SUPPLEMENTARY MATERIAL
METHODS
Study design
This narrative review protocol was written before the start of the literature search. We wrote this manuscript in accordance with the PRIMSA guidelines. For the purpose of this review we limited ourselves to the evidence available from MEDLINE (PubMed) published between January 1, 2017 and December 31, 2018 using the following search strategy:
("diseases category"[Mesh] OR "chronic somatic" OR “chronic disease” OR „chronic illness“
OR „chronic medical“ OR “chronic physical ” OR „chronic condition“ OR “non- communicable disease” OR “noncommunicable diseases” OR “somatic comorbidity” OR multimorbid* OR “medical comorbidity” OR diabetes OR metabolic OR obesity OR COPD OR “obstructive pulmonary” OR asthma OR lung OR respiratory OR cardiovascular OR coronary OR heart OR periodont* OR cancer OR malignant OR tumor OR tumour OR oncolog* OR renal OR kidney OR liver OR HIV OR hepatitis OR gastrointestinal OR ulcer OR gerd OR „sexually transmitted“ OR osteoarthritis OR neurological OR inflammation OR autoimmune ) AND ( “treatment outcome”[Mesh] OR rehospitalization OR re-admission OR readmission OR relapse OR „recurrence“ [Mesh] OR admission OR “clinical course” OR progression OR remission OR „prognosis“ [Mesh] OR decompensation ) AND ( schizophren* [tiab] OR psychotic [tiab] OR psychosis [tiab] ) AND ("2017/01/01"[PDAT] :
"2018/12/31"[PDAT]).
Studies screening and eligibility
We included studies whose targeted population were adult (age ≥18 years) patients of any gender, diagnosed with schizophrenia (ICD-10: F20), other schizophrenia spectrum disorder (ICD-10: F21-F29) excluding the studies who targeted only acute and transient psychotic disorder (ICD-10: F23). We included studies who used the following outcomes: relapse of schizophrenia, rehospitalization, severity of psychotic symptoms, neurocognitive functioning, symptomatic remission, functional remission, social functioning, quality of life and suicide attempts. We excluded the length of hospitalization as the targeted outcome. We included only primary randomized and non-randomized controlled trails, primary retrospective or prospective cohort studies and case-control studies. We excluded cross-sectional studies, case series and case studies, letters to the editors, commentaries, editorials, review articles, thesis and dissertations. Two reviewers (ISF, ZB) independently screened the studies based on title and abstract. Differences were resolved by consensus of two reviewers and principal investigator (IF).
Quality assessment and study inclusion
In the second step we retrieved full texts of all eligible studies and assessed the quality of each candidate study on both the study and the targeted outcomes levels. We assessed the quality of studies using the Downs and Black instrument (1). We did the quality assessment from the perspective of our review objectives and outcomes, and not from the perspective of the included studies primary objectives and outcomes. We did not calculate the Down and Black instrument total score because we think it is not appropriate to sum particular potential sources of bias implicitly assigning them the weight of 1. Instead, we calculated the percentage of studies having a good quality solution for the particular Down and Black
instrument’s item. Differences were resolved by consensus of two reviewers and the principal investigator.
Data extraction and synthesis
Data extraction from the included studies were: country, targeted population with the diagnostic and the main exclusion criteria used, sample size, mean age of participants and the proportion of participants being women, definition of CPI and CPM studied, outcomes relevant for our review domains and specific instruments, key findings, and the summary of each study limitations. We did not performed a meta-analysis, but only the qualitative synthesis, because the included studies were very heterogeneous with regards to the targeted population, design, exposures, that is to the CPI studied, therapies and outcomes.
Included studies characteristics
Four of the included studies were performed on the European (Italy, Denmark, Croatia, Germany)(2–5), two on the USA(6,7), and none on the populations from the other parts of the world (Table 1). One of the included studies were performed in the middle-income country (Croatia)(2), and non on the low-income countries. All studies included both genders but in different ratios. Proportion of women ranged from 29% in Anderson 2017. to 59% in Lambert 2017 (Table 1). The mean age of participants was from 34 to 50 years. The most important difference between included studies was with regard to the CPI studied. Two studies used the specific, but composite CPI: metabolic syndrome defined according to ATPIIIA criteria including waist circumference, triglycerides, HDL-cholesterol, blood pressure and glycaemia (Bosia 2018) or cardiometabolic CPI including cerebrovascular disease, coronary or ischemic
heart disease, diabetes mellitus, hyperglycemia, hyperlipidemia, and hypertension (Correll, 2017). Two studies used all CPI (Filipcic 2017, and Lambert 2017), but both operationalize the independent variable as the numeric indicators: Filipcic 2017 as the number of CPI, and Lambert 2017 as the dichotomous variable indicating having or not having any CPI. Finally, two studies used specific CPI and studied them separately: Nielsen 2018 cardiovascular disease, diabetes, COPD, and hepatitis, Anderson 2017 hypertension, hyperlipidemia, heart disease, diabetes, asthma and lung conditions. None of the included studies take the severity, duration and treatment of CPI into the account. One of the included studies was focused on the cognitive outcomes only using the specific Brief Assessment of Cognition in Schizophrenia instrument (Bosia 2018), three used the rehospitalization (Nielsen 2018, Filipcic 2017, Correll 2017), one used the remission status assessed by the Structured Clinical Interview for Symptoms of Remission (SCI-SR) and defined as score of ≤3 (mild) in all eight SCI-SR symptom areas (Anderson 2017), and one study used the set of four subjective outcomes: severity of psychopathology measured using Brief Psychiatric Rating Scale (BPRS), severity of illness measured using Clinical Global Impression - severity (CGI-S), functioning measured using Global Assessment of Functioning (GAF), and quality of life measured using Quality of Life Enjoyment and Satisfaction Question (Q-LES) (Lambert 2017). Follow-up period was short (up to one month) in two (Bosia 2018, and Correl 2017), moderate (up to a year) in (Nielsen 2018, Andersent 2017) and long (more than a year) in two studies (Filipcic 2017, Lambert 2017).
Quality assessment of the included studies
None of the included studies was a randomized controlled trial, two did not applied the adequate adjustment for possible confounders, and only one took the losses of patients to
follow-up properly into the account (Table 2). Therefore, although in all studies the participants were recruited from the same populations and over the same period of time, our review have a higher risk of lower internal validity due to the uncontrolled confounders. None of the included studies made the attempt to blind those measuring the main outcomes. Studies using the more subjectively assessed outcomes were more vulnerable to this potential source of bias (Bosia 2018, Anderson 2017, and Lambert 2017). Reporting, statistical analysis, and the main outcome measures were adequate in all included studies. Two studies had lower generalizability because they were designed to test the efficacy of two specific therapies:
Bosia 2018 to test the Cognitive remediation therapy, and Lambert 2017 The Integrated Care Treatment specifically focused on the primary and secondary prevention and treatment of CPI.
Supplementary material: Table 2. Risk of bias assessment using the Downs and Black instrument (0=no or unable to determine, 1=yes;
n/a=not applicable)
Bosia, 2018 (n=138)
Nielsen, 2018 (n=10509
)
Anderson, 2017 (n=763)
Filipcic, 2017 (n=301)
Correll, 2017 (n=57506)
Lambert , 2017 (n=187)
Total
Reporting
1. Is the hypothesis/aim/objective of the study clearly described? 1 1 1 1 1 1 100%
2. Are the main outcomes to be measured clearly described in Introduction or
Methods section? 1 1 1 1 1 1 100%
3. Are the characteristics of the patients included in the study clearly described? 1 1 1 1 1 1 100%
4. Are the interventions of interest clearly described? 1 1 1 0 0 1 67%
5. Are the distribution of principal confounders in each group of subjects to be
compared clearly described? 1 0 1 1 1 1 83%
6. Are the main findings of the study clearly described? 1 1 1 1 1 1 100%
7. Does the study provide estimates of the random variability in the data of the main
outcomes? 1 1 1 1 1 1 100%
8. Have all important adverse events that may be a consequence of the intervention
been reported? n/a n/a n/a n/a n/a n/a 0%
9. Have the characteristics of patients lost to follow-up been described? 0 0 1 0 0 0 17%
10. Have actual probability values been reported? 1 1 1 1 1 1 100%
External validity
11. Were the subjects asked to participate in the study representative of the entire
population from which they were recruited? 0 1 1 0 1 1 67%
12. Were those subjects who were prepared to participate representative of the entire
population from which they were recruited? 0 1 1 0 1 1 67%
13. Were the staff, places and facilities where the patients were treated, representative
of the treatment the majority of patients receive? 0 1 0 1 1 0 50%
Internal validity (bias)
14. Was an attempt made to blind study subjects to the intervention they have
received? 0 0 0 0 0 0 0%
15. Was an attempt made to blind those measuring the main outcomes of the
intervention? 0 0 0 0 0 0 0%
Bosia, 2018 (n=138)
Nielsen, 2018 (n=10509
)
Anderson, 2017 (n=763)
Filipcic, 2017 (n=301)
Correll, 2017 (n=57506)
Lambert , 2017 (n=187)
Total
16. If any of the results of the study were based on “data dredging”, was this mode
clear? n/a n/a 1 n/a n/a n/a 17%
17. Do the analysis adjust for the different lengths of follow-up of patients? n/a 1 1 1 1 1 83%
18. Were the statistical tests used to assess the main outcomes appropriate? 1 1 1 1 1 1 100%
19. Was compliance with the intervention reliable? 1 1 1 0 0 0 50%
20. Were the main outcome measures used accurate (valid and reliable)? 1 1 1 1 1 1 100%
Internal validity (confounding; selection bias)
21. Were the patients in different intervention groups recruited from the same
population? 1 1 1 1 1 1 100%
22. Were study subjects in different intervention groups recruited over the same
period of time? 1 1 1 1 1 1 100%
23. Were study subjects randomized to intervention groups? 0 0 0 0 0 0 0%
24. Was the randomized intervention assignment concealed from both patients and
health care staff until recruitment was complete and irrevocable? n/a n/a n/a n/a n/a n/a 0%
25. Was there adequate adjustment for confounding in the analysis from which the
main findings were drown? 0 1 1 1 1 0 67%
26. Were losses of patients to follow-up taken into account? 0 0 1 0 0 0 17%
Power 0%
27. Did the study have sufficient power to detect a clinically important effect where
the probability value for a difference being due to chance is less than 5%? 0 1 1 1 1 0 67%
