Skip to main content
  • Research article
  • Open access
  • Published:

Randomised control trial of the effectiveness of an integrated psychosocial health promotion intervention aimed at improving health and reducing substance use in established psychosis (IMPaCT)



People with psychosis have a reduced life expectancy of 10–20 years, largely due to cardiovascular disease. This trial aimed to determine the effectiveness of a modular health promotion intervention (IMPaCT Therapy) in improving health and reducing cardiovascular risk in psychosis.


A multicentre, two arm, parallel cluster RCT was conducted across five UK mental health NHS trusts. Community care coordinators (CC) were randomly assigned to training and supervision in delivering IMPaCT Therapy or treatment as usual (TAU) to current patients with psychosis (cluster). The primary outcome was the physical and mental health subscales of the Short form-36 (SF-36) questionnaire.


Of 104 care coordinators recruited, 52 (with 213 patients) were randomised to deliver IMPaCT therapy and 52 (with 193 patients) randomised to TAU. Of 406 patients, 318 (78%) and 301 (74%) attended 12- and 15-month follow-up respectively. IMPaCT therapy showed no significant effect on the physical or mental health component SF-36 scores versus TAU at 12 or 15 months. No effect was observed for cardiovascular risk indicators, except for HDL cholesterol, which improved more with IMPACT therapy than TAU (Treatment effect (95% CI); 0.085 (0.007 to 0.16); p = 0.034). The 22% of patients who received >180 min of IMPACT Therapy in addition to usual care achieved a greater reduction in waist circumference than did controls, which was clinically significant.


Training and supervising community care coordinators to use IMPaCT therapy in patients with psychosis is insufficient to significantly improve physical or mental health quality of life. The search for effective, pragmatic interventions deliverable in health care services continues.

Trial registration

The trial was retrospectively registered with ISRCTN registry on 23/4/2010 at ISRCTN58667926; recruitment started on 01/03/2010 with first randomization on 09.08.2010 ISRCTN58667926.

Peer Review reports


There is a wealth of evidence demonstrating that people with psychosis, including schizophrenia, schizoaffective disorder and bipolar disorder, have considerably worse physical health than the general population [1, 2]. Of particular concern are the high rates of diabetes, metabolic syndrome (MetS) and cardiovascular disease (CVD) [2]. This increased CVD risk profile is the largest risk factor for the reduction in life expectancy of between 10 and 20 years observed in people with psychosis [3, 4]. Furthermore this mortality gap is widening [3, 4].

The underlying causes of the increase in CVD and premature mortality are complex and multi-factorial. Whilst genetic factors and shared pathophysiological mechanisms contribute, unhealthy lifestyles including high rates of smoking, alcohol use disorders, poor diet, high levels of sedentary behaviour and low levels of physical activity are modifiable factors [1, 2]. In the general population, levels of CVD appear to have plateaued, but they are increasing for people with psychosis [5]. This may be in part because general public health initiatives, such as smoking cessation campaigns and calls to increase physical activity, do not effectively reach this population [6].

Our recent [7] large UK prevalence study, conducted as part of the baseline for this trial demonstrated that nearly two-thirds of people with established psychosis had MetS; similar to the 54.8% reported in the second Australian national survey of psychosis [8] with extremely high levels of central obesity. A Spanish study, using a different definition of MetS, found that patients with schizophrenia treated with antipsychotics had similar rates of MetS to people 10 to 15 years older in the general population [9, 10]. Additionally, 28% of the UK sample had “hazardous drinking” (AUDIT score ≥ 8), 23% had a high saturated fat intake [11], and on average participants spent over 8 h of their waking day being sedentary [7]. Of additional concern, 62% (n = 268/432) smoked tobacco, with a mean of 18 cigarettes per day, comparable to the 66.1% reported in the Australian study [7, 8].

A recent meta-analysis [12] of 17 randomised control trials (RCT) demonstrated a modest positive influence of lifestyle interventions on the prevention and reduction of weight gain and cardiometabolic risk in people with psychosis. Exercise [13] and health-mentor delivered nutritional and exercise interventions [14] also display promise. Whilst these results are encouraging, to date the majority of such interventions targeting people with psychosis have been ‘added on’ to standard care or focus on individual risk factors such as weight or BMI. Such interventions may exclude those patients not attending community mental healthcare services regularly or with difficulties in organisation or motivation. Nor do they fully allow for the multiple health risks experienced by this population. To date, few large-scale long-term RCTs have attempted to improve health in its widest sense in people with psychosis, with most concentrating on a single target, usually weight, with interventions largely run by therapists outside of the usual clinical team. Some studies have co-located the work; for instance, Daumit [15] et al. found that offering free exercise classes and healthy meals in community rehabilitation centres successfully reduced weight in people with psychosis, but again these were separate from usual mental health care and the intervention involved a high level of intense input. Moreover, the recent In SHAPE study [16] in the US relied upon attendance at a public fitness club, which can prove a barrier for some people with psychosis without adequate support. A well-designed large RCT in Denmark, running in parallel with IMPaCT, evaluated an intense intervention of add-on multifactorial lifestyle coaching and additional care co-ordination delivered in the patient’s usual healthcare setting but recently reported that this did not reduce cardiovascular risk [17]. Moreover, like much of the other work to date it involved significant additional input, which will have cost implications. The evidence-base therefore is not yet sufficient to plan services. There is a need for effective and cost-effective ways to reduce cardiovascular risk in a way that is accessible to all patients in community mental health services.

We therefore used the emerging knowledge base to design the content of a novel, manualised, modular health promotion intervention (HPI) to manage physical health and substance use disorders in people with psychosis – IMPaCT Therapy [18]. We chose an over-arching primary outcome measure, quality of life, to reflect both mental and physical health states, rather than focus on one specific marker of cardiometabolic risk. We hypothesised that the integration of IMPACT therapy with usual care would be more effective than usual mental health care in improving quality of life as measured by the physical (PCS) and mental (MCS) component of the Short form (SF)-36 scale [19], at 12-months follow-up in people with psychosis. We proposed that this would be sustained 3 months after completion of the formal intervention, at 15-months follow up. We also hypothesised that the addition of IMPaCT therapy to TAU would result in better cardiometabolic outcomes and healthier lifestyle/substance use choices than in those receiving TAU alone.


Development of the Health Promotion Intervention (IMPaCT Therapy)

We developed a novel integrated health promotion intervention (HPI), IMPaCT therapy, drawing on key principles of two existing effective interventions for physical health and substance use; the “Well-being Support Programme” [20] and “Managing Mental Health and Drug Use” [21], adapted for use in routine clinical care and implementation by the patient’s usual care coordinator. In UK secondary care community mental health services, care coordinators are the main point of clinical contact and are from multiple professions; usually mental health nursing but also social work, psychology or occupational therapy. The intervention aimed to be pragmatic enough to be deliverable within the National Health Service (NHS).

IMPaCT therapy was refined following a staged model of behavioural therapies development [22,23,24]. The focus and key principles were guided by current literature regarding effective physical health interventions in mental health populations, which proposed integration of physical and mental health treatment, included Motivational Interviewing and Cognitive Behavioural Therapy approaches, and addressed a broader range of physical health and substance use targets [25,26,27], delivered over 9 months [28, 29].

IMPaCT therapy went through 3 stages of design and analysis: (1) development of therapy and training, and manual writing, incorporating consultation between experts in mental and physical health, substance use and diabetes; (2) piloting, evaluation and refining of the training package with clinicians and (3) Delphi process to reach consensus on the therapy model and manual. The Delphi process comprised an initial consultation followed by 2 rounds of follow-up questionnaire feedback from (i) expert therapist researchers (ii) clinician providers within Community Mental Health Teams(CMHTs) and (iii) psychosis service users.

Specifically, IMPACT Therapy [18] used motivational interviewing (MI) techniques to address lifestyle choices, with modules targeting the key areas of exercise, diet, tobacco smoking, alcohol use, cannabis use, other illegal substances and diabetes (where applicable), plus integration of cognitive behaviour therapy (CBT) skills to support behaviour change and mental health. To support the intervention, we published in book form a manual and a reference guide for clinicians and a handbook for service users [18]. We developed a four-day training programme for practitioners, encompassing skills and knowledge on physical health, substance use, health promotion, running groups, cognitive behavioural therapy and motivational interviewing. This was well attended and well received with an increase in mean self-rated knowledge scores from pre to post training on all core areas of training (physical health, substance use, running groups and using motivational interviewing).

Study design and setting

We used a pragmatic multicentre, two arm, parallel cluster RCT design, integrated within community mental health teams (CMHTs) across five mental health NHS trusts in South London, Kent, Sussex, Somerset and Staffordshire, representing an urban to rural population. The study was planned and implemented in concordance with the Consolidated Standards of Reporting Trials (CONSORT) cluster trial extension standards [30] and details of method, measures, procedure, sample size calculation are described in the published protocol [31]. Ethical approval was obtained from the joint South London and Maudsley and the Institute of Psychiatry NHS Ethics Committee (REC Ref no 09/HO80/41). Colleagues with lived experience, both service users and carers, were involved throughout the research, from applying to funding to managing the steering group, to co-authoring this paper.


Care co-ordinators in participating CMHTs who were permanently employed and had a minimum of four psychosis patients on their on their caseload who were eligible to participate in the study. The participants eligible for inclusion were aged between 18 and 65 years with a diagnosis of a psychotic disorder (ICD 10 diagnosis F20–29, F31.2, F31.5). Patients were excluded if they a) had a primary diagnosis of learning disability, b) had a pre-existing physical health problem that would independently impact on metabolic measures (as judged by medical investigators), c) were currently pregnant or less than 6 months post-partum or d) had a life threatening or terminal medical condition. We did not recruit from first episode psychosis services.

Study procedure

The study procedure is published and available with open access [31], but will briefly be described. Recruitment of participants occurred in two waves; first, eligible community care coordinators were approached in a random sequence and invited to participate. Once a care co-ordinator gave informed consent, the patients from their caseload who met the inclusion criteria were likewise approached in a random order and sequentially invited to participate, until either 4 participants consented, or all eligible patients had been approached. Once baseline assessments were completed on all consenting patients in a care co-ordinator’s caseload, that care coordinator was randomised, stratified by borough (to allow for socio-economic differences between boroughs) using randomisation blocks of random sizes, to delivering IMPaCT Therapy or TAU alone to their own current patients (cluster). Researchers and the statistician remained blind to treatment allocation. Recruitment started on 01/03/2010 and the first date of randomization was 09.08.2010.

In the treatment arm, the intervention (IMPaCT Therapy) was provided by each patient’s usual community care coordinator who, within 3 months of randomisation, received the four-day IMPACT training course. Participating care coordinators were offered fortnightly supervision in IMPaCT Therapy throughout the subsequent 9-month intervention. Additionally, all care coordinators were offered a one-hour training session in best practice for physical health awareness to ensure more standardised treatment as usual (TAU).

Outcome measures

Change in outcome was defined as difference from pre-randomisation (baseline) and i) at completion of the supervised intervention (12 months) and ii) 3 months after the end of treatment (15 months). Time windows to collect this data around the defined time points were established (minus 6 weeks/plus 4 weeks at 12 months and plus/minus 4 weeks for 15 month follow up). Data collected outside these times were recorded but only used for sensitivity analyses and not the main analysis.

Primary outcome

Primary outcomes were the physical and mental health component scores of the SF-36 [19] at 12 and 15 months.

A range of socio-demographic data was collected including age, sex, self-report ethnicity, marital status and current medications. Full details of measures are included in the protocol [31] In brief, secondary outcome measures included:

Physical health measures: Fasting blood samples including total, high (HDL) and low density lipoprotein (LDL) cholesterol, triglycerides, glycated haemoglobin (HBA1c) and C-reactive protein (CRP). Anthropometric measurements (waist circumference, body mass index (BMI), blood pressure) were also taken [31]. The International Diabetes Federation (IDF) criteria for Metabolic syndrome (MetS) were used to define abnormalities [32].

Substance use measures: Alcohol use was recorded using the Alcohol Use Disorders Identification Test (AUDIT) [33], tobacco use with the Nicotine Dependence Questionnaire [34], while use of cannabis and other illegal substances (opiates, methamphetamine, cocaine) was recorded using the Time Line Follow Back [35].

Lifestyle measures: Dietary pattern and physical activity were quantified according to the Dietary Instrument for Nutrition Education (DINE, [11]) and the short form International Physical Activity Questionnaire (IPAQ-SF, [36]) respectively.

Mental Health status: Participants completed the Positive and Negative Syndrome Scale (PANSS [37]), Global Assessment of Functioning (GAF) [38], SF-36 and Montgomery Asberg Depression Rating Scales (MADRS, [39]).

Sample size

The power analyses were performed for the two subscales measures, physical and mental health components of SF-36 Quality of Life scale [19]. A sample size of 70 care coordinators each with an average of 4 patients (inflation factor 1.15 based on an assumed intraclass correlation of 0.05) and thus 280 patients, after allowing for 20% loss of care coordinator and an additional 30% loss of patients to follow up, was needed to detect a reduction of 5 points on both physical (d = 0.5) and mental health scale (d = 0.42) with at least 80% power using an alpha level of 0.05 and two-tailed assumptions.

Statistical analysis

The primary statistical analyses were based on the intention to treatment principle and targeted at estimating the difference in the mean outcomes between participants randomised to HPI and TAU at the two post-treatment observation time points (12 and 15 months) using mixed effects models. Bias due to missing follow-up data was assessed by comparing baseline characteristics of those with and without complete data. In the two linear mixed effects models, physical and mental health component scores, respectively, at 12 months and 15 months constitute the dependent variable. “Treatment randomisation group”, “time (with two levels 12- and 15- months post-randomization)”, the interaction between “treatment group and time”, “centre”, and the “baseline values of physical and mental health component scores, respectively, are the fixed part of the model”. An unstructured covariance pattern model was used to model the dependency of the repeated observations of the same subject while care coordinator was included as a random factor to account for the dependency of the subjects within a cluster. Model assumptions were assessed by visual inspection of the residuals. Standardised effect sizes (using pre-randomisation variability for standardisation) are also reported.

Sensitivity analyses

Because about 15% of the observations were collected outside the time window we repeated the analyses using all available data as a sensitivity analyses. Demographic and clinical baseline characteristics were similar for both arms and we therefore did not perform any sensitivity analyses with baseline covariates [40]. Nor did we pre-specify any subgroup analyses to assess treatment effects to avoid increasing false positive and false negative findings due to inadequate power [41].

Handling of missing data

Models were rerun with predictors related to outcome missingness included as further covariates in the model. For the main outcomes, a second sensitivity analysis of missing outcome data, using multiple imputations by chained regression equations, was performed separately for each treatment group, using all available clinical and demographic scores.

Secondary outcomes

Secondary outcomes were analysed using the same methods as for the primary outcome. However, for all models the interaction between treatment group and time was not significant and was removed from the final analyses. Treatment effects are therefore estimates for both time points. Logistic mixed models were used for binary outcomes (such as smoking) and Poisson mixed models for count data (such as number of cigarettes per day).

Role of the funding source

The funding body were not involved in the design, running or reporting of the trial.


In total, 104 care coordinators were randomised into the IMPACT health promotion intervention (HPI) and control group (TAU). Overall, 406 patients from randomised care coordinators were eligible and consented for the trial. Fifty-two care coordinators (with 213 patients, mean patients per care coordinator: 4.1, SD = 1.6, range 1–10) were randomized into HPI and 52 care coordinators (with 193 patients, mean patients per care coordinator 3.7, SD = 1.4, range 1–6) were randomised into TAU (Fig. 1 Consort diagram). Three hundred eighteen of the 406 patients (78.3%) attended the 12 months–follow-up and 301 (74.1%) attended 15 months follow-up. However, some of the patients who attended follow-up assessments were not seen within the required timeframe (10.5–13 months for 12 months and 14–16 for 15 months), so that the total sample size was reduced to 263 (64.8%) for 12 months (TAU: 132, HPI: 131) and 238 (58.6%) for 15 months (TAU: 114, HPI: 124). Three hundred and fourteen patients (77.3%, HPI: 75.6%, TAU: 79.3%) were seen at least at one time point within the time window and follow-up rates did not differ significantly between arms (12 months: chi2(1) = 1.81, p = 0.18, 15 months: chi2(1) = 1.80, p = 0.18). The required sample size based on the power analyses was achieved at both time points.

Fig. 1
figure 1

Consort diagram

Pairwise comparison of baseline characteristics revealed that subjects who did not attend at 12 months follow-up were significantly younger (41.5 years (SD = 10.6)) than patients who did attend (44.9 years (SD = 9.88), (t(404) = 1.77, p = 0.005) and tended to have smaller baseline waist circumferences (103.6 cm (SD = 19.56) versus 107.6 cm (SD = 17.43)). Non-attendance at 12 months follow-up differed between centres (p = 0.01). No other demographic and clinical baseline characteristics differed significantly between attenders and non-attenders.

There were no significant differences in baseline characteristics between patients who attended within and outside the time windows.

Demographic and clinical characteristics at baseline are presented in Tables 1 and 3. Patients at baseline were on average 45 years old (range 22 to 66), predominantly male (55%) white (55%) and single (65%). Black patients were the main ethnic minority group (34%). The mean SF-36 Mental health score was 43.3 (range 9.6 to 67.7) and the Physical health score was 48.0 (18.8 to 68.3). Demographic and clinical baseline characteristics were similar for both arms except for some small difference in the relative number of patients per centre which was associated with small differences in the number of care coordinator per arm per centre (note; one centre had only one participating care co-ordinator).

Table 1 Demographic characteristics of participants at baseline for each trial arm and all patients combined

Primary outcome analyses

The mixed effect model analyses revealed no significant treatment effect for change in either Physical or Mental health scores between TAU and IMPaCT at either 12 or 15 months (Fig. 2, Table 2). Estimated treatment effect sizes were small (Physical health score, d = −0.17 at 12 months and −0.09 at 15 months, Mental health score, d=: 0.03 at 12 months and −0.05 at 15 months). There were no significant differences in effect between centres. After removing the non-significant interactions between treatment arm and time, patients in the treatment group showed a non-significant decrease of 1.4 (95% CI: -3.2 to 0.41, p = 0.13, d = −0.12) in physical health scores and 0.13 (95% C.I -2.49 to 2.24), p = 0.92, d = −0.01) in mental health scores compared to the patients in the control group after adjusting for baseline differences.

Fig. 2
figure 2

Mean Physical health and mental health component score (95% Confidence interval) at baseline, 12 and 15 months follow-up for Treatment as usual (TAU) and Health intervention programme group (HPI)

Table 2 Results of the mixed effect linear model for main outcomes (physical and mental health score) with treatment arm at 12 months or 15 months

Sensitivity analyses

Including data obtained outside the observation period resulted only in minor changes and did not alter the conclusion. Similarly, including age as a predictor of missingness and a multiple imputation analyses for missing data resulted only in minor changes and did not alter the conclusion that there was no significant treatment effect at any time point (Models with age (raw figures) included: physical health score: d = −1.52 (95% CI -3.3 to 0.26, p = 0.094, Mental health score d = −0.09 (95% C.I. -2.45 to 2.27), p = 0.94; Multiple imputation: Treatment effect Physical health score, d = −0.78 (95% C.I.: -3.76 to 2.20), p = 0.61, mental health score, d = −0.27 (95% C.I.: -3.82 to 0.27), p = 0.90; all analyses without treatment x time interaction).

Secondary outcomes

Table 3 shows estimated treatment effects of the secondary clinical outcomes at 12 and 15 months follow-up. The only significant difference in any of the secondary outcomes at either 12 or 15 months was in HDL Cholesterol, which improved more with IMPACT therapy than in the TAU group (Treatment effect (95% CI); 0.085 (0.007 to 0.16); p = 0.034, Cohen’s d = 0.2). The other effect sizes (Cohen’s d) for continuous outcomes ranged from −0.15 to 0.19. Twenty-eight patients (13.15%) in the HPI group showed a serious adverse event (SAE) while 21 (10.88%) of the patients in TAU showed an SAE. This difference is not significant (chi2(1) = 0.41, p = 0.52).

Table 3 Clinical characteristics of participants at baseline for all participants and changes over time for each arm separately at baseline, 12 months and 15 months follow-up and estimated pooled treatment effects (95% confidence intervals), p value of test and effect size Cohen’s d for continuous outcomes

Care coordinators (CC) allocated to the HPI arm of the trial saw their patients for an average of 8.44 (SD 5.3; range 0–25) sessions over the course of the trial. These sessions lasted on average (50.7 min (SD 27.1). Of this, a median of 20 min was dedicated to the HPI rather than usual care. Minimum adherence to the intervention protocol was defined as at least 6 sessions. In interpreting the data, we later defined the minimum session duration as 30 min in addition to routine care; this was delivered by 9 (17.3%) out of the 52 HPI care coordinators to 19 of their patients. Forty-seven patients (22.0%), under the care of 19 care coordinators (36.5%) in the treatment group, either attended at least six 30-min sessions of IMPaCT Therapy in addition to their usual contact time or had the same therapy duration (180 min) spread over more than 6 sessions. These patients had a non-significant decrease of 2.45 (95% CI: -5.51 to 0.62, d = −0.23, p = 0.12) in physical health scores and 0.84 (95% C.I -4.44 to 2.76), d = −0.02, p = 0.65) in mental health scores compared to the patients in the control group over the 12 months, after adjusting for baseline differences. They had a significantly greater reduction in waist circumference over 12 months than did controls (−4.20 cm (95% C.I. -7.18 to – 1.23), p = 0.006). All other assessed variables remained non-significant, but only continuous outcomes could be assessed due to the smaller sample size.


Statement of principal findings

The current study, the largest pragmatic trial of a MI/CBT HPI intervention in the UK, delivered by care coordinators to their own patients with psychosis and integrated into secondary care practice, found the addition of IMPaCT therapy had no significant effect on our primary outcomes of PCS and MCS quality of life scores. In addition, there was little evidence that IMPaCT therapy delivered by CCs improved cardiovascular risk indicators, substance use or mental health measures compared to TAU alone. The only advantage observed with IMPaCT therapy over TAU at the intention to treat level was in the relatively small effect on HDL cholesterol, which must be interpreted with caution, in the context of multiple testing (see Table 3). Higher HDL levels are linked to cardiovascular health and respond to dietary change and exercise.

Strengths of the study

This is the first RCT investigating an integrated health promotion intervention designed to be used in routine clinical care and implemented by the patient’s usual care coordinator. The main strength of this trial is that it was a pragmatic study set in the NHS, designed to be as representative as possible. The NHS delivers the bulk of healthcare to people with psychosis, largely in secondary care, although if the course of illness is stable, people are discharged to primary care. Community secondary care treatment is led by care co-ordinators based in Community Mental Health Teams (CMHTs) who see their patients in various settings; it was they who were invited to take part in the trial and, if randomised to the treatment arm, deliver the intervention. The HPI was thus accessible to the greatest possible proportion of people with psychosis receiving secondary care, many of whom may not wish, or may be too unwell to attend add-on or group interventions. Thus it attempted to meet the needs of this very hard-to-reach group with high rates of cardiometabolic disease [7] by enhancing routine care. The study recruited to target from a diverse multi-ethnic sample of people with psychosis and had good follow-up rates. There was no difference in dropout between the active and control arms of the study, although those who did not attend at 12 months follow-up were younger. To limit multiple testing and due to the small numbers in some sub-categories we did not do sub-analyses of outcomes by ethnicity, age groups or sex.

The relationship between borough and non-attendance at 12 months follow-up is likely to be multifactorial.

This study attempted to influence lifestyle choices and substance use, which affect both physical and mental health. To date, few large-scale long-term RCTs have attempted to improve health in its widest sense in people with psychosis, with most concentrating on a single target, usually weight, and interventions largely run by therapists outside of the usual clinical team. Even where the work has been co-located, they have been separate to usual care [15, 17]. Therefore, our integrated personalised approach, conducted by the core mental health clinician adapting to the patient’s individual needs, attempted to address some of these concerns by maximising representativeness of the population studied and avoiding a piecemeal approach to behavioural change. Further, our choice of the SF-36 PCS and MCS scores as primary outcome measure reflected our hypothesis that the lifestyle interventions would lead to a general and perceptible improvement in both mental and physical health. Our study, like the recent Change RCT in Denmark [17], demonstrates the challenges of reducing cardiovascular risk among those with established psychosis.

