Shifting effects in randomised controlled trials of complex interventions: a new kind of performance bias?

OBJECTIVE: Randomised controlled trials (RCTs) aim to provide unbiased estimates of treatment effects. However, the process of implementing trial procedures may have an impact on the performance of complex interventions that rely strongly on the intuition and confidence of therapists. We aimed to examine whether shifting effects over the recruitment period can be observed that might indicate such impact. METHOD: Three RCTs investigating music therapy vs. standard care were included. The intervention was performed by experienced therapists and based on established methods. We examined outcomes of participants graphically, analysed cumulative effects and tested for differences between first vs. later participants. We tested for potential confounding population shifts through multiple regression models. RESULTS: Cumulative differences suggested trends over the recruitment period. Effect sizes tended to be less favourable among the first participants than later participants. In one study, effects even changed direction. Age, gender and baseline severity did not account for these shifting effects. CONCLUSION: Some trials of complex interventions have shifting effects over the recruitment period that cannot be explained by therapist experience or shifting demographics. Replication and further research should aim to find out which interventions and trial designs are most vulnerable to this new kind of performance bias.

Group art therapy as an adjunctive treatment for people with schizophrenia: a randomised controlled trial (MATISSE).

OBJECTIVE: To examine the clinical effectiveness and cost-effectiveness of referral to group art therapy plus standard care, compared with referral to an activity group plus standard care and standard care alone, among people with schizophrenia. DESIGN: A three-arm, parallel group, single-blind, pragmatic, randomised controlled trial. Participants were randomised via an independent and remote telephone randomisation service using permuted blocks, stratified by study centre. SETTING: Study participants were recruited from secondary care mental health and social services in four UK centres. PARTICIPANTS: Potential participants were aged 18 years or over, had a clinical diagnosis of schizophrenia, confirmed by an examination of case notes, and provided written informed consent. We excluded those who were unable to speak sufficient English to complete the baseline assessment, those with severe cognitive impairment and those already receiving arts therapy. INTERVENTIONS: Group art therapy was delivered by registered art therapists according to nationally agreed standards. Groups had up to eight members, lasted for 90 minutes and ran for 12 months. Members were given access to a range of art materials and encouraged to use these to express themselves freely. Activity groups were designed to control for the non-specific effects of group art therapy. Group facilitators offered various activities and encouraged participants to collectively select those they wanted to pursue. Standard care involved follow-up from secondary care mental health services and the option of referral to other services, except arts therapies, as required. MAIN OUTCOME MEASURES: Our co-primary outcomes were global functioning (measured using the Global Assessment of Functioning Scale - GAF) and mental health symptoms (measured using the Positive and Negative Syndrome Scale - PANSS) at 24 months. The main secondary outcomes were level of group attendance, social functioning, well-being, health-related quality of life, service utilisation and other costs measured 12 and 24 months after randomisation. RESULTS: Four hundred and seventeen people were recruited, of whom 355 (85%) were followed up at 2 years. Eighty-six (61%) of those randomised to art therapy and 73 (52%) of those randomised to activity groups attended at least one group. No differences in primary outcomes were found between the three study arms. The adjusted mean difference between art therapy and standard care at 24 months was -0.9 [95% confidence interval (CI) -3.8 to 2.1] on the GAF Scale and 0.7 (95% CI -3.1 to 4.6) on the PANSS Scale. Differences in secondary outcomes were not found, except that those referred to an activity group had fewer positive symptoms of schizophrenia at 24 months than those randomised to art therapy. Secondary analysis indicated that attendance at art therapy groups was not associated with improvements in global functioning or mental health. Although the total cost of the art therapy group was lower than the cost of the two comparison groups, referral to group art therapy did not appear to provide a cost-effective use of resources. CONCLUSIONS: Referring people with established schizophrenia to group art therapy as delivered in this randomised trial does not appear to improve global functioning or mental health of patients or provide a more cost-effective use of resources than standard care alone. TRIAL REGISTRATION: Current Controlled Trials ISRCTN 46150447. FUNDING: This project was funded by the NIHR Health Technology Assessment programme and will be published in full in Health Technology Assessment; Vol. 16, No. 8. See the HTA programme website for further project information.

Music therapy for depression.

BACKGROUND: Depression is a highly prevalent disorder associated with reduced social functioning, impaired quality of life, and increased mortality. Music therapy has been used in the treatment of a variety of mental disorders, but its impact on those with depression is unclear. OBJECTIVES: To examine the efficacy of music therapy with standard care compared to standard care alone among people with depression and to compare the effects of music therapy for people with depression against other psychological or pharmacological therapies. SEARCH STRATEGY: CCDANCTR-Studies and CCDANCTR-References were searched on 7/11/2007, MEDLINE, PsycINFO, EMBASE, PsycLit, PSYindex, and other relevant sites were searched in November 2006. Reference lists of retrieved articles were hand searched, as well as specialist music and arts therapies journals. SELECTION CRITERIA: All randomised controlled trials comparing music therapy with standard care or other interventions for depression. DATA COLLECTION AND ANALYSIS: Data on participants, interventions and outcomes were extracted and entered onto a database independently by two review authors. The methodological quality of each study was also assessed independently by two review authors. The primary outcome was reduction in symptoms of depression, based on a continuous scale. MAIN RESULTS: Five studies met the inclusion criteria of the review. Marked variations in the interventions offered and the populations studied meant that meta-analysis was not appropriate. Four of the five studies individually reported greater reduction in symptoms of depression among those randomised to music therapy than to those in standard care conditions. The fifth study, in which music therapy was used as an active control treatment, reported no significant change in mental state for music therapy compared with standard care. Dropout rates from music therapy conditions appeared to be low in all studies. AUTHORS' CONCLUSIONS: Findings from individual randomised trials suggest that music therapy is accepted by people with depression and is associated with improvements in mood. However, the small number and low methodological quality of studies mean that it is not possible to be confident about its effectiveness. High quality trials evaluating the effects of music therapy on depression are required.

User involvement in the planning and delivery of mental health services: a cross-sectional survey of service users and providers.

OBJECTIVE: To identify methods for involving service users in the planning and delivery of psychiatric services and factors which may assist and impede this process. METHOD: A cross-sectional postal survey of user groups and providers of psychiatric services throughout Greater London (UK). RESULTS: Seventeen (94%) service providers and 29 (48%) user groups responded to the survey. Service providers employed a wide variety of different methods for involving users but none met national standards for user involvement (UI). Service providers stated that the main obstacle to UI was that users who took part were not representative of local patients. User groups highlighted staff resistance as a major obstacle and 80% stated that they were not satisfied with current arrangements for UI. CONCLUSION: While users and providers of mental health services were able to identify changes resulting from UI the responsiveness of staff and the representativeness of service users may be impeding this process.

Xenopus adenine nucleotide translocase mRNA exhibits specific and dynamic patterns of expression during development.

We report the isolation and characterization of the Xenopus homolog to human T1 ANT (adenine nucleotide translocase). The 1290-nucleotide sequence contains initiation and termination signals, and encodes a conceptual protein of 298 amino acids. The sequence shares high amino acid identity with the mammalian adenine translocases. The transcript is present in unfertilized eggs, and it is expressed at higher levels during formation of the antero-posterior dorsal axis in embryos. Although low levels are expressed constitutively except in endodermal cells, adenine nucleotide translocase (ANT) expression is dynamically regulated during neurulation. At this stage, expression in ectoderm rapidly diminishes as the neural folds form, and then ANT expression increases slightly in mesoderm. At the culmination of neurulation, the neural tube briefly expresses ANT, and thereafter its expression predominates in the somitic mesoderm and also the chordoneural hinge. In addition, ANT expression is particularly high in the prosencephalon, the mesencephalon, the branchial arches, eye, and the otic vesicle. Treatment of embryos with retinoic acid has the effect of diminishing constitutive expression of ANT, but microinjection studies demonstrate that immediate and local repression cannot be induced in dorsal structures.

Subcellular localization of acetyl-CoA carboxylase in the apicomplexan parasite Toxoplasma gondii.

Apicomplexan parasites such as Toxoplasma gondii contain a primitive plastid, the apicoplast, whose genome consists of a 35-kb circular DNA related to the plastid DNA of plants. Plants synthesize fatty acids in their plastids. The first committed step in fatty acid synthesis is catalyzed by acetyl-CoA carboxylase (ACC). This enzyme is encoded in the nucleus, synthesized in the cytosol, and transported into the plastid. In the present work, two genes encoding ACC from T. gondii were cloned and the gene structure was determined. Both ORFs encode multidomain proteins, each with an N-terminal extension, compared with the cytosolic ACCs from plants. The N-terminal extension of one isozyme, ACC1, was shown to target green fluorescent protein to the apicoplast of T. gondii. In addition, the apicoplast contains a biotinylated protein, consistent with the assertion that ACC1 is localized there. The second ACC in T. gondii appears to be cytosolic. T. gondii mitochondria also contain a biotinylated protein, probably pyruvate carboxylase. These results confirm the essential nature of the apicoplast and explain the inhibition of parasite growth in cultured cells by herbicides targeting ACC.

Two isoforms of Xenopus retinoic acid receptor gamma 2 (B) exhibit differential expression and sensitivity to retinoic acid during embryogenesis.

We report the isolation of two retinoic acid receptor isoforms (RAR gamma), which differ only in the 5'untranslated and putative N-terminus A regions. The two isoforms appear to serve as early markers for the presumptive neural axis; however, their expression patterns differ. RAR-gamma 2.1 is first expressed at gastrulation at the dorsal lip and subsequently along the presumptive neural axis. RAR- gamma 2.2 represents the full-length sequence of a receptor cDNA already partially characterized and present as a maternal transcript [Ellinger-Ziegelbauer and Dreyer (1991); Genes Dev 5:94-104, (1993): Mech Dev 41:31-46; Pfeffer and DeRobertis, (1994) Mech Dev: 45:147-153]. Unlike RAR-gamma 2.2, the 2.1 variant is not expressed either in pre-somitic mesoderm or notochord. RAR-gamma 2.1 is strongly expressed in branchial arches and to a lesser extent in the neural floor plate. The two isoforms also exhibit differential sensitivity to retinoic acid. Constitutive expression of RAR gamma 2.2 following neurulation appears to be depressed by treatment with retinoic acid, but domains of highest expression, namely, the head and tail, remain relatively unaffected, as do patterns of expression prior to late neurulation. By contrast, RAR-gamma 2.1 is not transcribed in retinoid-inhibited structures. Using microinjection techniques, we show that changes of RAR-gamma 2.1 expression in presumptive head structures occur as an early and local consequence of retinoic acid administration. Since RAR-gamma 2.1 expression is inhibited by retinoic acid, we tested to see if other treatments that perturb axis formation had any effect. Surprisingly, UV irradiation did not suppress that its inhibition by retinoic acid is not due solely to inhibition of anterior neural development. These experiments demonstrate a new subdivision of isoforms that undergo differential expression during development and that exhibit differential sensitivity to retinoic acid and to UV. This sensitivity and the presence of this isoform variant in regions that are known to exhibit polarizing activity strengthen the hypothesis that these receptors play a primary role during morphogenesis.