Abstract
Objective:
To examine the evidence for adjunctive non-pharmacological interventions in the treatment of mania in an acute inpatient setting.
Method:
A selective review of original and review papers was conducted. The electronic databases PsycINFO and PubMed were searched using the following MeSH terms: mania, mania treatment and inpatient.
Results:
Four studies were identified in the search for non-psychopharmacological interventions for mania that commenced in an inpatient setting: Interpersonal and Social Rhythm Therapy (IPSRT), Group Cognitive Behavioural Therapy (G-CBT), sensory room, and dark room therapy. Only two of these were designed exclusively for patients with bipolar disorder and the other two included these patients in a heterogeneous group of acute psychiatric inpatients.
Conclusions:
Sleep and circadian regulation (Social Rhythm Therapy) that focuses on the establishment and maintenance of regular daily rhythms, particularly in relation to sleep–wake times, meal times and socialization, provides a potentially useful model for managing mania in the inpatient setting. However, there is an urgent need for further research into the effective treatment of mania.
Introduction
Episodes of mania are both very distressing and very disabling for patients with bipolar disorder (BD) and their families. As described by Kay Jamison (2011: 67), there is a:
particular kind of pain, elation, loneliness and terror involved in this kind of madness. At first when I’m high it’s tremendous … Ideas are fast … Like shooting stars you follow until brighter ones appear. All shyness disappears, the right words, the right gestures are suddenly there … Uninteresting people become intensely interesting. Sensuality is pervasive, the desire to seduce and be seduced is irresistible. Your marrow is infused with unbelievable feelings of ease, power, well-being, omnipotence, euphoria … You can do anything … But, somewhere this changes … The fast ideas become too fast and there are far too may … Overwhelming confusion replaces clarity … You stop keeping up with it – memory goes. Infectious humour ceases to amuse. Your friends become frightened … Everything is now against the grain … You are irritable, angry, frightened, uncontrollable, and trapped.
The potentially destructive nature of mania can be regarded as a psychiatric emergency requiring immediate intervention to enable the patient to re-establish control over his/her cognitions, behaviours and relationships. Clinical guidelines from the Royal Australian and New Zealand College of Psychiatrists (2003), the National Institute for Health and Clinical Excellence (2006), and the American Psychiatric Association (2005) recommend the need for hospitalization when the person poses a serious threat of harm to themselves or others; lacks adequate social support or demonstrates severely impaired judgment; has a complicating psychiatric or general medical condition; or has not responded to outpatient treatment.
BD accounts for approximately one quarter of admissions to acute inpatient units (Wheeler et al., 2011). Of these most are for mania and to a lesser extent mixed episodes (Olie and Levy, 2002). The average length of psychiatric inpatient stay for BD differs from country to country. In Sweden, the average length of inpatient stay for mania has been calculated at 29.2 days (Osby et al., 2009), Canada 72 days (Canadian Institute for Health Information, 2005), Australia 49 days, Germany 40 days and USA 15 days (Auffarth et al., 2008). It has been estimated that 44% of the hospitalization costs attributable to BD were due to patients suffering from mania or hypomania and 16% to those suffering from mania with psychotic symptoms (Gupta and Guest, 2002).
Although medication is recommended as the first-line treatment of mania, there are some concerns about the effectiveness of currently recommended medications. Nivoli et al. (2011) conducted a critical review of published guidelines and identified that there were problems with the generalizability of most pharmacological studies related to recommended medications. Mania in clinical practice does not have homogenous characteristics but is often complicated by factors such as depressive symptoms (Martin-Carrusco et al., 2012), anxiety (Gonzales-Pinto et al., 2012) and substance abuse (Goldberg et al., 1999; Hoblyn et al., 2009). A meta-analysis of the efficacy of 56 drug–placebo comparisons of 17 anti-manic agents (Yildiz et al., 2011) found that the pooled response rate was 48%, which was 17% more effective than placebo. Therefore, pharmacological anti-manic agents have a relatively modest response rate in the short duration of standard trials. Although medications are likely to remain first-line treatment for inpatient mania, this modest effect, and the possibility that the effect may be delayed, suggests the need to examine adjunctive treatments to improve this.
Inpatient care
Although the guidelines all focus on biomedical treatment, the ‘inpatient milieu is not an incidental backdrop to patients’ hospitalization experience but a powerful and integral aspect of the experience of illness and recovery’ (Thibeault et al., 2010). Acute hospital psychiatric care remains an integral component of treatment for BD despite the increased emphasis on community-based mental health care (Wheeler et al., 2011). Studies have identified that 84% of admissions to acute inpatient units were for risk containment but that the length of stay was often determined by a lack of sub-acute care including day-care, respite, outreach teams and supervised accommodation. Despite its integral role in the process of treatment for severe mental disorders, there is a sparse body of research into acute inpatient psychiatry (Bowers, Chaplin, Quirk, & Lelliot, 2009).