Limitations of the study

There are a number of contextual factors that may have influenced the outcome of the trial. First, this study was funded before the promotion of improved physical health and lifestyle choices among people with psychosis was rising on the national and local NHS agenda. Since that time in the UK alone, the ‘parity of esteem’ movement to improve physical healthcare for people with psychosis was successfully launched by the Royal College of Psychiatrists, NICE published guidance on physical health and the Quality and Outcomes Framework (QOF) offered payment to GPs to monitor and advise on physical health outcomes and lifestyle choices (e.g. encouraging patients to stop smoking). It is likely that participants in both treatment arms benefited from the concerted efforts to improve the very outcomes we were targeting within the study, although we did not demonstrate a significant improvement in either arm over the time frame of the study. Secondly, multiple re-organisations among the participating hospital trusts affected continuity of care. Staff turnover was slightly greater in the treatment than control arms, although this was not statistically different. In this context however, it is interesting to note that where patients remained with the same trained care coordinator throughout the trial, there were indications that the intervention may result in significantly lower cholesterol levels and more exercise in the treatment arm than in controls. Thirdly, given the integration with routine care, it was not possible to control for conditions such as minutes per session. Despite pilot development work and regular supervision being offered by research therapists, care co-ordinators struggled to deliver 6 or more sessions of 30 min each of the HPI in addition to routine care, with HPI sessions lasting a median of 20 min. This highlights the recognised challenges for care co-ordinators delivering targeted psychosocial interventions to people with psychosis in a busy secondary care environment. The proportion of care co-ordinators who delivered the minimum dose (17–36% dependent on the measure taken), was in keeping with Harding et al. (under review [42]) who found that only 25% of care co-ordinators delivered a brief CBT-based therapy despite adequate training and supervision, compared to 62% of assistants without case management responsibilities. This is also consistent with other studies that have found that care co-ordinators struggle to implement therapy alongside case management [43, 44]. Where participants received the intervention of at least 180 min in addition to routine care, the significantly greater reduction in waist circumference than in controls was of a clinically significant degree, suggesting a dose effect once a minimum therapy time was delivered.

A further consideration is that our modular interventions were broad and participants self-selected the target behaviours to be focused on. It is possible that more structured interventions targeted at specific health behaviours, such as those described elsewhere [15, 16, 45] may be better placed to demonstrate statistically significant improvement in defined aspects of the physical health of this population. However, for the purposes of routine care, a more long term, sustainable and integrated approach to global health behaviours is urgently required.

Clinical and service implications

This is the first trial to address the question of whether a health promotion intervention delivered by the core mental health clinician as part of enhanced routine care can achieve greater clinically significant health gains in comparison to treatment as usual. The answer is that this integrated approach did not appear to match the modest health gains (usually measured by weight loss) seen by adding group interventions to usual care – although it would be of interest to compare representativeness of samples between this integrated (and thus very accessible) approach and add-on interventions. Nor was additional time for the intervention regularly delivered when rolled out across community mental health teams. Thus the message to service planners is clear and important – by itself focussed training and supervision provision for front-line mental health workers does not appear to be enough to change therapeutic practice sufficiently to reverse cardiometabolic risk indicators in their patients with established psychosis. However our data suggests that continuity of care and protected time to deliver HPI work may well enhance outcomes, an important practical point in today’s rapidly evolving health services.

A recent qualitative review of reviews [46] considering non pharmacological interventions seeking to reduce cardiometabolic risk concluded that interventions with multiple components, personalized, with more frequent face-to-face contact, and professionally trained treatment providers are associated with better outcomes. Given that our trial showed little effect at the intention to treat level of training and offering supervision to CMHT staff in delivering a HPI, it may be that widening mental health teams to include specialists in exercise and nutrition is necessary, as in the Keeping the Body in Mind preventative programme in Sydney [47]. This will have cost implications for service providers and thus may introduce further barriers to accessing care, pending evaluation of the cost-effectiveness of such models. Our results strongly support the notion that prevention of emergence of cardiometabolic risk factors in early psychosis is key, as reversing these in established illness is extremely difficult [17]. Of note, only people with established illness were included in our study.

Focussing on reducing BMI and waist circumference is notoriously challenging [46]. Future multicomponent interventions might wish to consider cardiorespiratory fitness (the ability of the circulatory and respiratory systems to supply oxygen to working muscles during sustained physical activity) as the primary outcome, since there is mounting evidence that ‘fitness’ is also an important predictor of mortality [48].


The current paper is the first randomised control trial investigating a health promotion intervention designed to be used in routine clinical care and implemented by the patient’s usual care coordinator. Our study found that training and supervising care co-ordinators to work with their own patients with psychosis to change lifestyle choices is not in itself sufficient to effect clinically or statistically significant improvement in quality of life or overall cardiovascular risk in people with established psychosis, although there was an effect on waist circumference when the minimum dosage was delivered. Given the need to reduce the life expectancy gap between people with psychosis and the general population, the search across health systems for more effective, deliverable, affordable and sustainable interventions remains a priority.



Alcohol Use Disorders Identification Test


Body mass index


Cognitive behaviour therapy


Care Coordinator


Community mental health teams


Consolidated Standards of Reporting Trials cluster trial extension standards


C-reactive protein


Cardiovascular disease


Dietary Instrument for Nutrition Education


Global Assessment of Functioning


Glycated haemoglobin


High density lipoprotein


Health Promotion Intervention


International Diabetes Federation


Improving health and reducing substance use in established psychosis


International Physical Activity Questionnaire


Low density lipoprotein


Montgomery Asberg Depression Rating Scales


Mental component of the Short form (SF)-36


Metabolic syndrome


Motivational Interviewing


Positive and Negative Syndrome Scale,


Physical component of the Short form (SF)-36


Randomized controlled trial


Short form −36


Treatment as usual


Treatment as usual


  1. Correll CU, Detraux J, De Lepeleire J, De Hert M. Effects of antipsychotics, antidepressants and mood stabilizers on risk for physical diseases in people with schizophrenia, depression and bipolar disorder. World Psychiatry. 2015;14(2):119–36.

