Background
It is estimated that each year, 38.2% of the European population suffer from a psychiatric disorder [
1]. The economic burden from psychiatric disorders is excessive, not only because of high direct health care costs, but also because of indirect costs like sick days, disability, and early retirement [
1]. Psychotherapy is among the recommended and widely used interventions for most disorders [
2]. Specific types of psychotherapy have already been systematically reviewed, but the appropriate length of psychotherapy for all adult psychiatric disorders has not been reviewed previously. To present a complete overview of the evidence and to increase the statistical power, we will therefore in the present review include any adult psychiatric disorder. The major categories of adult psychiatric disorders listed in the
Diagnostic and Statistical Manual of Mental Disorders,
5th edition (DSM-V) [
3] are the following.
Attention Deficit Hyperactivity Disorder (
ADHD) is characterized by a persistent pattern of inattention and/or hyperactivity and impulsivity that significantly interferes with functioning and development [
3]. ADHD is one of the most common psychiatric disorders of childhood and adolescence, and it often persists into adulthood. The predominant characteristics of adult ADHD differ from typical ADHD characteristics in children. Symptoms of hyperactivity or impulsivity are typically less obvious in adults, whereas symptoms of inattention are more dominant [
4]. Epidemiologic studies of adult ADHD have estimated the current prevalence to be 4.4% in the USA and 3.4% internationally [
4,
5]. The total economic burden of ADHD in America has been estimated to be 31.6 billion US dollars in 2010 [
6] including both direct costs, other health care costs, health care costs for family members, and work loss of patients and their relatives.
Psychotic disorders are characterized by abnormalities in one or more of the following five domains: delusions, hallucinations, disorganized thinking, disorganized or abnormal motor behavior, and negative symptoms [
3]. The estimated annual prevalence of all psychotic disorders is 2.6% [
7]. The most common psychotic disorder is schizophrenia with an estimated median lifetime prevalence of 4.0 per 1000 and a lifetime morbid risk of 7.2 per 1000 [
8]. Annual costs for the schizophrenia population have been systematically reviewed and estimated to range from 94 million to 102 billion US dollars. Indirect costs contributed to 50–85% of the total costs associated with schizophrenia [
9].
Bipolar disorders are characterized by serious mood changes involving mood elevation (mania or hypomania) either alone or followed by major depressive episodes [
3]. Bipolar disorder subtypes include bipolar I and bipolar II. Bipolar I disorder is associated with manic episodes nearly always followed by major depressive and hypomanic episodes. Bipolar II disorder is associated with at least one hypomanic episode, at least one major depressive episode, and the absence of manic episodes. The international annual prevalence is estimated to be 0.4% for bipolar I disorder and 0.3% for bipolar II disorder [
10]. In 2009, the estimated annual direct and indirect costs of bipolar I and II disorders were 30.7 and 120.3 billion US dollars, respectively [
11].
Depressive disorders are characterized by the presence of a sad, empty, and irritable mood often accompanied by somatic and cognitive changes resulting in significant functional impairment [
3]. The most common depressive disorder is major depressive disorder (unipolar depression) with an annual prevalence of approximately 7% both in Europe [
1] and in the USA [
12]. The estimated annual economic burden of adults with major depressive disorder, including direct medical costs, workplace costs, and costs associated with comorbidities exceeded 200 billion US dollars [
13] in the USA in 2010.
Anxiety disorders are characterized by excessive and counterproductive feelings of fear and anxiety often accompanied by behavioral disturbances such as pervasive avoidance behaviors [
3]. Different anxiety disorders exist, which differ from one another in the types of objects or situations that induce hyperarousal or avoidance behavior [
3]. The prevalence of anxiety disorders is estimated to be 18% in the USA [
14] and 14% in European countries [
1], placing them among the most prevalent psychiatric disorders worldwide. Costs associated with anxiety disorders have previously been reported to be 46.6 billion US dollars in the USA [
15] including both direct and indirect costs.
Obsessive-
compulsive disorder (
OCD) is characterized by recurrent intrusive thoughts, images, or urges (obsessions) with or without repetitive mental or behavioral acts (compulsions) [
3]. OCD among adults has an annual prevalence of 1.2% and a lifetime prevalence of 2.3% [
16,
17]. The annual economic burden of OCD is estimated to be 2272 euros per patient when including both direct and indirect costs [
18].
Trauma-
and stressor-
related disorders are characterized by psychological distress following exposure to a traumatic or stressful event. The most common trauma disorder is post-traumatic stress disorder (PTSD) [
3]. PTSD is a prevalent and disabling disorder associated with delayed help seeking [
19]. The estimated annual prevalence of PTSD is 2% in Europe [
1] and 4.7% in the USA [
20], and the estimated lifetime prevalence is 3.9% across 26 countries ranging from low to high income [
21]. The total costs of PTSD per patient have been estimated to 1082 million euros including both direct and indirect costs [
18].
Eating disorders are characterized by a persistent disturbance in eating behavior resulting in altered consumption or absorption of food that significantly impairs health and psychosocial functioning [
3]. The most common eating disorders are anorexia nervosa, bulimia nervosa, and binge-eating disorder. Lifetime prevalence of anorexia nervosa, bulimia nervosa, and binge-eating disorder are estimated to be 0.9, 1.5, and 3.5%, respectively, among women and 0.3, 0.5, and 2.0%, respectively, among men [
22]. The estimated annual prevalence of eating disorders is 0.9% in the European population [
1]. Annual costs per patient are estimated to range from 1288 to 8042 US dollars [
23].
Personality disorders are characterized by enduring and inflexible patterns of emotional, behavioral, and interpersonal problems that deviate markedly from cultural expectations. According to DSM-V, the following nine personality disorders exist. Personality disorders onset in adolescence or early adulthood and are associated with great psychosocial distress and impairment [
3]. In a systematic review of the economic burden of personality disorders, the estimated direct and indirect costs were 11,126 euros for patients 12 months prior to seeking treatment. Direct medical costs accounted for two thirds of these costs, while the remaining costs were related to productivity losses [
24].
Description of the interventions
Different schools of psychotherapy exist. They are often divided into the following categories:
psychodynamic therapies,
cognitive and behavioral therapies,
humanistic therapies, and
systemic therapies [
2].
Psychodynamic (
or psychoanalytical)
therapies encompass the many approaches that are influenced by Freud’s psychoanalysis but have developed into different independent schools [
2]. Traditionally, psychodynamic therapies have been considered as long-term therapies, perhaps due to the notion that the uncovering of unconscious emotions and conflicts cannot be achieved with a fixed time limit [
25]. Long-term psychoanalytical psychotherapy has previously been systematically reviewed yielding different results [
26,
27]. Today, different lengths of psychodynamic therapies have been developed to treat different forms of psychopathology. In addition to traditional psychoanalysis, examples of long-term psychodynamic treatments are transference-focused psychotherapy (TFP), a psychodynamic treatment rooted in object relations theory lasting up to 3 years [
28,
29], and mentalization-based therapy [
30], an 18-month psychodynamic treatment rooted in attachment theory. Both are developed specifically to treat borderline personality disorder. Further, different variations of short-term psychodynamic therapy have been developed to treat a variation of common psychiatric disorders, most notably anxiety disorders, depressive disorders, certain behavior disorders, and personality disorders [
31]. Short-term psychodynamic therapies vary in treatment duration but typically last between 12 and 24 sessions [
31].
Cognitive and behavioral therapies (CBT) encompass many integrative approaches. Historically, behavior therapy (first wave CBT) developed from the learning theories of Pavlov [
32] and Skinner [
33]. An integration of a cognitive component to classical behavioral theories was first established by Beck [
34,
35], who developed what is now often referred to as second-wave CBT. CBT is now often delivered as a short-term treatment, typically lasting between 12 and 20 sessions, for a variation of common psychiatric disorders like depressive disorders [
34], anxiety disorders [
36], obsessive-compulsive disorder [
37], personality disorders [
38], and eating disorders [
39]. Different durations of CBT are also available for the treatment of schizophrenia [
40,
41]. Today, so-called third-wave cognitive therapies have emerged, characterized by more integrative approaches to psychotherapy, incorporating techniques from Buddhist mindfulness, psychodynamic therapies, or Gestalt therapy [
2]. These include dialectical behavior therapy [
42] and schema-focused therapy (SFT) [
29,
43], which are both long-term therapies for borderline personality disorder (up to 3 years), and acceptance and commitment therapy (ACT) [
44] and compassion-focused therapy (CFT) [
45], which are often delivered as short-term treatments for various psychiatric disorders [
46,
47].
Humanistic therapies are characterized by psychotherapy approaches derived from humanistic and existentialist philosophy. Major approaches within this orientation are person-centered therapy [
48], Gestalt therapy [
49], existential psychotherapy [
50], and process-experiential/emotion-focused therapy [
51]. All humanistic therapies share the notion of empathic understanding, the promotion of in-therapy experiencing, and a belief in the uniquely human growth tendency by applying a consistent person-centered view involving concern for each patient’s individual experience and differing needs [
2]. Humanistic therapies have not been developed to treat specific types of disorders and are traditionally considered open-ended, which is also aligned with the person-centered way of thinking. However, different lengths of humanistic therapies have been studied, e.g., PE-EFT as a short-term treatment (down to 5 weeks) for depressive disorders [
52] and as a 20-week treatment for trauma-related disorders [
53].
Systemic therapies are characterized by a systemic approach to psychotherapy defining patients’ problems as contextually rather than individually derived. Most often, the context of interest is the partner or the family, but it can also be a broader context, such as the extended family or a classroom [
2]. Different systemic therapies exist for different types of psychopathology. Examples are family-based therapy for eating disorders [
54], attachment-based family therapy for depressed adolescents (ABFT) [
55], parent management training for childhood conduct disorders [
56], psychoeducational family interventions for schizophrenia [
57] and bipolar disorder [
58], and systemic treatments for substance-use disorders [
59,
60]. Different lengths of systemic therapies exist. However, the typical duration is between 10 and 25 sessions.
Other forms of psychotherapy exist, e.g., interpersonal therapy (IPT) [
61] or cognitive-analytic therapy (CAT) [
62]. However, it is beyond the scope of this review to mention all new approaches to psychotherapy since the field is constantly expanding. Further, despite the existence of well-established manualized and evidence-based approaches to psychotherapy, a large proportion of practicing psychotherapists define themselves as eclective or integrative [
63].
How the interventions might work
It is a common opinion among clinicians and researchers that patients suffering from complex psychiatric distress require longer and more intensive psychotherapy [
27]. Complex psychiatric distress can be defined as disorders, which by definition are enduring and inflexible [
27], such as personality disorders or schizophrenia, chronic psychiatric disorders (defined as lasting at least a year), or multiple psychiatric disorders. A related assumption is that complex and severe problems typically take longer to improve than less complex or acute psychiatric distress [
25,
64]. This is due to the inherent inflexibility of the psychopathology and the complexity of the required therapeutic techniques. Such potential therapeutic techniques could be provocation of affect or working with the therapeutic alliance [
25]. These are techniques that are potentially hard to carry out when faced with time constraints. However, it is often argued that such techniques are essential to effective psychotherapy [
65].
In contrast, one could argue that long-term therapies can become counterproductive, given that the same therapeutic techniques will be repeated for a long period of time without continuous assessment of their effects. It is possible that given the limited therapeutic time, planned short-term psychotherapy forces both patients and therapists to establish and maintain a focus throughout the treatment process [
66]. Further, issues regarding termination of treatment are particularly important when conducting short-term psychotherapy, where concerns about termination are, almost by definition, always present [
67,
68]. Thus, a possible advantage of short-term therapies is that both therapist and patient are forced to address difficult themes associated with separation and loss from the very beginning instead of postponing them for later.
Why is it important to do this review?
It is essential to investigate the optimal duration of psychotherapy for psychiatric disorders, because of the potential patient and health economic burden from long-term psychotherapy and because of the potential harmful effects of terminating treatment prematurely [
69]. If short-term psychotherapy is the optimal treatment approach, then this could result in a reduction of waitlists and thus a greater access to evidence-based care. On the contrary, if long-term psychotherapy is the most optimal treatment, then it becomes sensible for mental health systems to invest in these treatments, as they would translate into greater health and occupational benefits [
70].
The relationship between the number of sessions (dose) and patient improvement (effect) in psychotherapy has previously been studied with mixed results [
70,
71]. There are studies indicating that increased number of sessions is associated with diminishing results [
72]. There are also studies indicating that the speed of improvement is dependent on patients pretreatment functioning [
73] and that some patients require different dosages to receive the same effect. However, most research on the association between dose and effect is based on uncontrolled studies [
70‐
72,
74,
75] which can only show that patients improve during treatment. Whether this improvement can be attributed to the treatment, can only be established with randomized controlled trials, in which shorter and longer therapies are directly compared. A systematic review of such randomized clinical trials might allow us to assess the safety profile of the different treatment options directly. We are already aware of two randomized clinical trials comparing a short-term and a long-term version of the same psychotherapy type for one or more adult psychiatric disorders [
76,
77]. We have performed a preliminary literature search in the Cochrane Database of Systematic Reviews (search terms, short-term or brief and long-term or standard psychotherapy) for previous systematic reviews comparing a short-term and a long-term version of the same psychotherapy type for one or more adult psychiatric disorders. We identified 1114 hits. From this preliminary literature search, we have only identified one empty systematic review [
78].
The present systematic review aims at forming the basis for evidence-based guideline recommendations for the optimal duration of psychotherapy for adult psychiatric disorders taking bias risk (systematic errors), play of chance (random errors), and certainty of the findings into consideration. The objective of this review will be to assess the beneficial and harmful effects of short-term psychotherapy compared with long-term psychotherapy for adult psychiatric disorders.
Discussion
This protocol aims at comparing the effects of short-term psychotherapy with the effects of long-term psychotherapy for common adult psychiatric disorders to determine the best length of treatment. The outcomes will be quality of life, serious adverse events, symptom severity, suicide or suicide attempts, self-harm, and level of functioning.
This protocol has a number of strengths. The predefined methodology is based on the
Cochrane Handbook for Systematic Reviews of Interventions [
83], the eight-step assessment suggested by Jakobsen et al. [
94], Trial Sequential Analysis [
84], and GRADE assessment [
105‐
107]. Hence, this protocol considers both risks of random errors and risks of systematic errors. Another strength of this protocol is that we pragmatically compare two overall treatment strategies with each other, i.e., the results of this review will potentially reflect the effects of the two strategies in clinical everyday practice.
Our protocol also has some limitations. The primary limitation is the potential for large heterogeneity as a result of including all psychiatric disorders and all types of psychotherapy. Therefore, we may ultimately decide that a meta-analysis is not warranted. Further, psychotherapy always consists of multiple treatment elements and it is likely that different interventions have different effects. Hence, if we show a difference between the compared strategies, it will be difficult to conclude what exactly caused the difference in effect. To minimize this limitation, a number of subgroups are planned, but results of subgroup analyses should always be interpreted with great caution. Another limitation is the large number of comparisons which increase the risk of type 1 error. We have adjusted our thresholds for significance according to the number of primary outcomes, but as mentioned, we have also included multiple subgroup analyses. This large risk of type 1 error will be considered when interpreting the review results. Further, we expect that no trials will have blinded treatment providers and patients. Even though blinding of patients should be relatively easy, blinding of treatment providers is theoretically possible but much more difficult to carry out. Finally, we rely on the trialists defining their compared interventions as short-term and long-term (or similar terminology). Hence, we will not include trials comparing a short-term and a long-term version of the same psychotherapy type, if the trialists did not explicitly define their interventions with such terminology. Using trialists’ definitions of short-term and long-term psychotherapy potentially introduces problems with heterogeneity. However, we believe that our choice of methodology from a pragmatic point of view is the best solution there is. First, trialists often report poorly and often do not themselves use thresholds and important data might be excluded from our review if we demand exact definitions of lengths. Further, we do not expect to include many trials in this systematic review. Hence, relying on trialists definitions of short-term versus long-term psychotherapy may increase the number of trials being eligible for inclusion. Finally, we believe this pragmatic methodology will lead to the inclusion of the most relevant trials.
Publisher’s Note
Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.