Bowers et al. (2005) have suggested that the objectives of acute inpatient services include the provision of containment to keep people safe, the presence of staff spending time with service users, using interpersonal skills to build relationships, and treatment provision using an array of psychotherapeutic approaches. However, many service users have reported experiencing acute inpatient care as coercive (Husum et al., 2011; Newtown-Howes and Mullen, 2011), which was associated with feeling dehumanized and unheard, and has important implications for treatment outcomes.
Mania is associated with marked behavioural and cognitive disturbance, with psychosis occurring in approximately 60% of acute episodes (Malhi et al., 2009). Of patients admitted to inpatient units with psychosis, 50% meet criteria for post-traumatic stress disorder in the immediate aftermath and for 25% of these this is related to the hospital experience itself (Bendall et al., 2006). What happens in the inpatient setting may play a direct role in the treatment of mania – it may directly impact on the course of illness. Because of the issues associated with the phenomenology of mania, the efficacy of available anti-mania medications, and the inpatient context, it is important to explore non-pharmacological inpatient interventions that provide therapeutic containment for the patient while aiming to modify the course of illness.
Method
The aim of this paper is to examine the evidence for non-pharmacological interventions in the treatment of mania in an acute inpatient setting and to identify clinical and research implications.
A selective review of original and review papers was conducted. The electronic databases PsycINFO and PubMed were searched using the following MeSH terms: mania, mania treatment and inpatient. Papers were excluded if they were solely focused on psychopharmacology.
Results
Only four studies were identified in the search for non-psychopharmacological interventions for mania commenced in an inpatient setting: Interpersonal and Social Rhythm Therapy (IPSRT), Group Cognitive Behavioural Therapy (G-CBT), sensory room, and dark room therapy. Only two of these were designed exclusively for patients with BD and the other two included these patients in a heterogeneous group of acute psychiatric inpatients. Three of the studies were pilot studies.
The only large-scale study (Frank et al., 2005) included 175 participants in both inpatient and outpatient settings who had a score of 15 or greater on the Hamilton Depression Rating Scale or a score of 15 or greater on the Bech–Rafaelsen Mania Scale and were recruited into a trial IPSRT versus Intensive Case Management (ICM). Seventy-seven patients were experiencing a manic episode on entry to the study, but it is not clear how many were inpatients at the time. Acutely ill patients were randomly assigned to one of four treatment strategies: acute and maintenance IPSRT; acute and maintenance ICM; acute IPSRT followed by maintenance ICM; or acute ICM followed by maintenance IPSRT. All therapies were additive to pharmacological intervention per study protocol. Participants assigned to IPSRT in the acute phase survived significantly longer without a new mood episode in the maintenance phase. The authors suggested that this effect appeared to be mediated by the substantially increased regularity of social routines among subjects receiving IPSRT. They found that increasing social rhythm stability during acute treatment was associated with reduced risk of recurrence regardless of the treatment received in the maintenance phase. There was no evidence that either psychotherapy facilitated more rapid recovery from mixed or pure manic states or that later stability varied across therapies depending on index episode.
A pilot, open-label study of G-CBT (Raune and Daddi, 2011) assessed the feasibility of delivering stand-alone CBT sessions in a group format in an acute inpatient setting. Patients chose one or two problems as the target for each group that was conducted using CBT principles. Outcomes measures focused on attendance and patient feedback. The study recruited 137 patients to attend a total of 291 times across 31 groups. Ten per cent of participants had a diagnosis of BD but the paper does not identify how many patients with BD may have been categorized as being amongst the psychotic group (61%). Participants diagnosed with BD without psychotic symptoms were significantly more likely to re-attend the group than participants without this diagnosis. The one-off sessions covered a wide variety of topics including emotional management, depression, reducing self-harm, coping with family problems, stress management, anger management, and improving sleep.
One pilot study examined the use of a sensory room in an acute inpatient psychiatric unit (Novak et al., 2012). The ‘sensory room’ was a specialized room with various items to relax and perform self-soothing routines. It included a ‘homely environment with scenic pictures, comfortable furnishings and a range of sensory modulation items’. The room was made available to patients in a 40-bed acute psychiatric unit in order to reduce distress, reduce disruptive behaviour and reduce rates of seclusion and aggression. There were 75 occasions of use with a mean duration of 39 minutes (range: 10–150 minutes). Approximately a quarter of the usage was by patients experiencing mania or BD. The most common self-soothing activities used were: use of a weighted blanket, listening to music, reading, and a rocking chair. The use of the sensory room was found to reduce self-rated distress and the following clinician-rated behaviours: pacing, loudness, irritability, intrusiveness, elevation and anxiety. There was no significant reduction in physical aggression and paranoia. There were also no significant changes in seclusion use.