    Article  PubMed  PubMed Central  Google Scholar 

  2. De Hert M, Correll CU, Bobes J, Cetkovich-Bakmas M, Cohen D, Asai I, et al. Physical illness in patients with severe mental disorders. I. Prevalence, impact of medications and disparities in health care. World Psychiatry. 2011;10(1):52–77.

    Article  PubMed  PubMed Central  Google Scholar 

  3. Walker ER, McGee RE, Druss BG. Mortality in Mental Disorders and Global Disease Burden Implications: A Systematic Review and Meta-analysis. JAMA Psychiatry. 2015;72(4):334-41. doi:10.1001/jamapsychiatry.2014.2502.

  4. Lawrence D, Hancock KJ, Kisely S. The gap in life expectancy from preventable physical illness in psychiatric patients in Western Australia: retrospective analysis of population based registers. BMJ (Clinical Research Ed). 2013;346:f2539–f.

    Google Scholar 

  5. Speyer H, Nørgaard HCB, Hjorthøj C, Madsen TA, Drivsholm S, Pisinger C, et al. Protocol for CHANGE: a randomized clinical trial assessing lifestyle coaching plus care coordination versus care coordination alone versus treatment as usual to reduce risks of cardiovascular disease in adults with schizophrenia and abdominal obesity. BMC Psychiatry. 2015;15(1):1–11.

    Article  Google Scholar 

  6. Mitchell AJ, Vancampfort D, De Hert M, Stubbs B. Do people with mental illness receive adequate smoking cessation advice? A systematic review and meta-analysis. Gen Hosp Psychiatry. 2015;37(1):14-23. doi:10.1016/j.genhosppsych.2014.11.006. Epub 2014 Nov 21.

  7. Gardner-Sood P, Lally J, Smith S, Atakan Z, Ismail K, Greenwood KE, et al. Cardiovascular risk factors and metabolic syndrome in people with established psychotic illnesses: baseline data from the IMPaCT randomized controlled trial. Psychol Med. 2015;45(12):2619–29.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  8. Morgan VA, Waterreus A, Jablensky A, Mackinnon A, McGrath JJ, Carr V, Bush R, Castle D, Cohen M, Harvey C, Galletly C, Stain HJ, Neil AL, McGorry P, Hocking B, Shah S, Saw S. People living with psychotic illness in 2010: the second Australian national survey of psychosis. Aust N Z J Psychiatry. 2012;46(8):735–52. doi:10.1177/0004867412449877. Epub 2012 Jun 13.

    Article  PubMed  Google Scholar 

  9. Vancampfort D, Stubbs B, Mitchell AJ, De Hert M, Wampers M, Ward PB, et al. Risk of metabolic syndrome and its components in people with schizophrenia and related psychotic disorders, bipolar disorder and major depressive disorder: a systematic review and meta-analysis. World Psychiatry. 2015;14(3):339–47.

    Article  PubMed  PubMed Central  Google Scholar 

  10. Bobes J, Arango C, Aranda P, Carmena R, Garcia-Garcia M, Rejas J, on behalf of the CLAMORS Study Collaborative Group. Cardiovascular and metabolic risk in outpatients with schizophrenia treated with antipsychotics: results of the CLAMORS study. Schizophr Res. 2007;90(1–3):162–73. Epub 2006 Nov 21

    Article  PubMed  Google Scholar 

  11. Roe L, Strong C, Whiteside C, Neil A, Mant D. Dietary intervention in primary care: validity of the DINE method for diet assessment. Fam Pract. 1994;11(4):375–81.

    Article  CAS  PubMed  Google Scholar 

  12. Caemmerer J, Correll CU, Maayan L. Acute and maintenance effects of non-pharmacologic interventions for antipsychotic associated weight gain and metabolic abnormalities: a meta-analytic comparison of randomized controlled trials. Schizophr Res. 2012;140(1–3):159–68. 10p

    Article  PubMed  Google Scholar 

  13. Firth J, Cotter J, Elliott R, French P, Yung AR. A systematic review and meta-analysis of exercise interventions in schizophrenia patients. Psychol Med. 2015;45(7):1343–61.

    Article  CAS  PubMed  Google Scholar 

  14. Van Citters AD, Pratt SI, Jue K, Williams G, Miller PT, Xie H, et al. A pilot evaluation of the in SHAPE individualized health promotion intervention for adults with mental illness. Community Ment Health J. 2010;46(6):540–52.

    Article  PubMed  Google Scholar 

  15. Daumit GL, Dickerson FB, Wang N-Y, Dalcin A, Jerome GJ, Anderson CAM, et al. A behavioral weight-loss intervention in persons with serious mental illness. N Engl J Med. 2013;368(17):1594–602.

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  16. Bartels SJ, Pratt SI, Aschbrenner KA, Barre LK, Jue K, Wolfe RS, et al. Clinically significant improved fitness and weight loss among overweight persons with serious mental illness. Psychiatr Serv. 2013;64(8):729–36. 8p

    Article  PubMed  PubMed Central  Google Scholar 

  17. Speyer H, Christian Brix Norgaard H, Birk M, Karlsen M, Storch Jakobsen A, Pedersen K, et al. The CHANGE trial: no superiority of lifestyle coaching plus care coordination plus treatment as usual compared to treatment as usual alone in reducing risk of cardiovascular disease in adults with schizophrenia spectrum disorders and abdominal obesity. World Psychiatry. 2016;15(2):155–65.

    Article  PubMed  PubMed Central  Google Scholar 

  18. Greenwood K, Smith S, Atakan Z. Impact - the reference guide: improving physical health and treating substance use in mental illness. 2011.

    Google Scholar 

  19. Ware JE Jr, Sherbourne CD. The MOS 36-item short-form health survey (SF-36). I. Conceptual framework and item selection. Med Care. 1992;30(6):473–83.

    Article  PubMed  Google Scholar 

  20. Smith S, Yeomans D, Bushe CJP, Eriksson C, Harrison T, Holmes R, Mynors-Wallis L, Oatway H, Sullivan G. A well-being programme in severe mental illness. Baseline findings in a UK cohort. Int J Clin Pract. 2007;61(12):1971–8. Eur Psychiatry. 2007 Oct; 22(7):413-8

    Article  CAS  PubMed  PubMed Central  Google Scholar 

  21. James W, Preston NJ, Koh G, Spencer C, Kisely SR, Castle DJ. A group intervention which assists patients with dual diagnosis reduce their drug use: a randomized controlled rial. Psychol Med. 2004 Aug;34(6):983–90.

