, depersonalization/derealization, paranoid symptoms, and obsessional and compulsive symptoms) deliver additional clinical data while not affecting the final score. The other items are graded on a range of 3–5 points, and the overall score determines how severe the patient’s depression is. Since its publication, it has evolved into the benchmark for diagnosing depression [18], however, the HAMD is not a diagnostic instrument [19]. The HAMD-17 has been validated in different languages, including but not limited to French [20], Turkish [21], Arabic [22] and Chinese [23]. The French version consists of three factors, the Arabic one four factors, the Chinese version five factors and the Turkish one six factors. Methodological limitations, cultural context and clinical groups studied [24] might explain the differences in the results at the factor analysis level. Approaches within the item response theory framework, which are especially pertinent to the field of transcultural psychiatry, allow for a thorough study of cultural variations in the responses to individual items [25].
A shortened version of the HDRS composed of 7 items was implemented in 2003 to assess the depression gravity according to the fourth version of the Diagnostic and Statistical Manual of Mental Disorders criteria [26]. An extraction technique was used to identify the most commonly endorsed and sensitive depressive items from the HAMD-17, a standardized clinician-rated depression scale [26]. It is interesting to note that this shortened version aimed to identify a cutoff score and evaluate the effectiveness of the current treatment by differentiating between symptomatic remission and full clinical response to treatment. Results of the latter study showed that a score of ≤3 indicate a complete remission of symptoms with the HDRS [26]. Noting that the HAMD-7 scale provides the same level of accuracy while being timesaving than the original version [27].
A single published review regarding the validity of the HAMD scale [18] criticized its factorial and content validity, while its convergent, discriminant, and predictive validity are generally satisfying. This was supported by previous authors [28] who showed a high correlation between the HAMD-17 and BDI scores (r = 0.73). The only subscale that has been repeatedly evaluated and recommended as a formal measure in itself is the 6-item melancholia subscale [29]. The latter reported the validity of the HAMD-6 for the assessment of the central symptoms of depression [30]. In 1993, Gibbons et al. [31] recommended a more condensed, one-dimensional tool composed of 8 items (HAMD-8) for the assessment of depression severity.
Only a few cross-cultural studies are included in Bagby and colleagues’ review [18]. Therefore, it is unclear whether the review’s conclusions apply to non-Western populations. In this context, a systematic review conducted in 2019 revealed that a number of depression items remain common in different cultures, however, other items differ between cultures as seen by the variability of Cronbach’s alpha values and number of factors retained [32].
The HAMD-7 scale was previously validated in Arabic in Saudi Arabia [27] and demonstrated to have good test-retest reliability [27]; however, there is a need to implement research among other diagnosed Lebanese patients with depression in view of assessing its adequacy in evaluating the intensity of depressive symptoms, hence the importance of our study. Furthermore, Lebanon differs from Saudi Arabia on cultural, economic, and political levels. For example, in Lebanon the economy is smaller than the average Arab country, do not derive a significant share of his revenues from natural resources, has a significantly more than average quality of democracy/participatory political system in opposite to Saudi Arabia classified as "politically restrictive with high gender inequality" [33]. Therefore, our main objective in this study was to validate the HAMD-7 scale among non-clinical and clinical samples of Lebanese adults, and compare its psychometric properties with two accepted measurement tools, MADRS and LDS-19. We hypothesize that the items of HAMD-7 would load on a one-factor solution.
Methods
Ethics approval and consent to participate
The Psychiatric Hospital of the Cross Ethics and Research Committee agreed on this study protocol (HPC-023-2021). Submitting the form online was considered equivalent to obtaining a written informed consent.
Study 1 (non-clinical/community sample)
General design.
This cross-sectional study was performed between June and September 2021, during which 443 Lebanese citizens were recruited from the community from all governorates via a chain sampling snowball technique and filled an online survey. The research team asked acquaintances to forward the link along to their friends and family members. The link was distributed to participants via social media applications (WhatsApp, Instagram, Messenger, etc.). Everyone above 18 years old was qualified for the study. Mundfrom et al. suggest a minimum sample based on 3–20 times the number of items of the scale [34]. Therefore, since the short form of the HAM-D scale is composed of 7 items, two sample of 140 participants each were needed at least to perform the exploratory and confirmatory factor analyses respectively.
Participants characteristics (sample 1)
The mean age of the total sample was 31.71 ± 13.90 years, with 57.1% females; 193 (43.6%) had depressive symptoms (scores ≥4). A significantly higher percentage of participants who had a physician-diagnosis with depression had a secondary level of education or less, whereas a higher mean age was significantly found in patients who had a physician-diagnosis with depression (Table 1).
Study 2 (clinical sample)
We conducted another cross-sectional study in September 2022 on another sample of Lebanese patients (independent from the sample enrolled in study 1); 150 patients, attending two psychology clinics, enrolled in study 2 (mean age = 33.21 ± 14.26 years; 52% females; 29.3% married and 64.7% university level of education). The HAM-D-7 scale’s items were administered via a face-to-face interview with the clinical psychologist.
Measurement
The Arabic questionnaire consisted of questions about sociodemographic aspects (sex, age, marital status, educational level, and monthly income). It also asked participants to self-report a physician diagnosis of depression and anxiety (no/yes type of answer). It included the following scales:
Hamilton Depression Rating Scale (HAMD-7).
The HAMD was used to measure depression that occurred in the last 7 days [35]. It consists of 7 items (e.g. “depressed mood: Have you been feeling down or depressed this past week?—How often have you felt this way, and for how long?”), rated on a 5-point scale from 0 (not present) to 4 (severe). Higher scores indicate higher depression. Scores ≥4 indicate the presence of depressive symptoms [27].
Montgomery–Asberg Depression Rating Scale.
The MADRS is validated in Lebanon [36]. It is composed of 10 items assessing the severity of depression exhibited over the preceding two weeks [11] (e.g. “Apparent sadness”, “Reported sadness”). Items are scored from 0 to 6 [37].
Lebanese Depression Scale.
The LDS‐19 scale evaluates depression over the last 2 weeks and is composed of 19 items (e.g. "Retardation (slowness of thought and speech, impaired ability to concentrate, decreased motor activity”)) [38]. Higher scores indicate higher depression.
Hamilton Anxiety Scale.
The Hamilton Anxiety Scale (HAM-A) is validated in Lebanon [39] and includes 14 items [40] rated from 0 (symptoms not present) to 4 (very extreme symptoms). The higher the scores, the more the anxiety (e.g. “Anxious mood: Worries, anticipation of the worst, fearful anticipation, irritability”).
Lebanese Anxiety Scale.
It contains 10 items and is validated among adults and adolescents [41,42] for the assessment of anxiety symptoms experienced during the past week (e.g. “I have insomnia (Difficulty in falling asleep, broken sleep, unsatisfying sleep and fatigue on waking, dreams, nightmares, night terrors)”. Higher scores indicate higher levels of anxiety.
Statistical analysis
SPSS software version 23 was used to perform the statistical analysis. Missing values represented less than 10% of the total and were replaced with the mean value [43]. The HAMD-7 score’s normal distribution was validated by its skewness (= 0.722) and kurtosis (= .224) values varying between ±1 [44]. The Chi-square test was used to compare two percentages. Student t test was used to compare two means. P <0.05 was envisaged as significant.
We employed an exploratory-to-confirmatory (EFA-to-CFA) factor analysis method to investigate the HAMD-7 factor structure [45]. We divided Sample 1 using the SPSS-generated random approach to provide suitable sample sizes for both EFA and CFA. There were no significant differences between the two subsamples in terms of mean age, t(824) = 1.52, p = .129, d = .11, and BMI, t(824) = 0.07, p = .944, d = .004, as well as the distribution of women and men, χ2(1) = .015, p = .901, of single and married participants, χ2(1) = .083, p = .773, education level χ2(1) = 1.603, p = .205, and monthly income, χ2(3) = 3.936, p = .268.
First, we computed a robust EFA using the principal analysis component with the first split-half subsample using the FACTOR software [46]. We confirmed that all conditions related to item-communality [47], average item correlations, and item-total correlations [48], Weighted Root Mean Square Residual (WRMR) [49], Kaiser-Meyer-Olkin (KMO) measure and Bartlett’s test of sphericity [50], were met. The dimensionality of the instrument was determined through the optimal implementation of Parallel Analysis [51], using the Pearson correlation matrix. Items retained when having a factor loading >0.4 and communality >0.3 [52].
CFA was performed using SPSS AMOS v.29 using the maximum likelihood estimation methods based on the factor solution obtained in the EFA. Multivariable normality was verified since the Bollen-Stine bootstrap p-value was 0.098 (>0.05). The root mean square error of approximation (RMSEA), Tucker Lewis Index (TLI), and Comparative Fit Index (CFI) assessed the goodness-of-fit of the model [53]. Values of RMSEA ≤0.08, and CFI and TLI ≥0.90 stipulate a good-fitting model [53]. The average variance extracted (AVE) values of ≥ .50 ensured convergent validity [54].
Gender invariance. We applied multi-group CFA on the complete sample from study 1 to explore the gender invariance of HAMD-7 scores [55]. Measurement invariance was assessed at the configural, metric, and scalar levels [56], evidenced if ΔCFI ≤ .010 and ΔRMSEA ≤ .015 or ΔSRMR ≤ .010 [55,57].
McDonald’s ω was chosen to estimate reliability due to known issues with the application of Cronbach’s α [58] (values >.70 considered adequate [59]).
The cutoff point of the scale versus physician-diagnosed depression was determined using the Receiver-Operating Characteristics (ROC) curve; sensitivity, specificity, and negative and positive predictive values were calculated for the HAMD-7 scale; for this purpose, we relied on the participant’s self-report of physician diagnosis of depression (yes/no). In addition, positive likelihood ratio was calculated according to the following formula: , whereas negative likelihood ratio was calculated according to:
.
Finally, to evaluate convergent validity, we tested the correlations between HAMD-7 and MADRS, LDS, HAM-A and LAS scales since depression and anxiety usually coexist [60].
Results
McDonald’s omega values in the total sample (study 1) were as follows: MADRS (ω = 0.85), LDS (ω = 0.87), HAM-A (ω = 0.95) and LAS (ω = 0.94). The mean HAM-D score in sample 1 was 5.24 ± 4.43 and 4.54 ± 5.06 in sample 2.
Principal Component Analysis (subsample 1; study 1)
In both men and women, the advised number of dimensions according to the parallel analysis was one. The common variance explained was 49.15% and 40.54% respectively. The WRMR values were also adequate (.131; 95% CI .114-.142 and .120; 95% CI .106-.130) respectively. The factor loadings are shown in Table 2. McDonald’s ω was satisfactory in this subsample (ω = .78), in men (ω = .82) and in women (ω = .76).
Confirmatory Factor Analysis (subsample 2; study 1)
Fit indices for the CFA of the one-dimensional model of the HAM-D-7 was acceptable: χ2/df = 27.88/14 = 1.99, RMSEA = .066 (90% CI = .028, .102), SRMR = .043, CFI = .960, TLI = .939. The standardised estimates of factor loadings (Table 2) and AVE = .55 were acceptable. McDonald’s omega was good in the second subsample (ω = .79).
Gender invariance
All indices suggested that configural, metric, and scalar invariance was supported across gender (Table 3). The Student t test results showed that a higher mean HAMD-7 score was seen in women (M = 5.88, SD = 4.39) compared to men (M = 4.52, SD = 4.55) in the second subsample from study 1, t(211) = 2.201, p = .029, d = .304.
Convergent validity (total sample)
The HAMD-7 scale score positively correlated with the MADRS (r = 0.809; p<0.001), LDS (r = 0.872; p<0.001), HAM-A (r = 0.645; p<0.001) and LAS (r = 0.651; p<0.001) scales scores.
Validity measures (ROC curve) (sample 1)
The performance of the HAM-D-7 scale scores in terms of classifying participants by their depression status (yes/no; self-report physician diagnosis of depression) was evaluated with ROC curve analysis The HAMD-7 scale had a smaller area under the curve compared to the LDS and MADRS scales. The HAMD-7’s ideal cutoff score between healthy people and depressive patients was at a value of 5.50 (Se = 82.8% and Sp = 62.4%) (Fig 1). This indicates that this cutoff value, the HAMD-7 scale is capable of capturing persons with depressive symptoms by 82.8%, whereas it is capable to capturing persons who are not depressed by 62.4%.
Patients with a self-report physician’s diagnosis of depression were analyzed (sample 1). Area under the curve of LDS-19 scale = 0.833 [0.781–0.885] (p<0.001); at value = 16.50, Se = 82.8% and Sp = 72.1%. Area under the curve of HAMD-7 scale = 0.788 [0.731–0.844] (p<0.001); at value = 5.50, Se = 82.8% and Sp = 62.4%. Area under the curve of MADRS scale = 0.827 [0.773–0.882] (p<0.001); at value = 9.50, Se = 81.0% and Sp = 73.4%.
Predictive values (total sample)
The positive and negative predicted values for the HAMD-7 were 25.1% and 96.0%, respectively.
Discussion
Factorial validity supporting the one-factor model
We examined the psychometric properties of the HAMD-7 among non-clinical and clinical samples of Lebanese adults. Researchers are constantly criticized for the administration of long surveys, with other scales created to screen for the presence of depressive symptoms (such as the HAMD-17 and the LDS) being lengthy and time-consuming. Therefore, we found a necessity to evaluate the psychometric properties of the HAMD-7 scale. Our results confirm the psychometric properties of the one-factor solution. Furthermore, this scale can be completed in a couple of minutes.
Internal consistency and convergent validity
The HAM-D 7 items appeared to be internally consistent, reaching a value of .78 and .79 respectively, higher than the value obtained in the previous Arabic validation in Saudi Arabia (Cronbach’s alpha values < .75 in the pre-and post-tests). Concerning the convergent validity of the proposed factorial solution, the HAMD-7 total score correlated exceptionally well with other validated depressive symptoms’ screening scales (LDS and MADRS). It also highly correlated with anxiety scales (HAM-A and LAS), proving the divergent validity of the scale. Summing up, the current study’s findings support the construct validity of the Arabic HAMD-7. In addition, the non-significant difference between the clinical and non-clinical samples suggest that there is no difference in the administration of the scale between a healthcare professional and the participant him/herself (self-adminsitered). Participants may now have more privacy when answering the questions, be at ease and more willing to provide honest answers.
ROC curve
The area under the curve of the HAMD-7 was lower than that of the LDS and MADRS scales. This was kind of expected since the latter scales include items that cannot be found in the HAMD-7. Despite the aforementioned benefits of the HAMD-7, there are some drawbacks to this brief version. For example, insomnia, disturbed appetite, loss of libido, and hypochondriasis are only a few of the HAMD-17 items that the HAMD-7 does not include, so it may not detect some symptoms that the HAMD-17 may detect [61]. However, a high correlation with the MADRS and Clinical Global Impression Scale, as well as acceptable levels of sensitivity and specificity, indicate that the HAMD-7’s briefness does not seem to to jeopardize crucial details on patient progress and outcome [35].
The scale has high sensitivity and moderate specificity, indicating that participants scoring less than 5.5 can be considered not suffering from depression. Furthermore, the NPV value was excellent (96%), which means that the HAMD-7 scale can be used to rule out depression, whereas the PPV was low (25.1%), reflecting that this scale cannot be used for the diagnosis/screening of depressive symptoms. Those findings are similar to those obtained with the MADRS [62] and LDS [12] scales. Hence, a practitioner’s diagnosis is crucial to the screening test’s validation.
Clinical implications
Results of our study suggest that the HAMD-7, Arabic version, may be a possibly valuable clinical tool for clinicians and researchers aiming to assess depression severity among Arab subjects. Such research from an understudied cultural setting, on a subject that is culturally dependent, could support the instrument’s external validity [63]. Furthermore, we recommend clinical psychologists and educators to use the shortened version of the HAMD-7 to conduct a thorough assessment into the presence of depression among subjects who are more likely to experience depressive symptoms and to take a holistic approach to manage their condition. Additionally, young adults and those who have had significant life events, chronic medical conditions, or mental health disorders should regularly be screened for depression with the necessary management following.
Limitations
These results cannot be attributed to the entire population since the sample had a relatively low mean age, with the majority having a university education level. In addition, patients tend to exaggerate the severity of their symptoms. The snowball technique followed in the data collection, the inability to know the refusal rate/response rate implicate a selection bias because of the respondent-driven sampling. We failed to test the criteria validity that needs clinical diagnosis according to the standard clinical test (physician’s diagnosis). Additionally, relying only on a self-report of being diagnosed with depression (yes or no) might not be a reliable criterion to differentiate clinical and non-clinical populations to establish a cut-off. Other psychometric properties of the scale (test-retest) are recommended to strengthen our results. Divergent validity was tested using scales that screen for anxiety; literature has discussed whether anxiety and depression are two separate constructs or not, therefore, future studies should assess the divergent validity of the HAM-D 7 items using other scales.
Conclusion
The primary psychometric properties obtained for the Arabic HAMD-7 are satisfactory, warranting this short-scale use clinically and in research. This scale seems highly efficient for ruling out depression; however, those with positive scores need a referral to a healthcare practitioner for deeper evaluation. HAMD-7 might be self-administered by non-clinical subjects. Future studies are recommended to additionally confirm our results.
References
- 1. Liu Q, He H, Yang J, Feng X, Zhao F, Lyu J. Changes in the global burden of depression from 1990 to 2017: Findings from the Global Burden of Disease study. Journal of psychiatric research. 2020;126:134–40. pmid:31439359
- 2. Marcus M, Yasamy MT, van Ommeren Mv, Chisholm D, Saxena S. Depression: A global public health concern. 2012.
- 3. Machado MO, Veronese N, Sanches M, Stubbs B, Koyanagi A, Thompson T, et al. The association of depression and all-cause and cause-specific mortality: an umbrella review of systematic reviews and meta-analyses. BMC Med. 2018;16(1):112. pmid:30025524; PubMed Central PMCID: PMC6053830.
- 4. Ogbo FA, Mathsyaraja S, Koti RK, Perz J, Page A. The burden of depressive disorders in South Asia, 1990–2016: findings from the global burden of disease study. BMC Psychiatry. 2018;18(1):333. pmid:30326863; PubMed Central PMCID: PMC6192293.
- 5. Friedrich MJ. Depression Is the Leading Cause of Disability Around the World. JAMA. 2017;317(15):1517. pmid:28418490.
- 6. Santomauro DF, Herrera AMM, Shadid J, Zheng P, Ashbaugh C, Pigott DM, et al. Global prevalence and burden of depressive and anxiety disorders in 204 countries and territories in 2020 due to the COVID-19 pandemic. The Lancet. 2021. pmid:34634250
- 7. Obeid , Lahoud N, Haddad C, Sacre H, Akel M, Fares K, et al. Factors associated with depression among the Lebanese population: Results of a cross‐sectional study. Perspectives in psychiatric care. 2020;56(4):956–67. pmid:32314394
- 8. Kessler RC, McGonagle KA, Zhao S, Nelson CB, Hughes M, Eshleman S, et al. Lifetime and 12-month prevalence of DSM-III-R psychiatric disorders in the United States: results from the National Comorbidity Survey. Archives of general psychiatry. 1994;51(1):8–19.
- 9. Kessler RC, Sampson NA, Berglund P, Gruber M, Al-Hamzawi A, Andrade L, et al. Anxious and non-anxious major depressive disorder in the World Health Organization World Mental Health Surveys. Epidemiology and psychiatric sciences. 2015;24(3):210–26. pmid:25720357
- 10. Beck AT, Steer RA, Brown GK. Beck depression inventory. 1996.
- 11. Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. The British journal of psychiatry. 1979;134(4):382–9. pmid:444788
- 12. Obeid , Haddad C, Sacre H, Akel M, Chalhoub C, Rahme C, et al. The Lebanese Depression Scale: A scale to screen for depression in the Lebanese population. Perspect Psychiatr Care. 2021;57(2):620–6. pmid:32686132.
- 13. Spitzer RL, Kroenke K, Williams JB, Group PHQPCS, Group PHQPCS. Validation and utility of a self-report version of PRIME-MD: the PHQ primary care study. Jama. 1999;282(18):1737–44.
- 14. Radloff LS. A self-report depression scale for research in the general population. Applied psychol Measurements. 1977;1:385–401.
- 15. Zung WW. A self-rating depression scale. Archives of general psychiatry. 1965;12(1):63–70.
- 16. Hamilton M. The Hamilton Depression Scale—accelerator or break on antidepressant drug discovery. Psychiatry. 1960;23:56–62.
- 17. Carrozzino D, Patierno C, Fava GA, Guidi J. The Hamilton Rating Scales for Depression: A Critical Review of Clinimetric Properties of Different Versions. Psychother Psychosom. 2020;89(3):133–50. pmid:32289809.
- 18. Bagby RM, Ryder AG, Schuller DR, Marshall MB. The Hamilton Depression Rating Scale: has the gold standard become a lead weight? American Journal of Psychiatry. 2004;161(12):2163–77. pmid:15569884
- 19. Poirot I. Syndromes dépressifs et échelles d’évaluation (1 ère partie). Médecine du Sommeil. 2005;2(5):38–41.
- 20. Guelfi J. L’évaluation clinique standardisée en psychiatrie, tome I. Paris: Edition Pierre Fabre. 1997.
- 21. Akdemir A, Türkçapar M, Örsel S, Demirergi N, Dag I, Özbay M. Reliability and validity of the Turkish version of the Hamilton Depression Rating Scale. Comprehensive psychiatry. 2001;42(2):161–5. pmid:11244153
- 22. Obeid S, Hallit CAE, Haddad C, Hany Z, Hallit S. Validation of the Hamilton Depression Rating Scale (HDRS) and sociodemographic factors associated with Lebanese depressed patients. L’encephale. 2018;44(5):397–402. pmid:29307705
- 23. Zheng Y, Zhao J, Phillips M, Liu J, Cai M, Sun S, et al. Validity and reliability of the Chinese Hamilton depression rating scale. The British Journal of Psychiatry. 1988;152(5):660–4. pmid:3167442
- 24. Nixon N, Guo B, Garland A, Kaylor-Hughes C, Nixon E, Morriss R. The bi-factor structure of the 17-item Hamilton Depression Rating Scale in persistent major depression; dimensional measurement of outcome. PloS one. 2020;15(10):e0241370. pmid:33104761
- 25. DeMars CE. A Review of “The Theory and Practice of Item Response Theory” de Ayala, RJ (2009). New York: Guilford Press. Taylor & Francis; 2011.
- 26. McIntyre RS, Fulton KA, Bakish D, Jordan J, Kennedy S. The HAM-D~ 7: A Brief Depression Scale to Distinguish Antidepressant Response From Symptomatic Remission. Primary Psychiatry. 2003;10(1):39–42.
