Observational study of the implementation of a Norwegian adaptation of a tool for monitoring, detection and intervention of cardiovascular risk factors in severe mental disorders. The project investigates if the medical care is improved or the cardiovascular risk is reduced after implementation. Patients and service provision/clinical interventions is assessed at baseline and at 12 months, with a historic patient comparison group with the same assessments except for service provision/clinical interventions.

1. AIMS

This study aims at answering whether a Norwegian adaptation of a tool for monitoring, detection and intervention of cardiovascular risk factors (hereafter: the CMR-tool) may be effectively implemented and thus improve the care for this patient group, and ultimately reduce cardiovascular risk. Specific research questions:

1. Is there any improvement in received services between units implementing the CMR-tool after one year compared to other non-implementation units and the same unit before implementation? (Health care assessment)
2. Is there any improvement in patients' cardiovascular risk factors between units implementing the CMR-tool after one year compared to other non-implementation units and the same unit before implementation? (Patient related factors)

2. MATERIAL AND METHODS 2.1 Sample and setting Mental Health Specialized Units: Out and inpatient departments providing services for patients with psychotic or psychosis-related conditions in the specialized health care service in Norway. Several university hospitals will cooperate. The implementation of the Tool will be a part of the hospitals' quality improvement work. The CMR-tool will be implemented sequentially in the different hospitals according to the capacity of the hospital organizations.

Community Care: The current project will include patients receiving treatment in the specialized health care system. However, many/all participants will be followed on somatic issues by the family doctors after discharge from hospital treatment/the specialized health care system.

The research project was performed as an add-on to quality improvement interventions in different hospitals and regions in Norway, involving the implementation of a new clinical procedure. The hospitals had differences in details in the practical implementation of this quality improvement intervention; the research protocol describes the minimum common denominator. The research protocol is adjusted to meet the demands for broad scale feasible and effective interventions in the clinic.

2.2 Inclusion criteria Patients in the mental health care for adults in the participating hospitals, aged 18 years + and with a ICD-10 diagnosis of schizophrenia, schizoaffective disorder, schizophreniform disorder, psychotic disorder NOS, bipolar I and II disorder or bipolar disorder NOS or current use of antipsychotic medication have been included. Exclusion criterion is inability to give informed consent.

2.3 Study design Intervention project with a 12 month follow up period using a historical cohort for control. Time from start of first intervention until last follow up: 24 months. Long term (10y) follow up on registry data has been planned.

Each site/organizational unit defined a date for start of intervention from which the use of the CMR-tool was described and made mandatory in the hospitals'/units' clinical procedures. In units with close collaboration the intervention date should be the same because of potential "contamination effects" on the baseline conditions.

Baseline assessments were performed before intervention for the individual patient. This means that baseline assessments could be made after the intervention date for patients novel to the unit who have not been exposed to the intervention (CMR-tool) before. In outpatient clinics this was most often the case. However, in inpatient units the implementation of a new routine affected all patients, and baseline measures was done before the units' intervention date. For practical reasons (workload for the clinicians) there has been set a maximum limit of two months from baseline to the units' intervention date.

Some units chose also to include newly admitted inpatients after the units' intervention date, procuring that these patients are CMR-tool naïve at baseline.

The baseline assessments include all assessments necessary for complying with the CMR-tool's "Danger Zone" (Faresone) check.

Comparisons in baseline and follow up measures are performed in the intervention group, changes in measurement values are compared with the control group.

2.4 Assessments Baseline assessments were performed within two months before participants are exposed to the intervention. Follow-up assessments were performed 12 months after whichever comes last of baseline assessment or intervention date (+/- 1 month).

Assessments were organized in the following charts: BASELINE (baseline), IMPLEMENTERING (level of implementation/fidelity), 12 MND OPPFØLGING (12-months follow-up), BLODPRØVER (blood tests).

Performed at baseline: BASELINE, IMPLEMENTERING, BLODPRØVER Performed at 12 months follow up: 12 MND OPPFØLGING, IMPLEMENTERING, BLODPRØVER Below assessments' organization in charts is accounted for with name of chart type in brackets.

