Structural reform of outpatient psychotherapy

On June 16, 2016, the G-BA created new framework conditions for low-threshold therapy offers with the structural reform of outpatient psychotherapy, which has been in effect since April 1, 2017. The declared aim of the reform is to make psychotherapeutic care more effective, including reducing waiting times. Psychotherapeutic consultations, acute psychotherapeutic treatment and relapse prevention as new elements of care are intended to ensure flexible, low-threshold and timely access to psychotherapy. Acute treatment plays an important role in the reform. It should be carried out as a short-term intervention with a maximum of 24 sessions lasting 25 minutes each (G-BA 2017, p.7). These “new care offers” will diversify the range of services for mentally ill people, i.e. expand the variety of offers in this care area. The G-BA is currently also examining the inclusion of systemic therapy in standard care, for which IQWIG, which was commissioned to do this, published a preliminary benefit assessment on August 16, 2016 (IQWIG 2016), while many health insurance companies are working with specialized providers such as Ge.on Case Management GmbH is already improving the care of mentally ill patients.1 In addition, there are a number of so-called digital mental health programs, some of which have also made the jump to reimbursement by statutory health insurance, for example Deprexis or Novego2, others will follow here. With an increasingly differentiated supply of care, knowledge of the study situation and, as a result, the control and coordination function of providers and health insurance companies is becoming increasingly relevant. Therefore, both the psychotherapeutic consultation hours on the part of the providers and the offering of managed care providers on the part of the consumers will assume a very central control function in the future. As part of this article, Ge.on Case Management GmbH has therefore investigated the question of what benefits of short-term psychotherapeutic interventions have so far been proven in randomized controlled trials (RCT), whether there are any statements about the cost-effectiveness of such procedures and how the care situation in Germany compares internationally represents comparison.

>> A systematic literature search was conducted in MEDLINE for randomized, controlled trials determining the outcome of short-term psychotherapeutic interventions. In addition, the Cochrane Library was searched for systematic literature reviews, a hand search was conducted in Google Scholar, and suitable studies were evaluated in addition to the systematic search. Studies were included that were available in English, in which the short-term psychotherapeutic treatment lasted ≤ 600 minutes, and whose intervention was comparable to the procedures recognized in Germany according to Section 15 of the Psychotherapy Guidelines. Furthermore, only studies in which study participants were 17 years of age or older were considered. The control group in the studies had to be single- or multi-arm and tested against "treatment as usual" (TAU), "placebo" (PL), "waitlist" (WL), or "long-term therapy" (LTP). In this study, LTP refers to therapy lasting longer than 600 minutes and does not correspond to the definition of long-term therapy according to psychotherapy guidelines. The results were evaluated descriptively. A qualitative assessment of the included studies, e.g., using the Cochrane Risk of Bias Tool, was not performed. Since approximately 70% of the studies we examined were from Great Britain or the Netherlands, we expanded the analysis to include a synopsis of the services offered in these countries compared to Germany.
Results
Twenty studies met the inclusion criteria (see Figure 1). Of these, 15 studies investigated cognitive behavioral therapy (CBT), four studies investigated problem-solving therapy (PST), and one study investigated brief psychodynamic therapy (BPP) or brief cognitive behavioral therapy. The following outcome parameters were identified from the 20 studies (with multiple responses): (a) reduction in symptoms of the clinical complaint (N = 20); (b) remission (N = 3); (c) patient satisfaction (N = 2); (d) cost-effectiveness (N = 3).