    Article  CAS  PubMed  Google Scholar 

  22. Rounsaville BJ, Carroll KM, Onken LS. A stage model of behavioral therapies research: getting started and moving on from stage I. Clin Psychol Sci Pract. 2001;8:133–42. doi:10.1093/clipsy.8.2.133.

    Article  Google Scholar 

  23. Medical Research Council. Framework for the development and evaluation of RCTs for complex interventions to improve health. London: MRC; 2000. p. 18.

    Google Scholar 

  24. Medical Research Council. Developing and evaluating complex interventions: new guidance. London: MRC; 2006.

    Google Scholar 

  25. Drake RE, O’Neal EL, Wallach MA. A systematic review of psychosocial research on psychosocial interventions for people with co-occurring severe mental and substance use disorders. J Subst Abus Treat. 2008;34:123–38.

    Article  Google Scholar 

  26. Haddock G, Barrowclough C, Tarrier N, et al. Cognitive –behavioural therapy and motivational intervention for schizophrenia and substance misuse: 18-month outcomes of a randomised controlled trial. Br J Psychiatry. 2003;183:418–26.

    Article  PubMed  Google Scholar 

  27. Baker KD, Lubman DI, Cosgrave EM, Killackey EJ, Yuen HP, Hides L, Baksheev GN, Buckby JA, Yung AR. Impact of co-occurring substance use on 6 month outcomes for young people seeking mental health treatment. Aust N Z J Psychiatry. 2007;41(11):896–902.

    Article  PubMed  Google Scholar 

  28. Barrowclough C, Haddock C, Tarrier N, et al. Randomized controlled trial of motivational interviewing, cognitive behaviour therapy and family intervention for patient with comorbid schizophrenia and substance use disorders. Am J Psychiatr. 2001;158:1706–13.

    Article  CAS  PubMed  Google Scholar 

  29. Barrowclough C, Haddock G, Beardmore R, Conrod P, Craig T, Davies L, Dunn G, Lewis S, Moring J, Tarrier N, Wykes T. Evaluating integrated MI and CBT for people with psychosis and substance misuse: recruitment, retention and sample characteristics of the MIDAS trial. Addict Behav. 2009;34(10):859–66. doi:10.1016/j.addbeh.2009.03.007.

    Article  PubMed  Google Scholar 

  30. Campbell MK, Piaggio G, Elbourne DR, Altman DG. Consort 2010 statement: extension to cluster randomised trials. BMJ (Clinical Research Ed). 2012;345:e5661–e.

    Google Scholar 

  31. Gaughran F, Stahl D, Ismail K, Atakan Z, Lally J, Gardner-Sood P, et al. Improving physical health and reducing substance use in psychosis -- randomised control trial (IMPACT RCT): study protocol for a cluster randomised controlled trial. BMC Psychiatry. 2013;13(1):305–26.

    Article  Google Scholar 

  32. Alberti KGM, Zimmet P, Shaw J. The metabolic syndrome -- a new worldwide definition. Lancet. 2005;366(9491):1059–62.

    Article  PubMed  Google Scholar 

  33. Saunders JB, Aasland OG, Babor TF, de la Fuente JR, Grant M. Development of the alcohol use disorders identification test (AUDIT): WHO collaborative project on early detection of persons with harmful alcohol consumption--II. Addiction. 1993;88(6):791–804.

    Article  CAS  PubMed  Google Scholar 

  34. Fagerstrom KO, Heatherton TF, Kozlowski LT. Nicotine addiction and its assessment. Ear Nose Throat J. 1990;69(11):763–5.

    CAS  PubMed  Google Scholar 

  35. Sobell LC, Sobell MB. Timeline follow-back: a technique for assessing self-reported alcohol consumption. In: Litten RZ, Allen JP, editors. Measuring alcohol consumption: psychosocial and biochemical methods. Totowa: Humana Press; 1992. p. 41–72.

    Chapter  Google Scholar 

  36. Craig CL, Marshall AL, Sjöström M, Bauman AE, Booth ML, Ainsworth BE, et al. International physical activity questionnaire: 12-country reliability and validity… including commentary by Bassett DR Jr. Med Sci Sports Exerc. 2003;35(8):1381–96.

    Article  PubMed  Google Scholar 

  37. Kay SR, Fiszbein A, Opler LA. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr Bull. 1987;13(2):261–76.

    Article  CAS  PubMed  Google Scholar 

  38. American Psychiatric Association. Multiaxial assessment. In Diagnostic and Statistical Manual of Mental Disorders (4th edn, text revised) (DSM–IV–TR). Washington, DC: American Psychiatric Association; 2002.

  39. Montgomery SA, Åsberg M. A new depression scale designed to be sensitive to change. Br J Psychiatry. 1979;134:382–9.

    Article  CAS  PubMed  Google Scholar 

  40. Pocock SJ, Assmann SE, Enos LE, et al. Subgroup analysis, covariate adjustment and baseline comparisons in clinical trial reporting: current practice and problems. Stat Med. 2002;21:2917–30.

    Article  PubMed  Google Scholar 

  41. Burke JF, et al. Three simple rules to ensure reasonably credible subgroup analyses. BMJ. 2015;351:h5651.

    Article  PubMed  PubMed Central  Google Scholar 

  42. Garety PA, Craig TKJ, Iredale C, Basit N, Fornells-Ambrojo M, Halkoree R, Jolley S, Landau S, McCrone P, Tunnard C, Zala D, Waller H. Training the frontline workforce to deliver evidence-based therapy to people with psychosis: challenges in the GOALS study. Psychiatr Serv. In press.

  43. Couldwell A, Stickley T. The thorn course: rhetoric and reality. J Psychiatr Ment Health Nurs. 2007;14:625–34.

    Article  CAS  PubMed  Google Scholar 

  44. Slade M, Bird V, Clarke E, et al. Supporting recovery in patients with psychosis through care by community-based adult mental health teams (REFOCUS): a multisite, cluster, randomised, controlled trial. Lancet. 2015;2:503–14.