Level of implementation: The history of risk assessment and interventions during the past year were explored using the patient charts. Clinicians' adherence to the guidelines expressed in the Tool and existing national guidelines were monitored. (IMPLEMENTERING) Cardiovascular risk factors: Measurements were made according to established guidelines (De Hert et al 2008). Weight and height (BMI is calculated). Waist circumference is measured with a horizontal tape measurement from the top of the right iliac crest. Systolic and diastolic blood pressure is measured. (BASELINE, 12 MND OPPFØLGING) S-Glucose, HbA1c, fasting plasma lipids (total cholesterol, HDL-chol, LDL-chol, TGA), leptin, T4, TSH, (Cortisol measurements) and metabolic hormones are measured as well as hsCRP. (BLODPRØVER) Cardiovascular outcome: The main outcome measure on cardiometabolic risk is the risk assessment score "NORRISK" published by the Norwegian Health Directorate (1). The tool uses data on sex, age, smoking status, systolic blood pressure and total serum cholesterol, and a calculator is publicly available at the website of the Norwegian Health Informatics (Norsk helseinformatikk - NHI): http://nhi.no/forside/skjema-og-kalkulatorer/kalkulatorer/norrisk-10-arsrisiko-for-kardiovaskuler-dod-30384.html. (BASELINE, 12 MND OPPFØLGING) Symptoms and functioning: Global symptoms and psychosocial functioning are measured by the Global Assessment of Functioning Scale (GAF), the scores are split into scales of symptoms (GAF-S) and functioning (GAF-F) to improve psychometric properties (2). The Clinical Global Impression (CGI) Scale is a simple measure of the global impression of disease severity used extensively in intervention studies (3, 4) (BASELINE, 12 MND OPPFØLGING) Sociodemography and somatic health: Data were collected about age and sex, urban/rural living, ethnicity, education, occupation, housing and marital/civil status. (BASELINE) Further, information about exercise, diet, smoking and somatic disease is recorded from patient/charts. (BASELINE, 12 MND OPPFØLGING) Diagnosis and treatment history: Diagnosis was obtained by a psychiatrist or a specialist in clinical psychology according to criteria in ICD-10. The diagnosis will be quality controlled by an assessment team with regular diagnosis consensus meetings. They all received similar training. Age at onset of illness has been recorded as well as some basic clinical characteristics (hospital admissions, suicidality, violence, sleep). (BASELINE) Current treatment will be registered. (BASELINE, 12 MND OPPFØLGING) Substance use: The onset and use of different substances were registered. (BASELINE, 12 MND OPPFØLGING) Current level of physical activity is assessed by using the five questions (questions 32-36) of physical activity in the HUNT questionnaire which cover frequency, intensity, duration and time in inactivity (5, 6). (BASELINE, 12 MND OPPFØLGING) Life satisfaction is assessed by the question for life satisfaction in the HUNT3 questionnaire (question 54) (7). (BASELINE, 12 MND OPPFØLGING) Medication: Data about current psychopharmacological medication and the use of statins is collected. (BASELINE, 12 MND OPPFØLGING) Diet: Diet habits are recorded as "healthy", "low level unhealthy" and "high level unhealthy" according to a broad judgement by the clinician based on parameters as variation in foods, regularity of meals, content of white sugar and total calories. (BASELINE, 12 MND OPPFØLGING) The assessment charts are attached 2.5 Control group The NORMENT study has existing 12 months follow-up data on CMR in 1700 well-described patients with severe mental illness (schizophrenia, schizoaffective disorder, schizophreniform disorder, psychotic disorder NOS, bipolar I and II disorder or bipolar disorder NOS). These patients were included from representative clinical cohorts across Norway until 2014, independently of the current implementation of the CMR-Tool and without any systematic change in practice on CMR management.

2.6 Interventions The CMR-Tool: The interventions consisted of education and distribution of the Tool and establishment of clinical routines regarding the use of the Tool (see attachment). From the defined intervention start date, the participating clinicians started to use the Tools' instructions on clinical practice on all patients which comply with the inclusion criteria.

2.7 Organization and quality of assessments and interventions Simple somatic measures were done by nurses or MDs. Other measures were collected by the clinician or project assistants. Only certified laboratories were used for blood samples.

2.8 Procedure for informed consent and handling of data See Standard Operating Procedure (SOP)

Additional protocol:

The patients were asked to participate in the NORSMI "basic protocol" study -with a separate informed consent. This did not involve more investigations, but included collecting of blood or salive for biobank and consent to collect register info data (REC approval: 2009/2485/REK sør-øst C).

3. CONSIDERATIONS OF STATISTICAL POWER The main outcome measure for cardiometabolic risk is the NORRISK risk score. With an expected mean score of 5.0, an expected mean standard deviation of 4.0, a proposed effect of 30 % and a statistical power of 0.8, the necessary sample size will be 60 for paired t-tests pre-post comparisons and 120 for independent t-tests in group comparisons (8).

4. PROJECT PLAN AND TIME ESTIMATION The intervention started in the third quarter of 2016 and ended in 2020. The follow up period will be 10 years, and the project will end in 2028.

The study is coordinated with a programme for improvement of quality of the health care services in the same departments. The Tool is primarily aimed at physicians working in the mental health care system, but other professions may also have important roles in the use of the Tool, and it can be integrated with different types of local procedures.

The patients are only subject to evaluations and interventions which are considered to be part of a regular "best-practice" treatment regimen. It is therefore not expected that the project has implied or could further imply any negative effects for the participants.

Inclusion Criteria:

• Age 18+. Diagnosis of ICD10 F2 or F3 and/or use of antipsychotics

Exclusion Criteria:

• inability to give informed consent

Severe mental disorder or use of antipsychotics