Regarding (a):
In the studies we examined, improvement in clinical symptoms was measured using generic instruments such as the EQ-5D, SF36, or the Beck Depression Inventory. The results across the studies varied considerably. King et al. demonstrated in two studies (N = 134 and N = 464) for anxiety disorders and depression that cognitive behavioral therapy produced significantly positive results compared to standard treatment by a general practitioner (King et al. 2000, 2014). For post-traumatic stress disorder, Sijbrandij et al. demonstrated in one study (N = 143) that cognitive behavioral therapy had a positive impact on the symptoms of this disorder, but these results were not stable over the long term (defined as a further measurement four months after the intervention). Sijbrandij et al. tested this in comparison to a waiting list, i.e., "doing nothing" (Sijbrandij et al. 2007). Kenardy et al. In their study (N = 186), they also showed positively significant effects for cognitive behavioral therapy compared to the waiting list treatment for panic disorder, but not in comparison to long-term therapy. Here the picture was reversed and short-term therapy even showed negatively significant effects, i.e. it was inferior to long-term therapy (Kenardy et al. 2003). Overall, however, the study confirmed the results of Clark et al. (N = 43), who were also able to show positively significant effects for panic disorder using cognitive behavioral therapy compared to the waiting list. In contrast to Kenardy et al., however, their study did not show any difference between long- and short-term therapy (Clark et al. 1999). Loerch et al. (N = 55) and Sharp et al. (N = 149) also showed positively significant effects for cognitive behavioral therapy in their studies for agoraphobia, a form of panic disorder. Both researchers tested against a placebo treatment in their study designs (Loerch et al. 1999; Sharp et al. 1997). For generalized anxiety disorder and panic disorder, cognitive behavioral therapy also demonstrated a positive effect on symptoms compared to a placebo, according to the studies by Power et al. (N = 101) and Black et al. (N = 75) (Power et al. 1990; Black et al. 1993).
At this point, it can be stated that nine studies found significantly positive results for cognitive behavioral therapy, while one study showed a significant negative result compared to long-term therapy. In addition to the aforementioned study by Clark et al., four other studies found no significant results for cognitive behavioral therapy. Neither Shapiro et al. (N = 104) for depression nor Marchand et al. (N = 100) for agoraphobia and panic disorder were able to identify a significant advantage when comparing short-term and long-term therapy (Shapiro et al. 1995; Marchand et al. 2009). Scott et al. (N = 48) were also unable to identify a significant advantage at the first measurement point; however, unlike Shapiro and Marchand, they tested against standard therapy and not long-term therapy (Scott et al. 1997). Bryant et al. (N = 90) investigated the effect of treatment compared to a waiting list for posttraumatic stress disorder and also found no statistical difference (Bryant et al. 2008).
From these results, it can be concluded that cognitive behavioral therapy offers benefits, measured in improved clinical symptoms, compared to the alternative of "no treatment" (e.g., defined by a waiting list and a placebo, but also by nonspecific medical treatment, i.e., standard treatment). Compared to long-term therapy, these benefits no longer appear to exist, which generally presents a challenge for health insurance providers when long-term therapy follows short-term therapy (add-on therapy) rather than replacing it.
A similar picture emerges for problem-solving therapy, which was investigated in four studies. Neither Schreuders et al. (N = 175) nor Kendrick et al. (N = 247) were able to measure a significant advantage of the short-term intervention over standard treatment by the general practitioner (Schreuders et al. 2007; Kendrick et al. 2005). Problem-solving therapy was tested against a placebo by Mynors-Wallis et al. (N = 91) and Barrett (N = 241), with the former finding a statistically significant positive advantage of the short-term intervention.
intervention in the treatment of depression and the latter, like the first two authors, were also unable to demonstrate any significant benefit of the short-term intervention in the treatment of dysthymia and mild depression (Mynors-Wallis et al. 1995; Barrett 2001).
Shulamit et al. conducted a study (N = 91) with a short-term psychodynamic therapy and could not demonstrate a statistically significant positive advantage of the short-term intervention over long-term therapy (Shulamit et al. 2012).
The conclusion for the first outcome parameter, improvement in clinical symptoms, is therefore generally indifferent. A total of 10 studies showed positive and one study a negative statistically significant result, while 9 studies showed no difference between the short-term therapy and the respective comparator therapy. Therefore, it cannot be said that the current study data demonstrate a clear benefit from the use of short-term interventions. This statement is even more true when comparing the short-term intervention with existing services for the mentally ill and even less so when comparing it with the alternative of "doing nothing."
Regarding (b):
Three studies considered remission as an outcome parameter, which from the perspective of health insurance companies represents a more relevant measurement than the generic instruments mentioned, since remission would be expected to have a lasting effect and thus lower service utilization. Two studies, Campbell-Sills et al. (N = 227) and Scott et al. (N = 48), showed a positively significant effect of cognitive behavioral therapy compared to standard treatment (Campbell-Sills et al. 2016; Scott et al. 1997). While Scott et al. included patients with depression in the study, the Campbell-Sills study also examined patients with post-traumatic stress disorder, panic disorder, and anxiety disorders. The Barrett study (N = 241) was able to demonstrate a statistically significant difference in an intergroup comparison with the drug paroxetine compared to placebo in the treatment of dysthymia. However, it remained unclear whether the effect of problem-solving therapy, considered on its own, also reached statistical significance. No statistically significant results were found for mild depression (Barrett 2001). The results regarding the outcome parameter confirm the results for the outcome parameter clinical symptoms: Brief cognitive behavioral therapy shows significantly positive results compared to nonspecific treatment, e.g., by a general practitioner. This was not demonstrated to the same extent for problem-solving therapy.

Regarding (c):
In two studies, patient satisfaction was also assessed as an outcome parameter in addition to clinical outcomes. Positively significant satisfaction scores were measured for both cognitive behavioral therapy and problem-solving therapy compared to standard treatment (King et al. 2000; Kendrick et al. 2005). This result is hardly surprising, given that patients received special attention compared to the alternative, standard treatment.

Regarding (d):
Hakkaart-van Roijen et al. (N = 702) and Kendrick et al. (N = 247) examined not only the clinical effects of a short-term intervention but also its cost-effectiveness (Hakkaart-van Roijen et al. 2006; Kendrick et al. 2005). Some of the other studies also made statements about cost variables, but did not relate them to a natural unit or benefit criterion, so only the two studies mentioned were able to make actual statements about cost-effectiveness. Hakkaart-van Roijen et al. were unable to demonstrate a statistically significant difference between cognitive behavioral therapy and either standard treatment or long-term therapy. The same applies to problem-solving therapy, where Kendrick et al. were unable to demonstrate a significant difference compared to standard treatment. These results illustrate that short-term therapy cannot be expected to contribute to reducing the overall cost burden for statutory health insurance. This would have been conceivable if the short-term therapy had a substitution character or demonstrably contributed to the avoidance of subsequent costs.

If we now summarize the results of the outcome parameters (a-c) and compare them with the statement on cost-effectiveness (d), the studies provide no reason to assume that the expansion of care offerings will result in a better cost-benefit ratio for health insurance companies. Rather, with a greater variety of therapies, it seems more important than ever to guide the right patients to the right therapy, and to the appropriate extent. Within the framework of the newly introduced psychotherapy consultation hours, this task will be assumed more extensively by psychotherapists than before.
The individual study results just described are summarized in Figure 2.
In addition to the available scientific evidence, we also examined the extent to which short-term interventions abroad are already covered by general health insurance. Since two-thirds of the studies we examined originate from Great Britain or the Netherlands, and the other studies were conducted outside the EU, we compared these countries. We compared these two countries with the German health care system in terms of therapy duration, types of therapy offered, indications, and forms of financing (see Figure 3). This shows that therapy types and duration are more differentiated in other European countries than in Germany, and that a larger variety of low-threshold counseling options, such as computer-assisted cognitive behavioral therapy, are offered there. At the same time, the scope of services covered by compulsory health insurance is significantly lower than in Germany. With the reform introduced, Germany is therefore moving closer to the therapy options offered abroad in terms of therapy types. An expansion of the range of therapeutic services will likely increase the number of patients to be treated and thus also the overall budget for psychotherapeutic services, especially since these are paid outside of the morbidity-related total remuneration.
discussion
In the so-called Psychotherapy Guidelines, the Federal Joint Committee (G-BA) specifies which psychotherapy methods are covered by statutory health insurance. In Germany, these currently include behavioral therapy, depth psychology-based psychotherapy, and analytical psychotherapy. With the structural reform of outpatient psychotherapy and the resolutions of June 16 and November 24, 2016, the access requirements for these psychotherapy methods were expanded to include psychotherapy consultations, acute psychotherapy treatment, and relapse prevention. The goal of acute psychotherapeutic treatment is "...to relieve patients of acute symptoms using outpatient psychotherapeutic means. Acute psychotherapeutic treatment is aimed at a short-term improvement in the patient's symptoms. It does not aim to comprehensively address the underlying etiopathogenetic factors of the mental illness, but rather serves to improve acute mental crises and exceptional situations. Patients for whom acute treatment is not sufficient should be stabilized so that they are prepared for treatment according to Section 15..." (G-BA 2017, p. 7). Acute psychotherapeutic treatment is thus a classic short-term intervention, as it is also known abroad. The study of the scientific evidence regarding the benefits of short-term psychotherapeutic interventions was the subject of this study.
However, based on the current study situation, it is currently not possible to say with certainty whether these short-term interventions generate any benefits, particularly compared to existing therapies. This will not be possible to clarify with certainty in everyday German care, as Section 13, Paragraph 4 of the Psychotherapy Directive stipulates that in the event of a need for further treatment, the hours of acute treatment provided are credited towards the therapy quota pursuant to Section 28 (short-term or long-term therapy). The therapy offered will therefore transition from one care option to the other, and a possible benefit for the patient could then be seen in the duration of treatment rather than in early access. In contrast to the outcome parameters, however, a clear picture emerges with regard to cost-effectiveness based on the study situation, as no RCT has yet proven that short-term interventions are cost-effective. Accordingly, short-term therapies do not appear to be a substitute, but rather, if treatment is required, are preceded by follow-up therapies in the sense of add-on therapy. The planned evaluation of the new care offer by the G-BA will therefore face the challenge of separating the individual interventions (e.g. acute treatment from short-term therapy) in terms of their effect and in considering their cost-effectiveness.
At this point, the success of the newly introduced form of care will also have to be measured in practice: If, on the one hand, patients who would previously have remained untreated are referred to a form of therapy through acute treatment, then positive effects for acute treatment can be inferred from the studies presented here. On the other hand, if patients are referred to acute treatment who previously could have immediately accessed short- or long-term therapy, then acute treatment is unlikely to offer any added benefit. For this reason, the psychotherapeutic consultation will also play a key steering function, and one would therefore expect that patients with mild symptoms in particular will be referred to the new acute treatment.
Overall, it could therefore be concluded for the statutory health insurance system that, with an overall increase in the number of patients treated, indifferent evidence of benefit and a lack of evidence of cost-effectiveness, expenditure in account group 40 is likely to increase.
In addition to these newly introduced forms of therapy, other forms of therapy will also become established in the healthcare landscape. These include systemic therapy, but also the increasingly extensive range of digital mental health programs. It remains to be seen whether psychotherapy consultations can and will regulate access to these services, or whether this will be controlled more by health insurance companies and their managed care providers. Overall, broader access to care is a positive development for patients. In the context of this broader access to care, the ability to select the right patient for the right type of care will become increasingly important.
limitations
The biggest limitation of this study is that the identified studies were not subjected to a quality assessment (quality criteria), e.g., using the Cochrane risk of bias tool. Therefore, the studies were only subjected to a descriptive evaluation, i.e., vote counting. Thus, all studies were included in this evaluation with equal value, although they do not necessarily have to be of equal value. A high-quality study may therefore be offset by a lower-quality study in the evaluation. Nevertheless, the evaluation conducted here provides a very good overview of the existing evidence for short-term interventions and can thus serve as a basis for the evaluation of short-term therapy in Germany.