Predictors in use of mental health resources: The role of behaviour problems in patients with severe mental illness

Abstract

Background:

In recent years, more variables are being included in the use of mental health resource prediction models. Some studies have shown that how well the patient can function is important for this prediction. However, the relevance of a variable as important as behaviour problems has scarcely been explored.

Aim:

This study attempted to evaluate the effect of behaviour problems in patients with severe mental illness on the use of mental health resources.

Method:

A total of 185 patients at a Community Mental Health Unit were evaluated using the Behaviour Problem Inventory. Later, a bivariate logistic regression was done to identify what behaviour problems could be specific predictors of use of mental health resources.

Results:

The results showed that the general index of behaviour problems predicts both use of hospitalization resources and outpatient attention. Underactivity/social withdrawal is the best predictor of all the different areas.

Conclusion:

These results confirm the role of behaviour problems as predictors of the use of mental health resources in individuals with a severe mental illness.

Keywords

Schizophrenic disorders bipolar disorder behaviour problems hospital admissions community services

Introduction

Several different mental disorders (schizophrenia, bipolar disorder, severe depressive disorder and other psychotic disorders) are grouped together under the term Severe Mental Illness. Those who suffer from them may experience very diverse symptoms, and share the same diagnosis, even though they are quite different from each other. The only common denominator seems to be a significant effect on functioning and social adaptation (American Psychiatric Association (APA), 2013). Some authors have proposed using this criterion to understand severe mental illness better by developing instruments able to relate behaviour variables observed to social functioning (Carpenter & Strauss, 1991). Two main approaches have derived from this criterion, one concentrating on social functioning (Birchwood, Smith, Cochrane, Wetton, & Copestake, 1990) and the other related to behaviour problems that interfere with social integration (Wykes & Sturt, 1986).

Individuals who suffer from a severe mental disorder require intense use of mental health resources to be able to function well socially and reduce their limitations due to their behaviour problems as far as possible (Knapp, Mangalore, & Simon, 2004), thereby increasing the cost of their care above other types of mental disorders (Andrews, Issakidis, Sanderson, Corry, & Lapsley, 2004; Carr et al., 2003; Somaiya, Grover, Avasthi, & Chakrabarti, 2014).

Previous studies have attempted to determine what variables influence the extent to which mental health resources are used by persons with mental disorders. These studies have mainly explored socioeconomic variables such as age (Jin et al., 2003), gender (Lindamer et al., 2003; Usall et al., 2012), ethnic group (Lee, Laiewski, & Choi, 2014; Mann, Fisher, & Johnson, 2014), socioeconomic status (Kilian, Matschinger, Becker, & Angermeyer, 2003; Tello et al., 2005) or education (Donisi et al., 2013; Have, Oldehinkel, Vollebergh, & Ormel, 2003; Jin et al., 2003; Kilian et al., 2003; Lindamer et al., 2003; Tello et al., 2005; Usall et al., 2012).

Later studies have added to the number of variables included, such as those related to urban context and neighbourhood (Donisi et al., 2013; Ngamini Ngui, Perreault, Fleury, & Caron, 2012) and other clinical variables (Moreno-Kustner et al., 2011), leading to a more complete vision of what determines use of mental health resources. Finally, the most recent studies have started to include social functioning variables (Bellido-Zanin, Pérez-San-Gregorio, Martín-Rodríguez, & Vázquez-Morejón, 2015; