Reliability,Criterion and Concurrent Validity of the Farsi Translation of DIVA-5: A Semi-Structured Diagnostic Interview for Adults With ADHD

Abstract

Objectives: This study evaluated the psychometrics of the Farsi translation of diagnostic interview for attention-deficit hyperactivity disorder (ADHD) in adults (DIVA-5) based on DSM-5 criteria. Methods: Referrals to a psychiatric outpatient clinic (N = 120, 61.7% males, mean age 34.35 ± 9.84 years) presenting for an adult ADHD (AADHD) diagnosis, were evaluated using the structured clinical interviews for DSM-5 (SCID-5 & SCID-5-PD) and the DIVA-5. The participants completed Conner’s Adult ADHD Rating Scale-Self Report-Screening Version (CAARS-S-SV). Results: According to the SCID-5 and DIVA-5 diagnoses, 55% and 38% of the participants had ADHD, respectively. Diagnostic agreement was 81.66% between DIVA-5/SCID-5 diagnoses, 80% between SCID-5/CAARS-S-SV, and 71.66% between DIVA-5/CAARS-S-SV. Test–retest and inter-rater reliability results for the DIVA-5 were good to excellent. Conclusion: Findings support the validity and reliability of the Farsi translation of DIVA-5 among the Farsi-speaking adult outpatient population.

Keywords

adult ADHD DIVA-5 reliability validity

The current prevalence of attention-deficit hyperactivity disorder in adults (AADHD), as a serious and under-diagnosed worldwide condition has been reported as 2.8% (ranging 1.4–3.8% in low-income compared to high-income countries). Among the adults whose ADHD was clinically diagnosed in childhood, 57% still met diagnostic criteria of the disorder as adults (Fayyad et al., 2017).

Common characteristics of AADHD differ from the typical presentation of ADHD symptoms in children, where they are associated with functional impairments, as well as emotional dysregulation. The individuals with AADHD mostly suffer from being forgetful, having problems in focusing on one task especially for a long period, difficulty in organizing and prioritizing activities, initiating, continuing and ending their tasks, and problems with time management (Barkley et al., 2006, 2008). Their impulsive behavior often faces them with serious outcomes such as early termination of relationships or jobs (Weiss & Hechtman, 1993), more accidents, suicide attempts, and higher mortality rates (Barbaresi et al., 2013). Adults with ADHD are also frequently presented with other psychiatric disorders (Barkley & Brown, 2008) and show mood instability, irritability, and anger, and lose their motivation and tolerance for fatigue, although these emotion regulation problems are not specific to this disorder (Rapport et al., 2002).

Because of the negative consequences of AADHD and its high comorbidity with other psychiatric disorders (Asherson et al., 2012), it is essential to diagnose the affected individuals accurately to distinguish them from individuals with other common psychiatric disorders. However, most adults with ADHD are unfortunately left underdiagnosed and untreated (Aragonès et al., 2010). To assess and screen adults with ADHD, it is needed to identify the symptoms and behaviors which match with the DSM-5 criteria for ADHD, to evaluate the impairment and distress which can be attributed to these lifetime symptoms, and to check for other conditions such as drug and alcohol abuse, personality, sleep, mood, and anxiety disorders.

Among the scales and instruments which have been designed and used for assessment of AADHD, most are questionnaires and rating scales that provide quantitative data for the severity of symptoms and disorders (Haavik et al., 2010; Taylor et al., 2011). These tools are usually completed by the individuals themselves or other informants familiar with them. The Conner’s Adult ADHD Rating Scale-Self Report (CAARS-S) and the Wender Utah Rating Scale (short version) have been reported as the tools with the best psychometric properties among the 14 existing tools (Taylor et al., 2011). However, these tools are not diagnostic, and careful evaluation of AADHD needs more elaborated clinical assessment and structured interviews. The best approach to assess adult ADHD can be the use of a combination of structured or semi-structured diagnostic interviews and scoring scales (Dulcan, 1997).

As far as we know, there are three semi-structured interviews including the Conner’s Adult ADHD Diagnostic Interview for DSM-IV (CAADID), the ACE+ (https://www.psychology-services.uk.com/adhd) and the Diagnostic Interview for ADHD in adults (DIVA; www.divacenter.eu). The CAADID interview provides a thorough set of information including the symptoms, age of onset, and type of the disorder (Conners et al., 1999). The CAADID has been shown to have a fair to good kappa statistics between current and past symptomatology and good concurrent validity (Epstein & Kollins, 2006). Moreover, its Spanish version was reported as a reliable interview to diagnose AADHD in clinical and research settings (Ramos-Quiroga et al., 2012). The ACE+ is a new diagnostic interview translated into several languages, but as yet there are no validating data published.

The DIVA 2.0 was more recently developed instrument than the CAADID, being published online by DIVA Foundation in the Netherlands in August 2010. This interview was based on the DSM-IV criteria and the first Dutch-language structured interview to diagnose ADHD in adults. The new edition, DIVA-5 is based on the DSM-5 criteria, and has the advantage of being very affordable, found in many languages, and easily available through the DIVA Foundation website. It assesses the ADHD symptoms and criteria and their related functioning impairment based on the DSM-5 criteria.

Limited psychometric studies have been conducted on the previous edition, the DIVA-2.0, although it has been widely used to evaluate the diagnosis in recent published research studies of AADHD. Ramos-Quiroga et al. (2019) supported the concurrent validity of the DIVA 2.0 and CAADID. They found a 100% diagnostic agreement between the two interviews and the raters as well. These two interviews were totally in agreement with each other on adults with ADHD. They reported an appropriate association among the DIVA 2.0 with three self-report scales. Discriminative validity of the DIVA 2.0 has also been confirmed compared to a battery of neuropsychological tests and diagnostic assessment tools (Pettersson et al., 2018). The interview was the most accurate instrument with the highest balanced sensitivity and specificity results.

Studies on the prevalence of AADHD in Iran are scarce. Amiri et al. (2014) used the CAARS-S and reported the rate of 3.8% for adults with ADHD in city of Tabriz recruiting a sample of 400 urban inhabitants. Arabgol et al. (2004) used the CAARS-S in a group of students and found that 3.7% of them had scores higher than the cutoff-point. This rate was higher than 11% in another group of students studying at Zahedan university completing the CAARS-S-Screening Version (CAARS-S: SV; Mosalanejad et al., 2013). Hamzeloo et al. (2016) using the Adult ADHD Self-Report Scale Screener found that 16.2% of the Iranian male prisoners had AADHD. As this literature review shows, all these studies have used the self-reported rating scales to assess AADHD, while precise diagnosis of ADHD in adults needs to be assessed using clinical/semi-structured interviews in combination with other rating scales. This fact confirms the need to test psychometric properties of a semi-structured interview which can be used to diagnose AADHD in a Farsi speaking population.

As far as we know this is the first study that uses the updated version of the DIVA (DIVA-5) based on the DSM-5 criteria, and in a non-Dutch/non-English language. We aimed to investigate the validity and reliability of the Farsi version of the DIVA-5 to evaluate whether this interview has an appropriate diagnostic value in the diagnosis of ADHD in Farsi-speaking adults.

Methods

Participants

This study was designed to investigate AADHD diagnosis among all referrals to the different outpatient clinics of a university affiliated psychiatric hospital (Roozbeh) in Tehran, the capital of Iran. These clinics receive referrals with different psychiatric complaints. The individuals who thought they had symptoms of AADHD and wished to be assessed in this regard responded voluntarily to the related advertisement requesting participation in the study. Hence, the patients’ recruitment was based on their own desire to be checked regarding their probable ADHD symptoms, not the clinical impression. Then the ADHD diagnosis was evaluated based on the study tools.

The participants contacted the research assistant (RA) who had a MSc degree in clinical psychology through the phone number provided and if they met the inclusion criteria, an appointment was made. The inclusion criteria were age between 18 and 55 years, educational level of at least high school (to be able to complete the questionnaires and to rule out any intellectual disability), and a negative history of serious neurological disorder. The aims and stages of the study were explained to the participants and written consent were received. All the instruments and assessments were conducted free and the participants did not receive any kind of compensation. They could exit the study whenever they wanted. The research assistant tried to match the time of assessments best to the participants’ daily routine. At the end of the study, each participant was in case referred to the special clinic based on the approved diagnoses.

According to Pettersson et al. (2018), considering Alpha of 0.05, the sample size was calculated to be 120. The study protocol was evaluated and confirmed by the Ethics Committee of Tehran University of Medical Sciences.

Among the 143 people who contacted the study RA, 120 individuals (38.3% male) with a mean age of 34.35 ± 9.84 years met the inclusion criteria and cooperated with completing the study assessments and questionnaires. Among the participants, 53.3% were married, 40% were single, and 4.2% were divorced.

Measures

Structured Clinical Interview for DSM-5

This semi-structured diagnostic interview (First, 2014) assesses the presence of axis I psychiatric disorders based on DSM-5. The interview is conducted by an experienced clinician and usually needs 1 hour to complete. As far as we know, the structured clinical interview for DSM-5 (SCID-5) has not yet validated in Farsi speaking population. However, there is a multi-center psychometric study on the SCID-I-Persian translation by Sharifi et al. (2009).

Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders

This clinical interview for DSM-5 axis II personality diagnoses was designed by First et al. (2016). It assesses 17 different personality disorders. Test–retest reliability and content validity of the Farsi translation of the SCID-II have been supported (Bakhtiari, 2000), however, there is not yet any published psychometrics on the structured clinical interview for diagnostic and statistical manual of mental disorders (SCID-5-PD) Farsi translation.

Structured Clinical Interview for DSM-5^® Screening Personality Questionnaire

This questionnaire has been designed to be used as a screener for the structured clinical interview for DSM-5^® screening personality questionnaire (SCID-5-PD; First et al., 2016). This self-report questionnaire was used as a screening tool and if the participant’s score was higher than the cut-off, the SCID-5-PD would be performed.

Conner’s Adult ADHD Rating Scale-Self Report-Screening Version

This questionnaire (Hirsch et al., 2013) includes 18 items that are scored between 0 and 3 to assess the ADHD diagnosis. It has several versions and a short self-report form. For the 26-item short version, a minimum score of 65 is needed to diagnose AADHD. The Farsi version of the scale showed excellent internal consistency and high test–retest reliability (Davari-Ashtiani et al., 2014).

The Diagnostic Interview for ADHD in Adults

The Diagnostic Interview for ADHD in Adults (DIVA-5) is a semi-structured interview based on the DSM-5 criteria for ADHD. The DIVA-5 evaluates all the DSM-5 criteria for AADHD based on the clinical judgment of the interviewer regarding the participant’s answers and the accompanying person who knows the interviewee. DIVA provides multiple examples for each criteria and can help for a better decision about the existence or non-existence of the symptoms. Finally, the interview assesses the functional impairment in five domains due to AADHD. DIVA-5 should always be part of a broader assessment including biography, comorbidity, and developmental history.

This tool was firstly provided in the Dutch language named the DIVA 2.0 based on the DSM-IV diagnoses (Kooij & Francken, 2010). The DIVA 2.0 was translated into many different languages which were freely available at www.divacenter.eu. After the publication of the DSM-5, the DIVA-5 version of the interview was provided (Kooij et al., 2019), which is available at affordable costs and is updated in all languages.

After receiving permission from the DIVA Foundation and based on the designers’ standards, first, the interview was translated from Dutch to Farsi and then back-translated to the original language by two separate professional translators. The back-translated text was compared to the original interview and some modifications were done to keep the original meaning. The Farsi text was sent to four psychiatrists, four child and adolescent psychiatrists, and two general physicians to be checked regarding its face validity. All feedbacks were considered and the best consensus was used to modify in some words and sentences. This modified text was again checked with the Dutch version. The final text was sent to the editors of the DIVA Foundation to be reassessed and finalized. It is now accessible through the official website.

Procedure

All outpatient volunteers who were willing to be assessed in terms of ADHD symptoms and met the study inclusion criteria were enrolled, regardless of meeting ADHD diagnosis or not. All participants were evaluated through a comprehensive clinical assessment by a board-certified child and adolescent psychiatrist based on the DSM-5 criteria checklist including adult ADHD and other accompanying disorders. Then, the structured clinical interview for DSM-5 (SCID-5) was conducted to confirm the clinical diagnoses.

If the participant had a high score in the self-report SCID-5-SPQ, the SCID-5-PD was conducted to evaluate personality disorders more comprehensively. Both SCID-5 and SCID-5-PD were performed by the clinical psychologist’s expert in clinical interview.

In order to evaluate adult ADHD diagnosis more precisely, the CAARS-S-SV was completed by the participants. For those who got a score higher than the cut-off point, a psychologist who was skilled in performing the DIVA-5 and was unaware of the psychiatrist’s diagnosis conducted the semi-structured interview.

The diagnostic agreement between the results of the DIVA-5 performed separately by the psychiatrist and each of the two collaborator psychologists were evaluated. To evaluate test–retest reliability, the DIVA-5 interview was repeated by another trained psychol-ogist within two to four weeks after doing the first interview.

Statistical Analysis

Data were analyzed by SPSS software version 22. If Fleiss kappa was not possible by SPSS software, the online reCal 3 system was used. To examine the agreement between the clinical diagnoses, the DIVA-5 results, and the CAARS-S scores, diagnostic accuracy tests, including sensitivity, specificity, negative likelihood ratio (NLR), positive likelihood ratio (PLR), positive predictive value (PPV), and negative predictive value (NPV), and Kappa statistical methods were done.

Results

One hundred and twenty participants entered the study. Among the adults with ADHD, 51.5% had combined, 40.9% had inattentive, and 7.5% had hyperactive-impulsive presentations. The demographic and clinical characteristics of the participants are listed in Table 1.

Table 1.

Demographic and Clinical Characteristics of the Participants With and Without ADHD.

Clinical diagnosis based on the DSM-5	Number (%) 120	With ADHD (66.55%)	Without ADHD (54.45%)	p
Mean age: (min = 18, max = 55)	34.35 + 9.84 years	34.80 + 10.39	33.81 + 9.20	.589
Gender				.003
Male	46 (38.3%)	33 (50%)	13 (24.1%)
Female	74 (61.7%)	33 (50%)	41 (75.9%)
Marital status
Married	64 (53.3%)	34 (51%)	30 (55%)
Single	48 (40%)	27 (40.9%)	21 (38%)
Divorced	5 (4.2%)	3 (4%)	2 (3%)
Unknown	3	2 (3%)	1 (1%)
Education
Under diploma	16 (13%)	9 (13%)	7 (12%)
Diploma and higher	91 (75%)	51 (77.27%)	40 (74.07%)
MS and higher	13 (10%)	6 (9%)	7 (12.9%)
Occupation
Self-employed	43 (3.5%)	22 (33%)	21 (38%)
Housekeeper	32 (26%)	14 (21%)	18 (33%)
Government-employed	26 (21%)	20 (3%)	6 (1%)
Student	8 (6%)	5 (7%)	3 (5%)
Unemployed	11 (9%)	5 (7%)	6 (11%)
Comorbid psychiatric disorders (based on SCID-5)
Obsessive Compulsive Disorder	32	18	14	.868
Major Depressive Disorder	34	16	18	.272
Bipolar Disorder	18	15	3	.009
GAD	10	7	3	.319
Cyclothymia	2	1	1
Dysthymia	4	2	2
Social phobia	4	2	2
Personality disorders
Cluster A	2	2		.919
Cluster B	3	3
Cluster C	1	1
Others	1	0

As Table 1 shows, most individuals in the two groups were married, self-employed, and had education equal or higher than diploma. The most common comorbid psychiatric disorders were major depressive disorders (MDD), obsessive compulsive disorder (OCD), and bipolar disorder (BD) in the whole group, while BD was significantly higher in adults with ADHD compared to the control group. However, the two groups had no significant difference regarding the prevalence of personality disorders.

Table 2 summarizes the prevalence of ADHD diagnosis and presentations derived separately from the CAARS-S-SV, DIVA-5, and clinical psychiatric assessment (based on the SCID-5 following the DSM-5 criteria). As the table shows, the number of participants diagnosed with ADHD according to the psychiatrist’s judgment was higher than those who were diagnosed based on the DIVA-5. The details on the diagnostic agreement have been shown in Table 3.

Table 2.

Comparison of ADHD Diagnoses Derived From the Three Methods of Assessments.

Psychiatric diagnosis	Based on the CAARS-S-SV scores	Based on the DIVA-5	Based on the SCID-5 clinical assessment (DSM-5 criteria)
ADHD(−)	64	74	54
ADHD(+)	56	45	66
Combined	33	20	34
Inattentive	13	14	27
Hyperactive-impulsive	10	11	5
Total	120	119	120

Table 3.

The Results on Diagnostic Agreement Among the SCID-5 (DSM-5 Criteria), the CAARS-S-SV, and the DIVA.

		DIVA-5				Total
		None	Combined	Inattentive	Hyperactive	Total
SCID-5 diagnosis	None	53	0	1	0	54
	Combined	5	18	4	7	34
	Inattentive	14	3	9	1	27
	Hyperactive	2	0	0	3	5
	Total	74	20	14	11	119
		CAARS-S-SV				Total
		None	Combined	Inattentive	Hyperactive	Total
SCID-5 diagnosis	None	47	4	0	3	54
	Combined	7	22	3	2	34
	Inattentive	10	7	9	1	27
	Hyperactive	0	0	1	4	5
	Total	64	33	13	10	120
		CAARS-S-SV				Total
		None	Combined	Inattentive	Hyperactive	Total
DIVA-5	None	52	11	6	5	74
	Combined	2	16	2	0	20
	Inattentive	7	3	4	0	14
	Hyperactive	2	3	1	5	11
	Total	63	33	13	10	119

The results based on these three methods were analyzed at two levels. At the first level considering the ADHD presentations, the mean overall diagnostic agreement was 67.5%. The diagnostic agreement between the SCID-5 and the DIVA-5 was 69.16%, between the SCID-5 and the CAARS-S-SV was 65.3%, and between the DIVA-5 and the CAARS-S-SV was 68.33%. Fleiss kappa for these three methods was 0.481. Cohen kappa according to the SCID-5 and the DIVA-5 was 0.519, the SCID-5 and the CAARS-S-SV was 0.516, and the DIVA-5 and the CAARS-S-SV was 0.417.

At the next level of analysis, the ADHD diagnosis was compared with these methods, regardless of the ADHD presentation. In this case, the diagnostic agreement rate was on average 77.77%. The two by two table of DIVA-5 and SCID-5 diagnoses has been shown below (Table 4). This rate was 81.66% between the SCID-5 and the DIVA-5, 80% between the SCID-5 and the CAARS-S-SV, and 71.66% between the DIVA-5 and the CAARS-S-SV. The Fleiss kappa calculated for this comparison was 0.554. Cohen kappa which was calculated pairwise showed 0.642 agreement between the diagnoses based on the SCID-5 and the DIVA-5, 0.603 between the SCID-5 and the CAARS-S-SV, and 0.415 between the DIVA-5 and the CAARS-S-SV.

Table 4.

Diagnostic Agreement Details of ADHD Diagnosis Based on DIVA-5 and SCID-5.

	DIVA-5+	DIVA-5−	Total
SCID-5+	45	21	66
SCID-5−	1	53	54
Total	46	74	120

Sensitivity, specificity, PLR, NLR, PPV, and NPV for and the DIVA-5 and the CAARS-S-SV were calculated. The results have been presented in Table 5. As can be seen, the DIVA-5 is less sensitive and more specific than the CAARS-S-SV when the results of the two tools were compared to the clinical diagnoses by the psychiatrist.

Table 5.

The Results on Sensitivity, Specificity, PLR, NLR, PPV, and NPV for the DIVA and the CAARS-S-SV Compared to the DSM5 Criteria.

Parameter	DIVA-5		CAARS-S-SV
Parameter	Ratio (%)	CI (95%)	Ratio (%)	CI (95%)
Sensitivity	68.18	55.56–79.11	74.24	61.99–84.22
Specificity	98.15	90.11–99.95	87.04	75.10–94.63
PLR	36.82	5.24–258.46	5.73	2.83–11.60
NLR	0.32	0.23–0.46	0.30	0.45–0.19
PPV	97.83	86.51–99.68	87.50	93.41–77.56
NPV	71.62	63.89–78.26	73.44	64.44–80.83

To assess the test–retest reliability for the DIVA-5, the interview was repeated with 22 participants during a 2- to 4-week interval after the first assessment. The analysis showed a kappa of 0.857 (p = .001).

Table 6 represents the results regarding the diagnostic agreement which was assessed between the two psychologists and the child and adolescent psychiatrist performing the DIVA-5 for 21 participants. The Fleiss Kappa ranged from 0.589 to 0.933.

Table 6.

The Diagnostic Agreement Among the Three Interviewers Performing the DIVA-5.

DIVA items	Expected agreement	Observed agreement	Fleiss Kappa
A_1a	0.545	0.933	0.853
A_1b	0.722	0.9	0.64
A_1c	0.52	0.933	0.861
A_1d	0.502	0.9	0.799
A_1e	0.625	0.933	0.822
A_1f	0.58	0.867	0.683
A_1g	0.567	0.933	0.846
A_1h	0.58	0.933	0.841
A_1i	0.527	0.867	0.178
A_2a	0.536	0.967	0.929
A_2b	0.509	0.933	0.853
A_2c	0.505	0.933	0.865
A_2d	0.501	0.967	0.933
A_2e	0.527	0.967	0.929
A_2f	0.502	0.933	0.866
A_2g	0.509	0.933	0.864
A_2h	0.527	0.867	0.718
A_2i	0.609	0.967	0.915
pA_1a	0.502	0.933	0.866
pA_1b	0.545	0.867	0.707
pA_1c	0.505	0.867	0.731
pA_1d	0.501	0.933	0.867
pA_1e	0.514	0.8	0.589
pA_1f	0.536	0.9	0.785
pA_1g	0.536	0.867	0.713
pA_1h	0.556	0.9	0.775
pA_1i	0.501	0.933	0.867
pA_2a	0.514	0.933	0.863
pA_2b	0.505	0.933	0.863
pA_2c	0.502	0.967	0.933
pA_2d	0.567	0.967	0.923
pA_2e	0.509	0.967	0.932
pA_2f	0.556	0.9	0.775
pA_2g	0.58	0.933	0.841
pA_2h	0.502	0.9	0.799
pA_2i	0.567	0.9	0.769

Note. A₁ = inattention criteria in adulthood; A₂ = hyperactivity-impulsivity criteria in adulthood; pA₁ = inattention criteria in childhood; pA₂ = hyperactivity-impulsivity criteria in childhood.

Discussion

This study evaluated the validity and reliability of the DIVA-5, a semi-structured interview to evaluate attention-deficit hyperactivity disorder in adults. As far as we know, this is the first study which assessed the psychometrics of the DIVA-5 based on the DSM-5 criteria and is the first using a DIVA-5 Farsi translation. The results supported the interview as a valid and reliable tool to diagnose AADHD in Farsi speaking individuals.

Among the participants who underwent the full assessments, 66.5% were diagnosed as AADHD based on the psychiatric comprehensive evaluation (the SCID-5 and open clinical assessment), of whom most had the combined presentation type (51.5%), followed by inattentive and hyperactive-impulsive presentation types. This kind of distribution of ADHD presentation types in adults is consistent with most related studies. For example, Deberdt et al. (2015) reported that 63.2% of their participants had combined type ADHD, followed by inattentive (29.6%) and hyperactive-impulsive type (7.2%). However, Epstein and Kollins (2006) reported the inattentive presentation as the most prevalent type in a group of adults presented to a university ADHD clinic. This finding may be the result of different setting for recruitment of their patients. It may also explain the different comorbidities in these AADHD presentations; major depression, substance use, and anxiety disorders were more prevalent concurrent conditions in this group to the groups with other ADHD presentation. In the current study, the most common comorbid disorder in both the ADHD and non-ADHD groups were OCD, depression, and bipolar disorder. Bipolar disorder was significantly higher in the AADHD group, which is consistent with several studies reporting high prevalence of bipolar disorders in the AADHD population (Marangoni et al., 2015; Wingo & Ghaemi, 2007). Deberdt et al. (2015) reported depression, followed by anxiety disorders as the most common comorbidities with adult ADHD participants. The group with AADHD in a study by Ramus-Quiroga et al. (2019) also showed a higher rate of psychological complications (74.1%) than the non-ADHD group (55.5%). As the comorbidity results derived from the current study is consistent with previous research, this further suggests the validity of the data reported here.

Diagnostic Agreement

The DIVA-5 was compared with the SCID-5 diagnoses and the CAARS-S-SV scores. The agreement was higher between the DIVA-5 and SCID-5 diagnoses compared to the DIVA-5 and CAARS-S-SV when considering either the ADHD diagnosis overall, or the specific clinical presentations of ADHD. This difference was lower for the specific subtypes which is not unexpected given they can vary depending on meeting threshold values by only one-item. Cohen’s kappa was highest for the agreement between SCID-5 and DIVA-5 (>0.6). It was fair to good for the others.

Criterion Validity

To evaluate the criterion validity, DIVA-5 results were compared with the clinical diagnosis of a child and adolescent psychiatrist as the gold standard based on the DSM5 criteria and after conducting the SCID-5. The agreement between the SCID-5 and DIVA-5 diagnoses was greater when the ADHD was considered regardless of the presentation (82% vs. 69%). This was the same for the Cohen kappa for the two conditions (0.64% vs. 52%). The analyses showed a sensitivity of 68%, a specificity of 98%, a PPV of 98%, and an NPV of 72% for the DIVA-5. Compared to these findings, in a study on the Spanish translation of the DIVA 2.0 which is based on the DSM-IV-TR (Pettersson et al., 2018), lower specificity (73%), and higher sensitivity (90%), PPV (80%) and NPV (85%) were found. They reported the most even distribution of specificity and sensitivity for the DIVA 2.0 in comparison with other research tools. The high specificity of the Farsi DIVA-5 which shows few false-positive cases of AADHD, suggesting that this interview can be used as an instrument to confirm the individuals who are diagnosed with AADHD by other methods including clinical interview and self-report questionnaires. It showed a great predictive ability to find adults with ADHD. Sensitivity of 68% supports the ability of the interview to find a good proportion of true positives. However, it suggests that somecases of the A-ADHD might have been missed. Therefore, in concordance with the general rule to use the DIVA-5, in clinical practice it should always be part of a broader assessment including biography, comorbidity and developmental history. It is better not to use the interview as the only method of assessment, and it can help mostly when added to other tools that have a higher rate of sensitivity. Ramos-Quiroga et al. (2019) evaluated the criterion validity of the Spanish translation of DIVA-2 using the Conners’ Adult ADHD Diagnostic Interview for DSM-IV (CAADID). The participants were referrals to an Adult ADHD clinic. They found 100% diagnostic agreement between the two interviews and could not analyze PPV, NPV, sensitivity, and specificity. Before the release of the DIVA 2.0, the CAADID was the only comprehensive semi-structured interview for the diagnosis of ADHD in adults. A study by Epstein and Kollins (2006) examined the criterion validity of the CAADID compared to DSM-IV criteria. They reported fair to a good range of Kappa statistics for ADHD symptoms for current and retrospective childhood reports.

Concurrent Validity

Comparison of the DIVA-5 diagnoses and the CAARS-S-SV scores was done to evaluate the concurrent validity of the DIVA-5. The agreement between the self-report questionnaire and the DIVA-5 diagnoses was lower for the ADHD subtypes entering the analysis compared to when examining the presentations of the disorder as well (68% vs. 71%). Cohen kappa was the same for the two conditions (0.41). We found a sensitivity of 74%, a specificity of 87%, PPV of 87%, and NPV of 73% for the CAARS-S-SV. Compared to these findings for the DIVA-5, the interview showed higher specificity and lower sensitivity in screening adults with ADHD. However, both tools were more specific than sensitive, supporting the issue that they are better to be used as adjuncts to clinical assessments.

The two above mentioned studies on the psychometrics of the DIVA-2 (Pettersson et al., 2018; Ramos-Quiroga et al., 2019) used other self-report instruments than the CAARS-S-SV. Hence, a direct comparison of our results with those is not possible. Ramos-Quiroga et al. (2019) found a good correlation between the DIVA 2.0 with the Wender Utah Rating Scale (WURS; r = .544, p < .0001) and the ADHD-Rating Scale (r = .720, p < .0001). Pettersson et al. (2018) used the Adult ADHD Self-Report Scale (ASRS v.1.1) with a battery of neuropsychological tests to discriminate adults with ADHD among other referrals with psychiatric problems. They showed high sensitivity and good specificity for the ASRS.

There are other studies evaluating the validity of different versions of the Conner’s Adult ADHD rating scale, however, not in comparison with the DIVA. For example, the German version of the CAARS showed moderate but significant criterion validity with DSM based ADHD subtypes (Christiansen et al., 2012). In the recent study, the agreement and Cohen kappa for the SCID-5 and CAARS-S-SV was lower than the rate between SCID-5 and DIVA-5 and higher than between the DIVA-5 and CAARS-S-SV.

Reliability

Test–retest of the DIVA-5 showed excellent reliability (Landis & Koch, 1977). Inter-rater agreement was good or perfect for most items of the interview too. None of the existing published articles on the DIVA reported reliability findings to be compared with the recent study. A test–retest analysis of the CAADID to make a diagnostic judgment for AADHD (Epstein & Kollins, 2006) showed adequate results for all symptoms in criteria (A).

Limitations

This study is the first evaluation of the psychometrics of DIVA-5, a DSM-5 version of a semi-structured interview to diagnose ADHD in adults. Moreover, it used a Farsi translation of the DIVA-5 as the first specific instrument in this age group of Farsi-speaking individuals in Iran. The study compared the DIVA-5 with a semi-structured clinical interview as well as a self-report questionnaire.

However, the findings should be considered in the light of several limitations as follows:

(1) The participants were all concerned about some problems related to ADHD and volunteered to be assessed. This might have led to selection bias, therefore, the findings cannot be generalized to adults with ADHD in the general population, or in clinical populations where the diagnosis of ADHD was not being sought by participants.

(2) The instructions of the DIVA-5 suggest that the interview may be more reliable when someone familiar with the interviewee’s (partner/parent) current problems and background attends the session. This way, they can answer the questions that the interviewee does not remember or is not sure about, or can help to clarify the details of any behavioral symptoms of functional impairments. Although the study research assistants had asked the participants to attend the interviews by accompanying persons, most of them did not. Relying on the reports of the interviewee alone can decrease the reliability of the information. Of course, as the DIVA-5 is a semi-structured tool, in the end, it is the clinical judgment of the interviewer to decide on the presence of individual symptoms and impairments, and thereby the diagnosis of ADHD.

(3) As the participants with AADHD have a high rate of other psychiatric disorders, we tried to detect axis I and II disorders using the SCID-5 and SCID-5-PD semi-structured interviews. However, the SCID-5-PD was only performed when the participant’s score was higher than the cut-off derived from self-report. Hence, some individuals with personality disorder might have been missed.

Conclusion

The Farsi translation of the DIVA-5 is a valid and reliable tool to diagnose adults with ADHD.

Footnotes

Acknowledgements

We are grateful to all adults who participated in this study. This study was granted by Tehran University of Medical Sciences as the thesis for child and adolescent psychiatry subspecialty of the first author.

Declaration of Conflicting Interests

The author(s) declared no potential conflicts of interest with respect to the research, authorship, and/or publication of this article.

Funding

The author(s) received no financial support for the research, authorship, and/or publication of this article.

ORCID iD

Zahra Shahrivar

Author Biographies

Lida Zamani is a child and adolescent psychiatrist working as an assistant professor at Tehran University of Medical Sciences. Her research interest is comorbidities of adult ADHD.

Zahra Shahrivar is an associate professor of child and adolescent psychiatry in the Department of Psychiatry at Tehran University of Medical Sciences. Her interests are neurodevelopmental disorders, parenting practice, community-based mental health and psychometrics.

Javad Alaghband-Rad, MD, is an associate professor of psychiatry in the Department of Psychiatry at Tehran University of Medical Sciences. His research interests include childhood psychiatric disorders, first episode psychoses, adult developmental disorders and magnetic stimulation treatments.

Vandad Sharifi, MD, is an associate professor of psychiatry in the Department of Psychiatry at Tehran University of Medical Sciences. His research interests include community-based mental health services, psychiatric epidemiology, and psychiatric rehabilitation.

Elham Davoodi is a MA in clinical psychology. Her research interests are mainly in the field of developmental psychopathology.

Shadi Ansari is a MA in clinical psychology working as a research assistant with Roozbeh hospital, Tehran University of Medical Sciences

Fatemeh Emari is a student of anatomy and cell biology collaborating with McGill University program: biological, biomedical and life sciences.

Dora Wynchank, MD, PhD, is a psychiatrist and researcher at the Expertise Centre, ADHD, in The Hague, The Netherlands. Her interests include sleep disorders, circadian rhythm, and adult ADHD and she is editor of the DIVA Foundation.

J. J. Sandra Kooij is a psychiatrist at the Expertise Center Adult ADHD at PsyQ in Hague, and associate professor at Amsterdam UMC, location VUMc, Amsterdam, The Netherlands. Her research interests include adult ADHD, sleep and health, ADHD in older people, and ADHD in women.

Philip Asherson is a professor of Neurodevelopmental Psychiatry at MRC Social Genetic and Developmental Psychiatry, Institute of Psychiatry, King’s College London. His research interests are in the aetiology, diagnosis, treatment, and delivery of effective clinical services for adults with ADHD.

References

Amiri

Ghoreishizadeh

M. A.

Sadeghi-Bazargani

Jonggoo

Golmirzaei

Abdi

Safikhanlo

Asadollahi

(2014). Prevalence of adult attention deficit hyperactivity disorder (adult ADHD): Tabriz. Iranian Journal of Psychiatry, 9(2), 83.

Arabgol

Hayati

Hadid

(2004). Prevalence of attention-deficit/hyperactivity disorder in a group of university students. Advances in Cognitive Science, 6(1), 73–87.

Aragonès

Piñol

J. L.

Ramos-Quiroga

J. A.

López-Cortacans

Caballero

Bosch

(2010). Prevalence of attention deficit hyperactivity disorder in adults according to the record of computerized primary care medical records. Spanish Journal of Public Health, 84(4), 417–422.

Asherson

Akehurst

Kooij

J. S.

Huss

Beusterien

Sasané

Gholizadeh

Hodgkins

(2012). Under diagnosis of adult ADHD: Cultural influences and societal burden. Journal of Attention Disorders, 16(Suppl. 5), S20–S38.

Bakhtiari

(2000). Mental disorders in patients with body dysmorphic disorder (MS dissertation). Iran University of Medical Sciences.

Barbaresi

W. J.

Colligan

R. C.

Weaver

A. L.

Voigt

R. G.

Killian

J. M.

Katusic

S. K.

(2013). Mortality, ADHD, and psychosocial adversity in adults with childhood ADHD: A prospective study. Pediatrics, 131(4), 637–644.

Barkley

R. A.

Brown

T. E.

(2008). Unrecognized attention-deficit/hyperactivity disorder in adults presenting with other psychiatric disorders. CNS Spectrums, 13(11), 977–984.

Barkley

R. A.

Fischer

Smallish

Fletcher

(2006). Young adult outcome of hyperactive children: Adaptive functioning in major life activities. Journal of the American Academy of Child & Adolescent Psychiatry, 45(2), 192–202.

Barkley

R. A.

Murphy

K. R.

Fischer

(2008). ADHD in adults: What the science says. Guilford.

10.

Christiansen

Kis

Hirsch

Matthies

Hebebrand

Uekermann

Abdel-Hamid

Kraemer

Wiltfang

Graf

Colla

Sobanski

Alm

Rösler

Jacob

Jans

Huss

Schimmelmann

B.G.

Philipsen

(2012). German validation of the Conners Adult ADHD Rating Scales (CAARS) II: Reliability, validity, diagnostic sensitivity and specificity. European Psychiatry, 27(5), 321–328.

11.

Conners

C. K.

Erhardt

Sparrow

M. A.

(1999). Conners’ Adult ADHD Rating Scales: Technical manual. Multi-Health Systems Inc.

12.

Davari-Ashtiani

Jazayeri

Arabgol

Razjouyan

Khademi

(2014). Psychometric properties of Persian version of Conners’ adult attention deficit/hyperactivity disorder rating scale (Screening form-self reporting). Iranian Journal of Psychiatry and Clinical Psychology, 20(3), 243–251.

13.

Deberdt

Thome

Lebrec

Kraemer

Fregenal

Ramos-Quiroga

J. A.

Arif

(2015). Prevalence of ADHD in nonpsychotic adult psychiatric care (ADPSYC): A multinational cross-sectional study in Europe. BMC Psychiatry, 15(1), 242.

14.

Dulcan

(1997). Practice parameters for the assessment and treatment of children, adolescents, and adults with attention-deficit/hyperactivity disorder. Journal of the American Academy of Child & Adolescent Psychiatry, 36(10), 85S–121S.

15.

Epstein

J. N.

Kollins

S. H.

(2006). Psychometric properties of an adult ADHD diagnostic interview. Journal of Attention Disorders, 9(3), 504–514.

16.

Fayyad

Sampson

N. A.

Hwang

Adamowski

Aguilar-Gaxiola

Al-Hamzawi

Andrade

L. H.

Borges

de Girolamo

Florescu

Gureje

Haro

J. M.

Karam

E. G.

Lee

Navarro-Mateu

O’Neill

Pennell

B. E.

Piazza

Posada-Villa

, et al. (2017). The descriptive epidemiology of DSM-IV adult ADHD in the world health organization world mental health surveys. ADHD Attention Deficit and Hyperactivity Disorders, 9(1), 47–65.

17.

First

M. B.

(2014). Structured clinical interview for the DSM (SCID). In The encyclopedia of clinical psychology (Vol. 1–6). Wiley.

18.

First

M. B.

Williams

J. B.

Benjamin

L. S.

Spitzer

R. L.

(2016). SCID-5-PD: Structured clinical interview for DSM-5^® personality disorders. American Psychiatric Association Publishing.

19.

Haavik

Halmøy

Lundervold

A. J.

Fasmer

O. B.

(2010). Clinical assessment and diagnosis of adults with attention-deficit/hyperactivity disorder. Expert Review of Neurotherapeutics, 10(10), 1569–1580.

20.

Hamzeloo

Mashhadi

Salehi Fadardi

(2016). The prevalence of ADHD and comorbid disorders in Iranian adult male prison inmates. Journal of attention disorders, 20(7), 590–598.

21.

Hirsch

Hauschild

Schmidt

M. H.

Baum

Christiansen

(2013). Comparison of Web-based and paper-based administration of ADHD questionnaires for adults. Journal of Medical Internet Research, 15(3), e47.

22.

Kooij

J. J. S.

Francken

M. H.

(2010). DIVA 2.0. Diagnostisch Interview voor ADHD bij volwassenen [DIVA 2.0. Diagnostic Interview ADHD in Adults]. DIVA Foundation.

23.

Kooij

J. J. S.

Francken

M. H.

Bron

T. I.

(2019). Diagnostic Interview for ADHD in adults (3rd ed.; DIVA-5). DIVA Foundation.

24.

Landis

J. R.

Koch

G. G.

(1977). An application of hierarchical kappa-type statistics in the assessment of majority agreement among multiple observers. Biometrics, 33(2), 363–374.

25.

Marangoni

De Chiara

Faedda

G. L.

(2015). Bipolar disorder and ADHD: Comorbidity and diagnostic distinctions. Current Psychiatry Reports, 17(8), 67.

26.

Mosalanejad

Lashkarpour

(2013). Prevalence of ADHD among students of Zahedan University of Medical Science in Iran. Iranian Journal of Psychiatry and Behavioral Sciences, 7(2), 83.

27.

Pettersson

Söderström

Nilsson

K. W.

(2018). Diagnosing ADHD in adults: An examination of the discriminative validity of neuropsychological tests and diagnostic assessment instruments. Journal of Attention Disorders, 22(11), 1019–1031.

28.

Ramos-Quiroga

J. A.

Bosch

Richarte

Valero

Gómez-Barros

Nogueira

Palomar

Corrales

Sáez-Francàs

Corominas

Real

Vidal

Chalita

P.J.

Casas

(2012). Criterion and concurrent validity of Conners Adult ADHD Diagnostic Interview for DSM-IV (CAADID) Spanish version. Revista de Psiquiatría y Salud Mental (English Edition), 5(4), 229–235.

29.

Ramos-Quiroga

J. A.

Nasillo

Richarte

Corrales

Palma

Ibáñez

Michelsen

Van de Glind

Casas

Kooij

J. J. S.

(2019). Criteria and concurrent validity of DIVA 2.0: A semi-structured diagnostic interview for adult ADHD. Journal of Attention Disorders, 23(10), 1126–1135.

30.

Rapport

L. J.

Friedman

S. L.

Tzelepis

Van Voorhis

(2002). Experienced emotion and affect recognition in adult attention-deficit hyperactivity disorder. Neuropsychology, 16(1), 102.

31.

Sharifi

Assadi

S. M.

Mohammadi

M. R.

Amini

Kaviani

Semnani

Shabani

Shahrivar

Davari-Ashtiani

Shooshtari

M. H.

Seddigh

Jalali

Seddigh

(2009). A Persian translation of the structured clinical interview for diagnostic and statistical manual of mental disorders: Psychometric properties. Comprehensive Psychiatry, 50(1), 86–91.

32.

Taylor

Deb

Unwin

(2011). Scales for the identification of adults with attention deficit hyperactivity disorder (ADHD): A systematic review. Research in Developmental Disabilities, 32(3), 924–938.

33.

Weiss

Hechtman

L. T.

(1993). Hyperactive children grown up: ADHD in children, adolescents, and adults (2nd ed.). Guilford Press.

34.

Wingo

A. P.

Ghaemi

S. N.

(2007). A systematic review of rates and diagnostic validity of comorbid adult attention-deficit/hyperactivity disorder and bipolar disorder. The Journal of Clinical Psychiatry, 68(11), 1776–1784.

Reliability,Criterion and Concurrent Validity of the Farsi Translation of DIVA-5: A Semi-Structured Diagnostic Interview for Adults With ADHD

Abstract

Keywords

Methods

Participants

Measures

Structured Clinical Interview for DSM-5

Structured Clinical Interview for Diagnostic and Statistical Manual of Mental Disorders

Structured Clinical Interview for DSM-5® Screening Personality Questionnaire

Conner’s Adult ADHD Rating Scale-Self Report-Screening Version

The Diagnostic Interview for ADHD in Adults

Procedure

Statistical Analysis

Results

Discussion

Diagnostic Agreement

Criterion Validity

Concurrent Validity

Reliability

Limitations

Conclusion

Footnotes

Acknowledgements

Declaration of Conflicting Interests

Funding

ORCID iD

Author Biographies

References

Structured Clinical Interview for DSM-5^® Screening Personality Questionnaire