4 Research recommendations

The Guideline Development Group has made the following recommendations for research, based on its review of evidence, to improve NICE guidance and patient care in the future.

4.1 Comprehensive assessment versus a brief assessment

For people with a suspected common mental health disorder, what is the clinical and cost effectiveness of using a comprehensive assessment (conducted by mental health professional) versus a brief assessment (conducted by a paraprofessional)?

Why this is important?

Uncertainty remains about the accuracy and consequent identification of appropriate treatment by paraprofessionals in primary care. An assessment by a mental health professional is likely to result in more accurate identification of problems and appropriate treatment, but is likely to entail greater cost and potentially significant longer wait times for interventions, both of which can have deleterious effects on care.

This question should be answered using a randomised controlled design that reports short- and medium-term outcomes (including cost-effectiveness outcomes) of at least 12 months' duration.

4.2 'Walking across' from one assessment instrument to another

What methodology should be used to allow 'walking across' from one assessment instrument for common mental health disorders to another?

Why is this important?

A number of different ratings scales for depression and anxiety disorders are in current use, both in research studies and clinical practice. This makes obtaining comparative estimates of clinical outcomes at the individual level difficult when moving between research and clinical settings, and also between clinical settings. A method that allows for prompt and easy 'walking across' between assessment instruments would have a potentially significant clinical benefit in routine care.

This question should be answered by developing a new method and subsequent data analysis of existing datasets to facilitate comparison between commonly used measures.

4.3 GAD-2 for people with suspected anxiety disorders

In people with suspected anxiety disorders, what is the clinical utility of using the GAD-2 compared with routine case identification to accurately identify different anxiety disorders? Should an avoidance question be added to improve case identification?

Why is this important?

There is good evidence of poor detection and under-recognition in primary care of anxiety disorders. Case identification questions for anxiety disorders are not well developed. There is reasonable evidence that the GAD-2 may have clinical utility as a case identification tool for anxiety disorders, in particular generalised anxiety disorder, but there is greater uncertainly about its utility for other anxiety disorders, especially those with an element of phobic avoidance. Understanding whether the GAD-2 plus or minus an additional phobia question would improve case identification for different anxiety disorders would be an important contribution to their identification.

These questions should be answered by a well-designed cohort study in which the GAD-2 is compared with a diagnostic gold-standard for a range of anxiety disorders. The cost effectiveness of this approach should also be assessed.

4.4 Routine outcome measurement

In people with a common mental health disorder, what is the clinical utility of routine outcome measurement and is it cost effective compared with standard care?

Why is this important?

Routine outcome measurement is increasingly a part of the delivery of psychological interventions, particularly in the IAPT programme. There is evidence from this programme and from other studies that routine outcome measurement may bring real benefits. However, there is much less evidence for pharmacological interventions on the cost effectiveness of routine outcome measurement. If routine outcome measurement were shown to be cost effective across the range of common mental health disorders it could be associated with improved treatment outcomes, because of its impact on healthcare professionals' behaviour and the prompter availability of appropriate treatment interventions in light of feedback from the measurement.

This should be tested in a randomised controlled trial in which different frequencies of routine outcome measurement are compared, for example at the beginning and end of treatment, at regular intervals and at every appointment.

4.5 Use of a simple algorithm compared with a standard clinical assessment

For people with a common mental health disorder, is the use of a simple algorithm (based on factors associated with treatment response), when compared with a standard clinical assessment, more clinically and cost effective?

Why is this important?

There are well-established systems for the assessment of mental states, in primary and secondary care services, for common mental health disorders. One key function of such assessment is to identify both appropriate treatments and to obtain an indication of likely response to such treatments, thereby informing patient choice and leading to clinically and cost-effective interventions. Although the reliability of diagnostic systems is much improved, data on appropriate treatment response indicators remain poor, with factors such as chronicity and severity emerging as some of the most reliable indicators. Other factors may also be identified, which, if they could be developed into a simple algorithms, could inform treatment choice decisions at many levels in the healthcare system. Treatment choice can include complex assessment and discussion of options but the validity of such assessments appears to be low. Would the use of a number of simple indicators (for example, chronicity, severity and comorbidity) provide a better indication of likely treatment response? Using existing individual patient data, could a simple algorithm be developed for testing in a prospective study?

This should be tested in a two-stage programme of research: first, a review of existing trial datasets to identify potential predictors and then to develop an algorithm; second, a randomised controlled trial in which the algorithm is tested against expert clinical prediction.

4.6 Priority of treatment for people with anxiety and depression

For people with both anxiety and depression, which disorder should be treated first to improve their outcomes?

Why is this important?

Comorbidity between depression and anxiety disorders is common. At present there is little empirical evidence to guide healthcare professionals or patients in choosing which disorder should be treated first. Given that for many disorders the treatment strategies, particularly for psychological approaches, can be very different, guidance for healthcare professionals and patients on the appropriate sequencing of psychological interventions would be likely to significantly improve outcomes.

This should be tested in a randomised trial in which patients who have a dual diagnosis of an anxiety disorder and depression, and where there is uncertainty about the appropriate sequencing of treatment, should be randomised to different sequencing of treatment. The clinical and cost effectiveness of the interventions should be tested at the end of treatment and at 12 months' follow-up.

  • National Institute for Health and Care Excellence (NICE)