- 27. Alhadi Alarabi MA, Alshomrani AT Shuqdar RM, Alsuwaidan MT McIntyre RS. Arabic Translation, Validation and Cultural Adaptation of the 7-Item Hamilton Depression Rating Scale in Two Community Samples. Sultan Qaboos Univ Med J. 2018;18(2):e167–e72. pmid:30210846; PubMed Central PMCID: PMC6132530.
- 28. Rehm LP, O’Hara MW. Item characteristics of the Hamilton rating scale for depression. journal of Psychiatric Research. 1985;19(1):31–41.
- 29. Bech P, Allerup P, Gram L, Reisby N, Rosenberg R, Jacobsen O, et al. The Hamilton Depression Scale: evaluation of objectivity using logistic models. Acta Psychiatrica Scandinavica. 1981;63(3):290–9.
- 30. Bech P, Wilson P, Wessel T, Lunde M, Fava M. A validation analysis of two self‐reported HAM‐D6 versions. Acta Psychiatrica Scandinavica. 2009;119(4):298–303. pmid:19032701
- 31. Gibbons RD, Clark DC, Kupfer DJ. Exactly what does the Hamilton depression rating scale measure? Journal of psychiatric research. 1993;27(3):259–73. pmid:8295158
- 32. Vindbjerg E, Makransky G, Mortensen EL, Carlsson J. Cross-cultural psychometric properties of the Hamilton Depression Rating Scale. The Canadian Journal of Psychiatry. 2019;64(1):39–46. pmid:29719964
- 33. Kuncic A. How similar are Arab countries and what are their characteristics. the 15th annual international. 2016.
- 34. Mundfrom DJ, Shaw DG, Ke TL. Minimum sample size recommendations for conducting factor analyses. International journal of testing. 2005;5(2):159–68.
- 35. McIntyre RS, Konarski JZ, Mancini DA, Fulton KA, Parikh SV, Grigoriadis S, et al. Measuring the severity of depression and remission in primary care: validation of the HAMD-7 scale. Cmaj. 2005;173(11):1327–34. pmid:16301700
- 36. Hallit S, Obeid S, El Hage W, Kazour F. Validation of the Arabic version of the MADRS scale among Lebanese patients with depression. L’encephale. 2019;45(3):195–9. pmid:30122299
- 37. Müller-Thomsen T, Arlt S, Mann U, Maß R, Ganzer S. Detecting depression in Alzheimer’s disease: evaluation of four different scales. Archives of Clinical Neuropsychology. 2005;20(2):271–6. pmid:15708735
- 38. Obeid S, Haddad C, Sacre H, Akel M, Chalhoub C, Rahme C, et al. The Lebanese depression scale: a scale to screen for depression in the Lebanese population. Perspectives in Psychiatric Care. 2021;57(2):620–6. pmid:32686132
- 39. Hallit , Haddad C, Hallit R, Akel M, Obeid S, Haddad G, et al. Validation of the Hamilton anxiety rating scale and state trait anxiety inventory a and B in Arabic among the Lebanese population. Clinical epidemiology and global health. 2020;8(4):1104–9.
- 40. Thompson E. Hamilton Rating Scale for Anxiety (HAM-A). Occup Med (Lond). 2015;65(7):601. pmid:26370845.
- 41. Hallit S, Obeid S, Haddad C, Hallit R, Akel M, Haddad G, et al. Construction of the Lebanese Anxiety Scale (LAS-10): a new scale to assess anxiety in adult patients. Int J Psychiatry Clin Pract. 2020:1–8. pmid:32228282.
- 42. Merhy G, Azzi V, Salameh P, Obeid S, Hallit S. Anxiety among Lebanese adolescents: scale validation and correlates. BMC Pediatr. 2021;21(1):288. pmid:34158020; PubMed Central PMCID: PMC8218523.
- 43. Downey RG, King CV. Missing data in Likert ratings: A comparison of replacement methods. The Journal of general psychology. 1998;125(2):175–91. pmid:9935342
- 44. Hair JF Jr, Hult GTM, Ringle C, Sarstedt M. A primer on partial least squares structural equation modeling (PLS-SEM): Sage publications; 2017.
- 45. Swami V, Barron D. Translation and validation of body image instruments: Challenges, good practice guidelines, and reporting recommendations for test adaptation. Body image. 2019;31:204–20. pmid:30220631
- 46. Lorenzo-Seva U, Ferrando PJ. FACTOR: a computer program to fit the exploratory factor analysis model. Behav Res Methods. 2006;38(1):88–91. pmid:16817517.
- 47. Worthington RL, Whittaker TA. Scale development research: A content analysis and recommendations for best practices. The counseling psychologist. 2006;34(6):806–38.
- 48. Clark L, Watson D. Construct validity: basic issues in objective scale development. Psychological Measurement. 1995;28:61–75.
- 49. Yu CY, editor Evaluation of model fit indices for latent variable models with categorical and continuous outcomes. Paper presented at the annual conference of the American Educational Research Association, April 4, 2002, New Orleans; 2002.
- 50. Hair JF. Multivariate data analysis. 2009.
- 51. Timmerman ME, Lorenzo-Seva U. Dimensionality assessment of ordered polytomous items with parallel analysis. Psychological methods. 2011;16(2):209. pmid:21500916
- 52. Tabachnick B, Fidell L. Using Multivariate Statistics (7th ed.). New York: Pearson Publishers; 2019.
- 53. Byrne BM. Structural equation modeling with Mplus: Basic concepts, applications, and programming: routledge; 2013.
- 54. Malhotra N, Dash S. Marketing Research: An Applied Orientation (;Pearson, Ed.). Delhi; 2011.
- 55. Chen FF. Sensitivity of goodness of fit indexes to lack of measurement invariance. Structural equation modeling: a multidisciplinary journal. 2007;14(3):464–504.
- 56. Vadenberg R, Lance C. A review and synthesis of the measurement in variance literature: Suggestions, practices, and recommendations for organizational research. Organ Res Methods. 2000;3:4–70.
- 57. Cheung GW, Rensvold RB. Evaluating goodness-of-fit indexes for testing measurement invariance. Structural equation modeling. 2002;9(2):233–55.
- 58. McNeish D. Thanks coefficient alpha, we’ll take it from here. Psychological methods. 2018;23(3):412. pmid:28557467
- 59. Dunn TJ, Baguley T, Brunsden V. From alpha to omega: A practical solution to the pervasive problem of internal consistency estimation. British journal of psychology. 2014;105(3):399–412. pmid:24844115
- 60. Dagher M, Farchakh Y, Barbar S, Haddad C, Akel M, Hallit S, et al. Association between problematic social media use and memory performance in a sample of Lebanese adults: the mediating effect of anxiety, depression, stress and insomnia. Head Face Med. 2021;17(1):6. pmid:33622360; PubMed Central PMCID: PMC7901207.
- 61. Lin C-H, Park C, McIntyre RS. Early improvement in HAMD-17 and HAMD-7 scores predict response and remission in depressed patients treated with fluoxetine or electroconvulsive therapy. Journal of affective disorders. 2019;253:154–61. pmid:31035216
- 62. Thase ME, Harrington A, Calabrese J, Montgomery S, Niu X, Patel M. Evaluation of MADRS Severity Thresholds in Patients With Bipolar Depression. CNS Spectr. 2021;26(2):182–3. pmid:33677183.
- 63. Fekih-Romdhane F, Obeid S, Malaeb D, Hallit R, Hallit S. Validation of a shortened version of the Eating Attitude Test (EAT-7) in the Arabic language. Journal of eating disorders. 2022;10(1):1–8.