The only study that was specifically focused on the treatment of mania in an inpatient setting was a pilot study of dark therapy (Barbini et al., 2005). This involved a light-sealed room in which patients experiencing mania spent 14 hours per day (6 p.m. to 8 a.m.) for 3 days. Thirty-two patients experiencing mania were recruited to either treatment-as-usual or dark room therapy plus treatment-as-usual. The two groups were matched for age, sex, age at onset, number of previous episodes and duration of current episode. There was a significant interaction between treatment and duration of mania whereby for those patients who had less than 2 weeks of mania, the dark therapy reduced their mania rating scores by 50% within 3 days. This group was also discharged from hospital 8.9 days earlier and had lower anti-mania drug use than those receiving treatment as usual. Those who had been in episode for longer than 2 weeks had no differences in improvement from the treatment-as-usual group. Patients treated with dark therapy also slept more than those who received treatment-as-usual. The authors propose a relationship between biological rhythms and severity of mania at the beginning of the manic episode. When patients were treated within 2 weeks from the onset of the current manic episode the dark therapy stabilized sleep–wake rhythms and thus avoided the vicious cycle in which sleep loss triggers mania and mania disrupts sleep.
All four studies were designed as adjunctive to treatment-as-usual and all the patients continued on their prescribed medication regime. The only study to demonstrate efficacy was dark therapy. This approach is underpinned by chronobiological understandings of circadian rhythms and their effects on mood disorders. Although small in scale, it was the only adjunctive intervention that apparently accelerated treatment response.
Discussion
The treatment of mania, particularly in an inpatient setting, is a relatively under-developed area of psychiatry. Given the relatively modest efficacy of anti-mania medications, it is somewhat surprising that only four interventions aimed at promoting a reduction in mania symptoms were identified in the review. The dark therapy study showed the most promise but may not be practical in current publicly funded inpatient units because of resource constraints. It may be more feasible in privately funded mood disorder units. This study focused on stabilization of circadian rhythms and it may be this concept that has potential for further development. It is possible to extrapolate a little further to suggest that regularity of social rhythms may also have some benefits because of their underpinning basis in circadian rhythms.
Although the biological underpinnings of BD are not well understood, there is consensus that instability of circadian rhythms is a significant pathway in the development and course of the condition, and the link between sleep and BD is well established (Murray and Harvey, 2010). A study of how people with BD managed to stay well, found that getting sufficient and regular sleep was the most important strategy for maintaining or re-establishing wellness (Murray et al., 2011). In the dark room therapy, Barbini et al. (2005) suggest that sleep loss during mania acts not only as a triggering factor but also plays a relevant role in reinforcing the mechanisms underlying the production of manic symptoms during the manic phase. Harvey (2008) has suggested that sleep disruption in BD may be due to a weaker coupling of the circadian system to the external environment. They have proposed a bidirectional relationship between daytime affect regulation and night-time sleep. Disturbances in affect regulation during the daytime interfere with sleep and circadian functioning, contributing to emotional dysregulation, thus creating a vicious cycle.
In this context, it is notable that a positive response to melatonin (3 mg per night) has been reported in a small trial to treat resistant circadian sleep alterations in mania and was also found to exert a global therapeutic action on the manic state (Bersani and Garavini, 2000). It has been found that the effects of melatonin are more effective if the patient is recumbent (Cagnacci et al., 1997).
Given the difficulty of imposing complete darkness in people with mania (Phelps, 2008), an alternative approach may involve the use of dim light. There is some evidence that light of an intensity higher than 500 lux suppresses melatonin secretion in controls but this was greater in patients with BD (Lewy et al., 1987). There is mounting evidence that excessive light-induced suppression of melatonin may be a risk factor for a mood episode (Srinivasan et al., 2006). Phelps (2008) suggests encouraging melatonin production using amber lenses to block blue light which stimulates a retinal photoreceptor connected to the biological clock region of the hypothalamus.
Clinical implications
Sleep and circadian regulation (Social Rhythm Therapy) focusing on maintaining regular daily rhythms, particularly in relation to sleep–wake times, meal times and socialization, provides a potentially useful model for managing mania in the inpatient setting. Specific aspects of this which may be practical and which should be researched further include the following.
Setting a regular time to wake and providing the patient with the opportunity for exposure to daylight on waking. This is crucial for re-establishing circadian rhythms as light is the primary zeitgeber (synchronizing agent) for the biological clock in the suprachiasmatic nucleus (SCN), via specialized circadian photoreceptors in the retina (Wirz-Justice et al., 2013). Current inpatient management of patients suffering from mania may, in fact, allow patients to sleep for longer in the morning on the basis that they may be lacking sleep and that they are at least not disturbed while asleep.
Ensuring the patient has regular meals, particularly breakfast. Food is required to initiate the rhythm for hormone release activated by the peripheral clock in the liver. This peripheral clock helps synchronize the day/night cycle of the SCN (Wirz-Justice et al., 2013).
While the patient is experiencing mania they continue to require small amounts of regular socialization. Regular social zeitgebers can act directly or indirectly on the SCN (Wirz-Justice et al., 2013). Frank et al. (2000) have hypothesized that disruptions in social cues that entrain circadian rhythm may serve as a link between biological and psychosocial processes. They have argued that losing a social zeitgeber could dysregulate biological rhythms.
The patient’s bedroom should only be used for sleep (stimulus control therapy). This is based on the concept of eliminating distractions from the bedroom and associating the bedroom with sleep only. While this intervention has good effectiveness for primary insomnia there is some evidence of its effectiveness in psychiatric conditions (Morin et al., 2006). A comparative review of pharmacotherapy and psychological/behavioural treatments has found comparative overall effect sizes for a range of sleep variables (Riemann and Perlis, 2009).
Patients could be set a minimum time of 6.5 hours in bed and this environment could include a dark therapy component (e.g. lighting dimmed, avoidance of blue light, environment as quiet as possible) (Harvey, 2008). This is based on the bi-directional understanding of sleep and mood (Talbot et al., 2012) and seeks to manipulate sleep factors in order to affect mood.
It may also be useful to teach the patient progressive muscle relaxation that involves learning to monitor tension in each specific muscle group in the body by deliberately inducing tension in each group. This tension is then released, with attention paid to the contrast between tension and relaxation. This has been found to be an empirically supported treatment for managing primary insomnia (Morin et al., 2006) and may have value in mania.
There is also considerable evidence for the use of melatonin to re-establish delayed sleep phase (van Geijlwijk et al., 2010). This would be administered when the patient commences their time in bed, in a dim environment and with the patient semi-recumbent (Cagnacci et al., 1997).
Research methodology
Two research issues are particularly important in this type of research. First, the design needs to take into consideration any likely Hawthorne effect that may impact on pre and post-testing of interventions. Any pre versus post-intervention design is highly likely to show a reduction in admission length or time to response based on increased structure and enthusiasm of nursing and medical staff. In general, a randomized design would be preferable. However, the second major problem in this case is likely to be contamination. This may be more of an issue if patients are randomized within units in a hospital. Patients on the same ward and receiving placebo (no additional intervention) may still benefit from the intervention. For example, increased routine regarding waking and meals might induce a general change which will occur for ‘placebo’ patients even if to a lesser extent than active patients. Treatment with dim light may be used (albeit less rigorously) in placebo patients in a dim light intervention. This issue may be less problematic if there is more than one inpatient unit and the randomization is at unit level. Problems in this case could be, however: (a) demoralization in the unit or units randomized to no change versus a Hawthorne effect in the other units; (b) contamination across units; (c) a tendency for there to be slightly different clinical characteristics across units or for clinicians deliberately to engineer admission for certain patients to the experimental unit.
A further way of addressing these problems would be to have a more active control which was, however, hypothesized to be unlikely to have an active effect other than by increased staff input. Care also needs to be given to ensuring that ratings are blind to treatment. The interventions would need to be both practical and feasible in often under-resourced inpatient settings.
Conclusion
This review suggests that there is limited evidence for the effectiveness of adjunctive treatments for mania, but given the personal, interpersonal, social and economic impacts of mania and the low response rates to medication there is an urgent need for further research. The design of such studies needs careful deliberation because unless they are conducted with real-life clinical populations in real-life clinical settings they are likely to have limited clinical utility. Interventions that incorporate sleep and circadian regulation need further investigation.
Despite these possible design difficulties, people with BD and their families deserve treatments that can moderate the destructive impact of mania by providing therapeutic containment while modifying the course of illness and reducing its duration.
Footnotes
Funding
This research received no specific grant from any funding agency in the public, commercial, or not-for-profit sectors.
Declaration of interest
The authors report no conflicts of interest. The authors alone are responsible for the content and writing of the paper.