In this study, we used a needs-based model adapted from previous studies that have proven useful for planning physician and nursing workforce requirements in other national health systems [7, 15]. The model was designed to estimate the number of clinical psychologists needed in the NHS based on population-level mental-health needs. It integrates epidemiological prevalence data with expert-defined treatment scenarios, each specifying the average number and duration of psychotherapy sessions by disorder type. The model followed a five-step process (Fig. 1).
Steps to estimate the number of clinical psychology professionals needed
The first step was to determine the prevalence of mental disorders in Spain using the National Health Survey (ENSE) published in 2017 [19] as the most recent and reliable official source available. The ENSE study population consists of individuals living in the main family households across the Spanish territory. The sampling design is stratified and executed in three stages, and data is collected through computer-assisted personal interviews conducted at participants’ homes. In the 2017 edition, data were collected from 37,500 households. The ENSE is representative at both the national and regional levels, and serves as a key instrument of territorial cohesion for the population monitoring of joint health strategies within the NHS. In addition, it proposes the following three broad and mutually exclusive categories for mental disorders, which were used in this study: depressive, anxiety (or common mental disorders), and other mental disorders (or severe mental disorders). The second step involved multiplying the prevalence of these three categories by the population over 15 years of age, to establish the number of identified cases susceptible to treatment. Individuals under the age of 15 years were not included in this study, because ENSE is conducted separately for children and adults [19]. Although the possibility of aggregating both age groups into a single-needs estimation model was considered, it was ultimately ruled out, because the two survey segments lacked sufficient methodological homogeneity in the type of questions, diagnostic criteria, classification systems and care pathways. In addition, disorders that are highly prevalent in childhood, such as attention–deficit/hyperactivity disorder (ADHD) and autism spectrum disorder, require specific modeling and will be addressed in a separate study. Consequently, future studies are planned to independently estimate the mental-health needs of children and adolescents. The third step establishes the percentage of cases susceptible to psychological treatment for the three categories based on theoretical–technical criteria and three treatment scenarios (protocolized, intermediate, and adjusted), as described below. These criteria were developed through expert consensus among the clinical psychologist authors of this study, who possess over 15 years of experience in clinical practice and/or clinical research. The consensus process was informed by their in-depth knowledge of the working conditions within the NHS, as well as by empirical evidence on psychological treatments synthesized in clinical practice guidelines [20,21,22]. In the fourth step, operationalized treatment models were assigned to each target population according to the three proposed scenarios. Each treatment scenario specifies the mean number of sessions and their duration (minutes), total average treatment per case (hours), and effective working hours for each professional (1586 h per year, according to the current collective bargaining agreement). Finally, in the fifth step, the average number of sessions and their duration per professional were used as criteria to estimate the number of full-time clinical psychologists required to deliver them.
Treatment scenarios for depression disorders
The protocolized scenario is consistent with the best recommendations from empirical research on psychological treatments for depression [16, 20], which are included in the National Institute for Health and Care Excellence (NICE) [21] and Spanish Ministry of Health’s clinical practice guidelines [22]. An optimal benchmark posits that 65% of patients diagnosed with a depressive disorder should be susceptible to receiving specialized psychological treatment as the first line of intervention. We excluded approximately 35% of patients considered susceptible to spontaneous remission [22]. The NICE also establishes a period of watchful waiting without specialized intervention for a similar percentage of cases [21].
Given that there is a disparity between what is efficacious in clinical trials and what is effective in routine clinical practice [23, 24], the protocolized scenario, although it may seem ideal, may not be feasible. Thus, in the absence of official data on demand, it can be reasonably assumed that the target population for treatment in Clinical Psychology could cover 45% of depression cases (intermediate scenario) or 30% of cases (adjusted scenario), as the remaining healthcare would be distributed among Primary Care, Psychiatry, or the private sector [21, 22]. In addition, the intermediate and adjusted scenarios warrant efficiency in the use of available resources by reducing the number and duration of sessions (see Table 1). This optimization is supported by the literature, which indicates an optimal dose of psychotherapy ranging from four to 24 sessions in routine practice settings. Nevertheless, the greatest percentage of change tends to occur in the first few sessions, and the improvement tends to plateau as the number of subsequent sessions increases [25, 26]. Regarding the duration of the sessions, usually around 45 and 40 min in the NHS, there is evidence to support the efficacy of this approach, even when the sessions are spaced monthly [27]. Consequently, the number of sessions and their duration in the intermediate and adjusted scenarios can be considered sufficient [26] if the frequency of sessions is close (i.e., weekly or every 2 weeks) [28].
Treatment scenarios for anxiety disorders
Anxiety disorders are clinical problems that respond well to evidence-based psychological treatments [16, 20, 26, 27]. However, these disorders are frequently polymedicated in Primary Care and arrive late for specialized treatment [29], despite the fact that patients with anxiety disorders generally prefer less invasive treatments such as psychotherapy [30, 31]. Similarly, we propose three treatment scenarios with targets of population coverage of 70%, 50%, or 40% of the cases. Likewise, each of these scenarios establishes treatment intensity criteria (mean number of sessions, minutes per session, and hours per specialist) ranging from ideal (protocolized) to minimally acceptable (adjusted), according to empirical literature [25,26,27,28]. This approach assumes, in each of the scenarios, what degree of improvement is desired in the treatment of anxiety [29], as well as what percentage of cases can be successfully resolved by other mental-health specialists, Primary Care in the public system, or by professionals in the private sector [32].
Treatment scenarios for other mental disorders
This category includes severe mental disorders, which are less prevalent than the previous ones but with significantly greater psychological suffering, lower functionality, and higher avoidable mortality [17, 33]. This category comprises psychotic, bipolar, personality, eating disorders, etc. However, this ENSE category is more heterogeneous than the other two categories in terms of diagnostic variability and, consequently, therapeutic implications. These conditions typically require tertiary care resources or highly specialized secondary prevention units/programs. As a result, the criteria and projections presented here should be considered preliminary, pending a more specific and detailed assessment that not only quantifies the required number of professionals, but also addresses the development of targeted programs, the creation of specialized units, and other structural considerations.
Table 1 shows that in the three scenarios, compared to depressive and anxiety disorders, there is a higher percentage of cases with severe mental disorders to be treated and a higher treatment intensity to be established based on the best available evidence [34] in such a way that clinical psychologists dedicate more hours of their clinical work to this type of patient. The reason for this is also to minimize the inverse care law, which shows that the availability of good healthcare tends to vary inversely with the need for it in the population served [35]. Therefore, an adequate approach should include Clinical Psychology services, together with Psychiatry and Specialized Mental Health Nursing, in a comprehensive treatment plan.
General considerations on treatment scenarios for the three diagnostic categories
The adjusted scenario estimates for the three broad diagnostic categories defined by the ENSE reflect a pragmatic balance, partially informed by scientific evidence that combines expert knowledge on clinical practice within the Spanish NHS with the criteria of efficiency and rational use of available resources. This adjusted scenario represents a substantial improvement over the current service provision scenario. However, from a perspective strictly aligned with clinical practice guidelines, this would likely still be insufficient and place a high workload on clinical psychologists. These limitations are acknowledged as part of a transitional strategy, subject to revision and improvement as better epidemiological tools are developed and investment in mental health increases.
Regarding the theoretical–technical criteria based on expert consensus, it should be noted that no formal consensus method, such as a Delphi technique, was used. First, the study was designed to inform the annual allocation of PIR positions, and a Delphi process was not feasible within the required timeframe. Second, there are few clinical psychologists in our setting with expertise in workforce planning. Third, existing clinical practice guidelines already provide indicators and parameters that offer a baseline consensus. Finally, the study adopts a flexible and exploratory approach aimed at generating reference scenarios rather than definitive figures.
Finally, the ENSE does not employ official diagnostic categories. As a broad guide, the following analogy can be made between the ENSE categories and the codes from the Chapter V of the International Statistical Classification of Diseases and Related Health Problems, 10th edition (ICD-10) [36]: the “Depression disorders” ENSE category corresponds to the following ICD-10 codes: F32–F39 (depressive mood [affective] disorders): the “Anxiety disorders” ENSE category corresponds to the following ICD-10 codes: F40–F48 (neurotic, stress-related and somatoform disorders). In addition, the “Other mental disorders” ENSE category encompasses the following ICD-10 codes: F20–F29 (schizophrenia, schizotypal and delusional disorders), F30–F31 (bipolar affective disorders), F50 (eating disorders) and F60–F69 (disorders of adult personality and behavior).
Sensitivity analysis
In addition, we performed a sensitivity analysis to calculate confidence intervals for the initially proposed workforce estimates. To do so, we computed the upper and lower limit of confidence intervals based on the common rate (15%) of scheduled psychotherapeutic appointments missed, because patients do not show up [37]. In this sense, these confidence intervals for workforce estimates take into account the possibility that patients attend fewer appointments than planned (no-shows) or require additional sessions.