Ambiguity and vagueness in the ontology of psychotherapy

Types of ambiguity and vagueness (Appendix) have been the subject of considerable attention in general ontology, linguistics, information technology and healthcare.^{Reference Codish and Shiffman19} ‘Disambiguation’ refers to the resolution of both vagueness and ambiguity^{Reference Codish and Shiffman19} and requires the development of international domain-specific taxonomies, related glossaries, vocabularies or dictionaries, and the semantic mapping of the international taxonomy to national listings, directories and classifications.^{Reference Salvador-Carulla, Amaddeo, Gutiérrez-Colosía, Salazzari, Gonzalez-Caballero and Montagni18} ‘Psychotherapy’, which shares similar issues with ‘psychiatry’ in regard to vagueness,^{Reference Keil, Keuck and Hauswald20} has been identified as a major problem in the ontology of mental health interventions.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10}

Definition of psychotherapist as a professional

The lack of specification regarding the professionals delivering psychotherapy amplifies the ambiguity and vagueness associated with this intervention. This problem is present in standard classifications, such as the International Standard Classification of Occupations (ISCO-08),²⁴ the Healthcare Provider Taxonomy Code Set²⁵ and the International Classification of Mental Health Care (ICMHC)^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} and in professional organisations from different countries, such as the UK and USA.

According to ISCO-08, psychologists are the core professionals providing psychotherapy, but in reality psychotherapy can be provided by a broad span of professionals with very different levels of qualification, such as social workers, specialised educators, nurses and physicians,^{Reference Lesage, Bland, Musgrave, Jonsson, Kirby and Vasiliadis12,Reference Kovess-Masfety, Alonso, Brugha, Angermeyer, Haro and Sevilla-Dedieu26} as well as delivered by apps and e-therapies with little or no human interaction.^{Reference Carlbring, Andersson, Cuijpers, Riper and Hedman-Lagerlof15} In addition, ISCO-08 identifies psychologists not in the ‘health professions’ group, but instead within ‘social and religious professionals’, together with sociologists, philosophers and social workers. Trainee psychologists or graduate psychologists not licensed and registered as clinical psychologists, who may also provide psychotherapy, are not classified in ISCO-08. Furthermore, the psychologist's tasks as defined by ISCO-08 are very broad, even noting that psychotherapist is included only among psychologists’ occupations²⁴ (Table 1; Appendix).

A similar picture can be found in the Health Care Provider Taxonomy Code Set in the USA, where providers such as ‘counsellor’, ‘psychologist’ and ‘psychoanalyst’ are classified under ‘Behavioural health and social service providers’ together with ‘poetry therapist’ and ‘social worker’.

In the ICMHC,^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} clinical psychologists are the core professionals providing psychotherapy, but other highly trained professionals without a psychology degree may also provide it. The level of training of the professional included in the ICMHC requires only that the psychological interventions be based on ‘well-defined theoretical models’, provided by professionals after ‘extensive training’ for high and intermediate levels of specialisation, albeit in these cases the exact meaning of ‘extensive training’ requires further specification. For instance, in France, residents in psychiatry are not really trained in psychotherapy, nevertheless psychiatrists bear the title of psychotherapist without the need for additional training.^{Reference Vasiliadis, Dezetter, Lesage and Drapeau3}

Some professional organisations, such as the Accredited Counsellors, Coaches, Psychotherapists and Hypnotherapists (ACCPH) in the UK (https://www.accph.org.uk/) or the Medicare refundable peer-support services in the USA,²⁷ do not make a clear distinction between these interventions and the professionals qualified to provide them, albeit counselling, emotional support^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} or peer-delivered therapies^{Reference Chibanda, Weiss, Verhey, Simms, Munjoma and Rusakaniko5} are usually differentiated from psychotherapy.

Clear requirements to be defined as a psychotherapist were introduced in Canada,^{Reference Vasiliadis, Dezetter, Lesage and Drapeau3} although the professionals involved vary from psychiatrists and psychologists to nurses and social workers. Continuous training is mandatory and is implemented by both the College of Psychology for the psychologists and other professionals practising psychotherapy and the College of Physicians for physicians practising psychotherapy.^{Reference Vasiliadis, Dezetter, Lesage and Drapeau3}

Health services providing psychotherapy

This level of vagueness and ambiguity makes it difficult to communicate with patients about where to go for help and who to see. Vagueness and ambiguity can affect the definition and cost of individual therapies in a single patient–professional care contact or visit. Moreover, they have an impact on the definition, planning and cost of services at the micro-level of the care system (individual services), the meso-level (local areas) and the macro-level (regions, states and countries)^{Reference Salvador-Carulla, Amaddeo, Gutiérrez-Colosía, Salazzari, Gonzalez-Caballero and Montagni18,Reference Meyer, Coroiu and Korner28} (Table 1; Appendix). According to the International Classification for Health Accounts’ Classification of Health Care Functions (ICHA-HC), in the System of Health Accounts (SHA),²⁹ individual psychotherapy should be provided by ‘psychological and behavioural rehabilitation’ (PBR) services. However, these PBR services are themselves defined very broadly. This makes it very difficult to assess the extent to which any particular PBR service is providing public psychotherapy interventions in comparison with PBR services in another region or another country and hence map relevant service coverage.^{Reference Salvador-Carulla, Amaddeo, Gutiérrez-Colosía, Salazzari, Gonzalez-Caballero and Montagni18,Reference Cetrano, Salvador-Carulla, Tedeschi, Rabbi and Gutierrez-Colosia30}

Definition of psychotherapy as an intervention

The vagueness and ambiguity of the distinction between psychotherapy and less-structured interventions, such as counselling or emotional support, is a major problem in the current classification of healthcare.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} We have previously conducted a content analysis of the different classifications of mental health interventions, such as the ICMHC, the Common Language for Psychotherapy (CLP) procedures and the KTL.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} When these classifications define psychotherapy as an intervention, different types of vagueness and ambiguity present, as summarised in Table 1. All current classifications suggest underspecification of the term psychotherapy. In the ICMHC for instance, psychotherapy refers more generally to ‘psychological interventions’. A ‘low-level’ psychological intervention may be limited to so-called ‘talking therapies’ in low-income countries,^{Reference Chibanda, Weiss, Verhey, Simms, Munjoma and Rusakaniko5} with no clear-cut theoretical models supporting these interventions. Therefore, ‘psychological intervention’ should be used not as a synonym of ‘psychotherapy’, but as a broader concept incorporating multiple forms of emotional support. In fact, health classifications are typically focused on defining different types of psychotherapy, such as cognitive–behavioural therapy (CBT), dialectical behaviour therapy (DBT), psychodynamic therapy and interpersonal therapy.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} Psychotherapists have identified about 100 different psychotherapy procedures to describe their interventions.^{Reference Borgo, Marks and Sibilia31} Similarly, SNOMED-CT in the UK has included 63 categories under the heading ‘psychotherapy’, but it does not provide a hierarchical ontology-based classification. For example, categories listed under ‘regime/therapy’ include ‘psychotherapy’, ‘eclectic psychotherapy’, ‘general psychotherapy’ and ‘integrative psychotherapy’, which all refer to general psychotherapy.³²

Table 1 Sources and types of ambiguity and vagueness with regard to psychotherapy

ISCO, International Standard Classification of Occupations; ICMHC, International Classification of Mental Health Care; ICHA-HC, Classification of Health Care Functions; PBR, psychological and behavioural rehabilitation; CLP, Common Language for Psychotherapy procedures; KTL, Klassifikation Therapeutischer Leistungen; CBT, cognitive–behavioural therapy.

Finally, several classifications define psychotherapy as a ‘conversation’ between a provider and a patient.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} This contributes to ambiguity and vagueness because it merely describes an observable activity and fails to provide enough information on the specific activity performed during the clinical encounter.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10}

Pathways to disambiguation

An unambiguous use of the term psychotherapy, the professionals involved and the settings, would permit the establishment and promotion of global standards, monitoring, resource allocation and accountability. We have therefore identified three international classifications that may be used together fulfil these purposes:

1. (a) the Description and Evaluation of Services and Directories for Long Term Care (DESDE-LTC): to standardise descriptions and classifications of long-term care services worldwide

2. (b) ISCO-08: to classify professionals providing psychotherapy

3. (c) ICHI: to define the term psychotherapy and differentiate it from other overlapping terms.

Strengths and limitations of these classifications

The DESDE-LTC distinguishes between ‘core health’, which refers to services providing direct clinical treatment, and ‘other care’, which includes more general facilities, such as accommodation, training, promotion of independence and autonomy, case management, employment support and social skills.^{Reference Salvador-Carulla, Alvarez-Galvez, Romero, Gutierrez-Colosia, Weber and McDaid33} Core health is usually provided by health professionals with over 3 years of training in health sciences, such as physicians and psychologists, whereas other care is provided by non-specialised staff.^{Reference Cetrano, Salvador-Carulla, Tedeschi, Rabbi and Gutierrez-Colosia30} A key advantage of the DESDE-LTC is that it does not merely list the names of the services or how they are funded. Instead, it clearly focuses on consistently describing the main activities or functions provided by the service. This approach has permitted regional and international comparisons between mental health services, a key step in systemic quality improvement typically missing from other classifications and indicators.^{Reference Cetrano, Salvador-Carulla, Tedeschi, Rabbi and Gutierrez-Colosia30,Reference Salvador-Carulla, Alvarez-Galvez, Romero, Gutierrez-Colosia, Weber and McDaid33} According to the DESDE-LTC, psychotherapy is an intervention that takes place in many services within a healthcare system. When this activity is the principal action or process that characterises a particular service, this activity is used as the main type of care (MTC) for defining a specific set of services in the local mental healthcare system. These services are ‘out-patient’ (characterised by a face-to-face contact for a duration lasting less than day care) and could be mobile (e.g. a vehicle-based clinic), non-mobile (e.g. hospital or health centre), online, individual or group therapy. Psychotherapy services are mainly but not only performed by a facility typically described as out-patient, non-acute, non-mobile care (Fig. 1).

Fig. 1 Example of how to classify a psychotherapeutic intervention using the International Standard Classification of Occupations (ISCO), the Description and Evaluation of Services and Directories for Long Term Care (DESDE-LTC) and the International Classification of Health Interventions (ICHI).

Moreover, professionals providing psychotherapy may be classified with ISCO-08 coding, as shown in Fig. 1.

Finally, WHO has been developing ICHI,^{Reference Fortune, Madden and Almborg34} which separates the coding of psychotherapy from counselling and emotional support.^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} This could usefully distinguish levels or types of intervention, from more highly standardised and theory-based techniques in psychotherapy, to other less defined or standardised treatments, better described as counselling, coaching, talking therapy and other interventions for emotional support. According to ICHI,^{Reference Castelpietra, Salvador-Carulla, Almborg, Fernandez and Madden10} psychotherapy is defined as an action ‘providing therapeutic communication, based upon the systematic application of psychological theory’. However, it does not specify setting and provider. This means that information on the professional who provided the intervention, and the setting where it was performed, would have to be captured using other classifications alongside the ICHI.^{Reference Fortune, Madden and Almborg34}

In Fig. 1, we show an example of how to integrate the three international classifications mentioned, in order to classify psychotherapeutic interventions.

There are some limitations to this integration. First, ICHI is still under testing. and there are problems related to its hierarchical rules to avoid ambiguity and excessive granularity.^{Reference Fortune, Madden and Almborg34} Second, ICHI's definition encompasses a broad list of different psychotherapies’ targets, such as mental functions or stress management, but does not detail any specific type of psychotherapy (CBT, DBT, psychodynamic and so on). Third, ISCO-08 does not provide a definition of which health professionals can be clearly identified as psychotherapists. A possible alternative to solve this problem could take into account the European Skills, Competences, Qualifications and Occupations (ESCO) classification.³⁵ Psychotherapist has a specific classification code in this system that has been updated recently. According to ESCO, however, ‘psychotherapists are not required to have academic degrees in psychology or a medical qualification in psychiatry’, which does not comply with the need to have specific occupations providing psychotherapy.

Possible future directions

In this paper we have emphasised the importance of the link between proper classification of mental health services providing psychotherapy and accountability for public expenditure on mental healthcare. Resources for mental health are too precious^{Reference Castelpietra, Nicotra, Pischiutta, Gutierrez-Colosía, Haro and Salvador-Carulla2} to sustain ambiguous investments. With growing mental healthcare needs, nations need to minimise the risks and maximise the benefits associated with public funding, including psychotherapy. They need to address the lack of clarity in how both psychotherapy treatments and professionals (e.g. psychotherapists) are described.

We have demonstrated the potential for ISCO-08, DESDE-LTC and ICHI to operate as a pathway to disambiguation of psychotherapy and to facilitate a better comparison between different standardised interventions across settings. Further improvement will require refinement and/or update of existing classifications, particularly ICHI and ISCO-08. This paper also sheds light on different issues related to health classifications which may be further scrutinised, suggesting extending this analysis to other critical terms in health systems research, such as case management. Future research should focus on the applicability of ICHI and DESDE-LTC for health classification purposes at the global level, including the costing of mental healthcare as part of future economic evaluations. This is underway already, within the PECUNIA project (an acronym derived from ‘ProgrammE in Costing, resource use measurement and outcome valuation for Use in multi-sectoral National and International health economic evaluAtions’). This work should continue to ensure that vital mental health resources are targeted to best effect, and real-world applications should be analysed in order to describe benefits in terms of supporting accountability and informing resource allocation.