Should psychological formulation replace diagnosis for psychiatrists?

Should psychological formulation replace diagnosis for psychiatrists?

Summary

Psychological formulation has been promoted as a replacement for diagnosis in mental health. This does not take into account the time pressures facing psychiatrists and their needs in clinical practice. Using an example of a published psychological formulation for psychosis it can be demonstrated that psychological formulation is inferior to psychiatric diagnosis for clinical functions (prognosis and choosing treatments). It is also inferior for research (classifying participants) and social roles (including access to welfare and administrative integration with the rest of healthcare). For psychiatrists, diagnosis is still superior but should be supplemented by diagnostic formulation which can incorporate psychological mechanisms.

Declaration of Interest

The author has written a book awaiting publication which has large sections focussed on the value of diagnosis in mental health.

Introduction

Psychological formulation – sometimes called case formulation - has been promoted to replace diagnosis in clinical practice ^1,2 particularly to increase understanding of the patient and choosing appropriate interventions ².

Psychological formulation involves the organising of relevant clinical material to map out and summarise a patient’s problems – in collaboration with the patient over the course of clinical work - using psychological theories and knowledge to understand the patterns of difficulties and suggest ways forward to resolve these problems ^2,3,4.

Statements on the superiority of psychological formulation compared to diagnosis tend to be made on the assumption that the clinician works in a particular way – that they can spend a long time with relatively few patients, will see these patients many times with short intervals between sessions (for example weekly), often in non-urgent situations at civilised hours and that psychosocial interventions are the main therapeutic activity.

For their outpatient practice psychiatrists average 60 minutes for first appointments and 18 minutes for follow-up appointments⁵ (usually separated by weeks or months between appointments) carrying caseloads up to ten times of their colleagues⁶. Emergencies/ overnight and inpatient assessments are often similarly brief and patients may only be seen once or twice by the same doctor. This makes providing psychosocial interventions difficult and where effective medications are available these tend to be used as these are usually taken outside of the assessments thus making effective therapeutic interventions possible.  

This paper briefly discusses the usefulness of psychological formulation for psychiatrists in clinical, research and social functions.

Usefulness of psychological formulation for clinical practice

To evaluate the usefulness of psychological formulation for clinical functions an example of psychological formulation will be used to illustrate the points; the formulation used to understand ‘Dan’ and his psychosis in Section 7 of ‘Understanding Psychosis and Schizophrenia’⁷. This example is chosen for two reasons. Firstly, it is given as a good example of psychological formulation to enable understanding of psychosis by the Division of Clinical Psychology so has been given a stamp of approval by professionals whose core professional skills include psychological formulation⁴. Secondly, it is used for a case of psychosis and my main clinical work is with patients with schizophrenia and other related conditions, so this is an example I can relate to my clinical practice.

The ‘medical model’ relies on advising on and helping deliver interventions that improve health based on the best possible evidence⁸. For clinical practice, classification systems (such as diagnosis or psychological formulation) utility is judged on whether they allow communication of useful information between clinicians/patients/carers, accurate conceptions of the nature of the problem, provide information on prognosis (such as rates of relapse), aetiology/ causation, complications and which interventions to use and their chances of success^9,10,11,12. This utility is context-sensitive (such as limitations imposed on the way the professional works by resource demands including limited time)⁹. This information is probabilistic in nature rather than giving absolute certainty. Psychological formulation also aims to allow understanding of the patient and their situation^1,2,3,4.

Looking at the formulation of ‘Dan’⁷ it is about 150 words in boxes connected by of arrows linking past experiences, current events, thoughts, beliefs, responses and feelings. It certainly communicates information such as how these different aspects may be connected and lists important aetiological factors for psychosis such as adverse childhood experiences and substances.

The formulation plausibly links hypervigilance to childhood bullying (but ignoring that this hypervigilance would presumably be directed towards others and not his own body); somatic hypervigilance may lead to a belief that there is a lump in the neck but does not explain why the patient believes that this lump is transmitting thoughts. The lacunae in this formulation reflects the gaps in the evidence for purely psychological mechanisms linking adverse childhood experiences and psychotic symptoms content and nature¹³. It is possible the substance misuse is more closely linked to the generation of psychotic symptoms and the other factors such as childhood bullying are linked to the substance misuse rather than as depicted in the formulation.

The formulation provides some prognostic information such as substance use, childhood bullying and criminal activity being linked to worse outcomes such as higher relapse rates or aggression. The formulation provides useful pointers for psychotherapy such as addressing substance misuse or the links between past experiences, current life difficulties and thoughts and feelings.

On the other hand, this formulation has important information gaps. It is not clear how severe the psychotic symptoms are and whether they persist when substances are not being abused (such as psychosis related to substance intoxication or relatively brief psychotic episodes triggered by substance misuse) or whether the psychosis is caused by mood problems caused by adverse childhood experiences and/or current life difficulties – all these distinctions have clinical relevance not just for prognosis but for appropriate interventions and treatment¹⁴. For example, if the psychotic symptoms are only present during substance intoxication or for short periods afterwards then prolonged antipsychotic treatment is unlikely to be indicated.

In the text before the formulation of ‘Dan’ the diagnosis of schizophrenia is mentioned. This diagnosis is associated with a great deal of information such as factors associated with increased risk, associated differences in biological or psychological factors and the effectiveness of interventions¹⁵. Diagnostic constructs are associated with information that is learned by the clinician that allows easier recall and use of this information under time/ fatigue/ workload and other pressures in the context of brief appointments or emergency/ overnight working practices of most psychiatrists.

Schizophrenia as a diagnostic construct is defined by the absence of known organic factors explaining the clinical picture, nevertheless it is associated with multiple factors known to increase the risk of meeting the diagnostic criteria such as childhood adverse experiences and cannabis^16,17. Like many general medical conditions there is no single fixed outcome but we have evidence for a range of probable outcomes including recovery and their likelihood¹⁸. We also have evidence for the likely effectiveness of interventions in people meeting schizophrenia criteria such as antipsychotics¹⁹ and cognitive behavioural therapy for psychosis²⁰ to guide choice of therapeutic interventions.

An additional diagnosis – such as substance harmful use- will need to be used to indicate the important role of substance use affecting prognosis and to influence clinical management to include providing information on the effect of substances to the patient and referring for a psychosocial intervention to address the substance misuse. Two diagnostic constructs may thus need to be used but the total word count is still far less than that of the formulation. Although the diagnostic construct schizophrenia does not explicitly state the causative factors of this case many general medical diagnostic constructs - such as type 2 diabetes - do not state the causative factors in their title and even when a necessary cause is included in the title – such as tuberculosis – other important factors in the aetiology are not included in the title (for instance the role of overcrowded housing in tuberculosis)²¹. Psychosis is a very broad concept that some extend to unusual experiences that cause little harm or dysfunction to more severe states that can meet diagnostic criteria for schizophrenia or bipolar disorder²². It is therefore useful to identify a more specific classification than psychosis alone¹⁴ to provide more useful information to guide clinical decision making for the individual patient.

Research and Social Functions

Psychological formulation – except psychodynamic formulation - has disappointing reliability²³ impairing its usefulness in classifying participants in research. Conceptually as a classification system for research, psychological formulation would have to be reduce the emphasis on the individual-specific factors as they will vary a lot between individuals – they will need to focus on a few common elements otherwise the number of participants in each group would be too small and underpowered to detect statistically significant differences between groups. Although not suitable as a classification system for research, psychological formulation can be used as part of an intervention in studies evaluating their effectiveness.

There are many social functions of a health classification such as diagnosis – administrative (activity measurement and billing), organisational (services organised according to diagnosis or group of conditions), statistical (incidence/ prevalence and planning healthcare provision on basis of need) and access to benefits or special educational support²⁴. Given the high word count of psychological formulation, it’s highly individualised nature and the lack of horizontal integration with general medical diagnostic systems it is not very suitable as system for administrative, organisational, statistical or access to benefits or special educational support. Most health activity involves payment from third parties such as health insurance or government and providers are required to provide activity data to justify their expenditure – psychological formulation would not satisfy the requirements of third party funders as it would differ greatly between individuals making counting categories difficult. Similar problems apply for statistical and organisational purposes. Psychiatric diagnosis for all its faults is the most convenient system to fulfil administrative/statistical social functions²⁵.

To gain access to benefits for example, a diagnosis – with empirical evidence of association with impaired functioning and a recognised classified entity in ICD10 for instance – with additional description of impaired functioning in the individual is more likely to convince welfare agencies of eligibility for benefits – it relies on empirical evidence (lacking for psychological formulation in comparison to psychiatric diagnosis²)  and a standardised recognised classification not just expert assertion and is compatible with existing systems for general medical conditions.

Discussion

Psychological formulation is unable to replace psychiatric diagnosis for psychiatrists’ clinical work for the contexts in which they work – such as brief appointments or emergency/ night work – and requirements for empirical-based advice and decision-making⁸ as much of the evidence base is based on diagnostic classifications². This statement carries caveats. For those – including some psychiatrists – whose clinical work uses primarily psychosocial interventions in the context of long appointments with relatively few patients during routine hours may find psychological formulation more useful.  For some ‘organic’ conditions such as dementia or general medical conditions with psychological problems then psychological formulation will be used to address specific problems rather than overall management of the primary condition. Conversely for conditions where medication has little benefit – such as where people meet borderline personality disorder diagnostic criteria – then management plans would preferably be based on psychological formulation rather than diagnosis. This difference in clinical utility of psychological formulation depending on presenting condition illustrates the usefulness of psychiatric diagnosis – different conditions have different implications such as range of likely outcomes, treatment responses and usefulness of psychological formulation.

For the social functions of classification – such as administrative, statistical or access to benefits – diagnosis is superior to psychological formulation. Psychological formulation would not easily used as a classification system for participants in research making it harder to close the evidence gap with diagnostic based research. Diagnostic formulations – although disagreement exists as to what they should contain ^26,27 and can exclude or include psychological theories³ including incorporating psychological formulations from other professionals or their own work – can incorporate both the diagnosis (and differential diagnosis) as well as individual predisposing, precipitating, perpetuating and protective factors in the patient³.

For general medical conditions it is common practice to use additional information apart from the diagnosis to help with estimating prognosis and guiding clinical decision-making such as the use of staging for cancers, Glasgow Coma Scale for acute brain injury, DAS28 for rheumatoid arthritis and Modified Blatchford Score in acute upper gastrointestinal haemorrhage²¹.

Diagnostic formulation – beginning with a diagnosis then expanding with further relevant individual information and can include psychological theorising – is the best complex classification system for most psychiatrists in the context of the way they work and their need to provide evidence-based advice to the patient and to guide decision-making. It can build on the empirical information attached to a diagnosis with further individual and contextual factors (which may provide additional empirical evidence affecting prognosis or clinical management or aid further understanding of the patient and their predicament³). Diagnosis can also be combined with psychological formulation as they provide supplementary information.

References

1.  Division of Clinical Psychology. Classification of Behaviour and Experience in Relation to Functional Psychiatric Diagnoses: Time for a Paradigm Shift. DCP Position Statement. British Psychological Society 2013

2. Macneil, CA, Hasty, MK, Conus, P. and Berk, M. Is diagnosis enough to guide interventions in mental health? Using case formulation in clinical practice. BMC Medicine 2012; 10: p.111

3. Summers A., Boland B., Dave S., Gill H., Ingrams C. & Padakarra S. Occasional Paper 103: Using formulation in general psychiatric care: good practice. Royal College of Psychiatrists, 2017

4. Division of Clinical Psychology. Good Practice Guidelines on the use of psychological formulation. British Psychological Society 2011

5. Creed F. How consultants manage their time.  BJPsych Advances 1995; 1: 65-70

6. Tyrer P, Al Muderis O & Gulbrandsen G. Distribution of case-load in community mental health teams. 2001; Psychiatric Bull 25: 10-12

7. Cooke, A., Basset, T., Bentall, R., Boyle, M., Cupitt, C. and Dillon, J. Understanding psychosis and schizophrenia. British Psychological Society, 2014

8. Shah, P & Mountain, D. The medical model is dead--long live the medical model Br J of Psychiatry 2007; 191: 375-377

9. Kendell R.E. & Jablensky A. Distinguishing between validity and utility of psychiatric diagnoses. Am J Psychiatry 2003; 160: 4-12

10. Kendler KS. The Nosologic Validity of Paranoia (Simple Delusional Disorder).  Arch Gen Psychiatry 1980; 37: 699-706

11. First MB, Pincus HA, Levine JB, Williams JBW, Ustun B and Peele R. Clinical utility as a criterion for revising psychiatric diagnoses.  Am J Psychiatry 2004; 161: 946-954

12. Tennant N & Thompson IE. Causes and Logic in Epidemiological Psychiatry 2: Causal Models and Logical Inference. Br J Psychiatry 1980; 137: 579-582

13. Gibson, LE, Alloy, LB & Ellman, LM. Trauma and the psychosis spectrum: A review of symptom specificity and explanatory mechanisms Clin Psychol Rev 2016; B: 92-105

14. Frances A. A report card on the utility of psychiatric diagnosis. World Psychiatry 2016; 15: 32-3

15. Matheson, SL, Shepherd, AM & Carr, VJ. How much do we know about schizophrenia and how well do we know it? Evidence from the Schizophrenia Library Psychol Med 2014; 44: 3387-3405

16. Belbasis L, Köhler CA, Stefanis N, Stubbs B, van Os J, Vieta E, Seeman MV, Arango C, Carvalho AF, Evangelou E. Risk factors and peripheral biomarkers for schizophrenia spectrum disorders: an umbrella review of meta‐analyses. Acta Psychiatr Scand 2018 Feb;137: 88-97.

17. Radua J, Ramella‐Cravaro V, Ioannidis JP, Reichenberg A, Phiphopthatsanee N, Amir T, Yenn Thoo H, Oliver D, Davies C, Morgan C, McGuire P. What causes psychosis? An umbrella review of risk and protective factors. World Psychiatry 2018;17: 49-66

18. Menezes, NM, Arenovich, T & Zipursky, RB. A systematic review of longitudinal outcome studies of first-episode psychosis Psychol Med 2006; 36: 1349-1362

19. Leucht S, Leucht C, Huhn M, Chaimani A, Mavridis D, Helfer B, Samara M, Rabaioli M, Bächer S, Cipriani A & Geddes JR. Sixty years of placebo-controlled antipsychotic drug trials in acute schizophrenia: systematic review, Bayesian meta-analysis, and meta-regression of efficacy predictors. Am J Psychiatry 2017; 174:927-42

20. Jauhar S, McKenna PJ, Radua J, Fung E, Salvador R, Laws KR. Cognitive–behavioural therapy for the symptoms of schizophrenia: systematic review and meta-analysis with examination of potential bias. Br J of Psychiatry 2014; 204: 20-9

21. Walker BR, Colledge NR, Ralston SH & Penman I. Davidson’s Principles and Practice of Medicine (22nd Edition), Churchill Livingstone (Elsevier) 2014

22. Van Os, J & Reininghaus, U.  Psychosis as a transdiagnostic and extended phenotype in the general population World Psychiatry 2016; 15: 118-124

23. Flinn, L, Braham, L & Nair, R. How reliable are case formulations? A systematic literature review. Br J Clin Psychol 2015; 54: 266-290

24. Rose, N. What is Diagnosis For?  Lecture given at the Institute of Psychiatry (London) in 2013. Transcript can be found at http://nikolasrose.com/wp-content/uploads/2013/07/Rose-2013-What-is-diagnosis-for-IoP-revised-July-2013.pdf

25. Boyle M. The problem with diagnosis. The Psychologist 2007; 20: 290-292