    PubMed  Google Scholar 

  45. Bartels SJ, Pratt SI, Aschbrenner KA, Barre LK, Naslund JA, Wolfe R, et al. Pragmatic replication trial of health promotion coaching for obesity in serious mental illness and maintenance of outcomes. Am J Psychiatry. 2015;172(4):344–52.

    Article  PubMed  Google Scholar 

  46. Ward MC, White DT, Druss BG. A meta-review of lifestyle interventions for cardiovascular risk factors in the general medical population: lessons for individuals with serious mental illness. J Clin Psychiatry. 2015;76(4):e477–e86.

    Article  PubMed  Google Scholar 

  47. Curtis J, Watkins A, Rosenbaum S, TEasdale S, Kalucy M, Samaras K, Ward PB. Evaluating an individualized lifestyle and life skills intervention to prevent antipsychotic-induced weight gain in first-episode psychosis. Early Interv Psychiatry. 2015; doi:10.1111/eip.12230.

  48. Lee DC, Sui X, Artero EG, Lee IM, Church TS, McAuley PA, et al. Long-term effects of changes in cardiorespiratory fitness and body mass index on all-cause and cardiovascular disease mortality in men: the aerobics center longitudinal study. Circulation. 2011;124(23):2483–90.

    Article  PubMed  PubMed Central  Google Scholar 

Download references


We wish to acknowledge the work of the wider IMPaCT Team: Bee Harries, Susan Moore, Stefania Bonaccorso, Anna Kolliakou, Conan O’Brien, Ali Featherman, Catherine Fung, Margaret Heslin, Keji Dalemo, Stella Anakwe-Umeh, Gill Todd, Manyara Mushore, Stanley Mutsatsa, Diana Orr, Oliver Howes, John Lally, Ruth Ohlsen, Evangelos Papanastasiou, Mohammad Firdosi, Hannah Sallis, Irene Sambath, Guilia Di Clemente, Josefine Breedvelt, Candice Joseph, Marta Di Forti, Adewale Odesanya, Dami Onagbesan; Philippa Case; Hannah Sallis, Augustine Musa, Christine Dalton, Hannah Antionades, Ben Reece, Andy Healy, Funda Sinan, Keerthana Rudhra, Hannah Kelly, Janet Treasure, Anthony Davis, Caroline Murphy, Joanna Kelly, Matthew Goldin.

We wish to thank the patients and staff of the participating Trusts for their support of the trial and the members of the Trial Steering Committee for their expert guidance; Prof David Osborn (chair), Dr. Jack Gibson, Dr. Adrienne Reveley, Prof Richard Drake.


This paper summarises independent research funded by the National Institute for Health Research (NIHR) under its IMPACT Programme (Grant Reference Number RP-PG- 0606-1049) and had support from the National Institute for Health Research (NIHR) Biomedical Research Centre at South London and Maudsley NHS Foundation Trust and King’s College London. BS and FG are, in part, funded by the National Institute for Health Research Collaboration for Leadership in Applied Health Research & Care Funding scheme. The views expressed in this publication are those of the authors and not necessarily those of the NHS, the National Institute for Health Research or the Department of Health. Neither the funding source nor the sponsor had any role in the design of this study or its execution, analyses, interpretation of the data, or decision to submit results.

Availability of data and materials

The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.

Author information

Authors and Affiliations




FG, DS, KI, KG, ZA, AP, RM, DH, AD and SS conceived the study design and objectives and interpreted the data. PGS, JL, BS, JE, SC significantly contributed to the analysis and interpretation of the study results and PL, MA, DO provided patient, carer and/or public input throughout. All authors read and approved the final manuscript. All authors meet the criteria for authorship in accordance with ICJME having made: a) substantial contributions to the conception or design of the work; or the acquisition, analysis, or interpretation of data for the work; b) drafting the work or revising it critically for important intellectual content; c) final approval of the version to be published; and d) agree to be accountable for all aspects of the work in ensuring that questions related to the accuracy or integrity of any part of the work are appropriately investigated and resolved. FG is guarantor for the manuscript and data contained within. FG affirms that the manuscript is an honest, accurate, and transparent account of the study being reported; that no important aspects of the study have been omitted; and that any discrepancies from the study as planned have been explained.

Corresponding author

Correspondence to Fiona Gaughran.

Ethics declarations

Ethics approval and consent to participate

Ethical approval was obtained from the joint South London and Maudsley and the Institute of Psychiatry NHS Ethics Committee (REC Ref no 09/HO80/41). Once participants were identified, they were approached by trained researchers who explained the trial and gave out an information leaflet regarding the study for consideration. The researcher then met the potential participant to answer any further questions they may have and learn if they wished to consent. Potential participants lacking capacity were ineligible for inclusion in the trial. Participants gave written informed consent, having been informed that their decision whether or not to participate in the study was purely voluntary and, in the case of patient participants, would have no effect on their clinical care. Participants were also told that they were free to withdraw their consent at any time, without their clinical care being affected.

Consent for publication

Not applicable.

Competing interests

FG has received honoraria for advisory work and lectures or CME activity support from Roche, BMS, Lundbeck, Otsaka, Janssen and Sunovion, is a collaborator on a NHS Innovations project co-funded by Janssen and has a family member with professional links to Lilly and GSK, including shares.

KI, speaker fees for Eli Lilly, Janssen, Sanofi.

RM Honoraria for lectures from Lundbeck, Janssen, and Sunovion.

JE is an employee of Lundbeck.

All other authors declare they have no conflict of interest.

Publisher’s Note

Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Additional information

Fiona Gaughran and Daniel Stahl are are joint first author.

Robin Murray and Shubulade Smith are joint last authors.

Rights and permissions

Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (, which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver ( applies to the data made available in this article, unless otherwise stated.

Reprints and permissions

About this article

Check for updates. Verify currency and authenticity via CrossMark

Cite this article

Gaughran, F., Stahl, D., Ismail, K. et al. Randomised control trial of the effectiveness of an integrated psychosocial health promotion intervention aimed at improving health and reducing substance use in established psychosis (IMPaCT). BMC Psychiatry 17, 413 (2017).

Download citation

  • Received:

  • Accepted:

  • Published:

  • DOI: