Internet‐based field trials of the ICD‐11 chapter on mental disorders

The ICD‐11 chapter on mental, behavioural and neurodevelopmental disorders has been now finalized, and a substantial effort to train mental health professionals in the use of the relevant Clinical Descriptions and Diagnostic Guidelines (CDDG) is ongoing worldwide1, 2, 3, 4, 5. The many contentious issues that have been debated in the development of the chapter have been extensively dealt with in this journal6, 7, 8, 9, 10, 11, 12, 13, 14, 15.

The finalization of the chapter has been preceded by a vast programme of international field studies. These included Internet‐based and clinic‐based studies. The Internet‐based field studies have been implemented through the World Health Organization (WHO) Global Clinical Practice Network (GCPN). This now includes more than 16,000 clinicians from 159 countries (51% psychiatrists, 30% psychologists; 40% from Europe, 25% from Western Pacific, 24% from the Americas, 5% from Southeast Asia, 3% from Eastern Mediterranean, and 3% from Africa; 63% from high‐income countries, 37% from middle‐ and low‐income countries. The clinic‐based field studies have been conducted with the participation of WHO Collaborating Centres.

The Internet‐based field studies have used the case vignette methodology to examine clinical decision‐making in relationship to the proposed diagnostic categories and guidelines; while the clinic‐based (or ecological implementation) field studies have assessed the reliability and clinical utility of the diagnostic guidelines with real patients.

Several papers reporting on the results of the Internet‐based field studies have been published during the past two years, covering some of the main sections of the ICD‐11 chapter on mental disorders ¹⁶ .

In a paper comparing the diagnostic accuracy and clinical utility of ten mental disorders (schizophrenia, schizoaffective disorder, bipolar type II disorder, recurrent depressive disorder, moderate personality disorder, adjustment disorder, complex post‐traumatic stress disorder, binge eating disorder, bodily distress disorder, and compulsive sexual behaviour disorder), it was reported that the ICD‐11 CDDG had a significantly higher diagnostic accuracy (71.9% vs. 53.2%), higher ease of use, better goodness of fit, higher clarity, and lower time required for diagnosis compared to the ICD‐10 CDDG. Diagnostic accuracy was consistent across WHO regions and independent of participants’ clinical experience. There were no differences between medical doctors and psychologists in the diagnostic accuracy, but participants representing other health professions had more difficulty in identifying the correct diagnoses on the basis of the ICD‐11 CDDG ¹⁶ .

In an Internet‐based field study conducted with the participation of 2,288 mental health professionals registered with the GCPN, the ICD‐11 CDDG significantly improved the accuracy of the diagnosis of all feeding and eating disorders compared to the ICD‐10 CDDG, and attained higher clinical utility ratings. Similar results were obtained across five languages (English, Chinese, French, Japanese and Spanish). The inclusion of binge eating disorder and avoidant‐restrictive food intake disorder in the ICD‐11 reduced the use of residual diagnoses ¹⁶ .

A further Internet‐based field study carried out with the participation of 1,357 members of the GCPN in four languages (English, Spanish, Japanese and Russian) found similar overall levels of accuracy in diagnosing mood disorders using ICD‐11 and ICD‐10 CDDG, but the use of the ICD‐11 CDDG was associated with a greater accuracy in identifying depressive episodes in recurrent depressive disorder ¹⁶ .

An Internet‐based field study conducted with 1,840 mental health professionals registered with the GCPN found that clinicians were significantly more accurate in diagnosing generalized anxiety disorder, specific phobia and adult separation anxiety disorder when using ICD‐11 vs. ICD‐10 CDDG, and provided high clinical utility ratings for these disorders. The participants found the ICD‐11 CDDG easy to use, clear, and with a good fit to patients they saw in their clinical practice, although having some difficulty in distinguishing the boundary between disorder and normality for subthreshold cases of anxiety ¹⁶ .

An Internet‐based field study with 1,717 mental health professionals who were members of the GCPN found that the use of the ICD‐11 CDDG resulted in a more accurate differentiation of presentations of obsessive‐compulsive and related disorders from one another compared with the use of the ICD‐10 CDDG ¹⁶ .

A small improvement was also found in accurately diagnosing vignettes depicting schizoaffective disorder using ICD‐11 vs. ICD‐10 CDDG in an Internet‐based field trial involving 873 clinicians. However, the problem in identifying whether the presence of mood symptoms was more consistent with a diagnosis of schizoaffective disorder or a mood disorder was still observed ¹⁶ .

Finally, an Internet‐based field study carried out with 196 clinicians in three languages (English, Spanish and Japanese) found that the ICD‐11 led to a more accurate identification of severe irritability, and a better differentiation from boundary presentations. Notably, participants using the DSM‐5 mostly failed to apply the diagnosis of disruptive mood dysregulation disorder when it was appropriate ¹⁶ .

These results, adding to those of the clinic‐based field studies – showing that the interrater reliability for the main groups of mental disorders in the ICD‐11 ranged from .45 to .88 and was generally superior to that obtained with the ICD‐10, and that the ICD‐11 CDDG were perceived as easy to use, clear and understandable, corresponding accurately to patients presentations, and providing useful guidance about distinguishing each disorder from normality and from other disorders ¹⁶ – are reassuring about the clinical utility of the ICD‐11 chapter on mental, behavioural and neurodevelopmental disorders, at a time in which the implementation of the new diagnostic system is being planned worldwide.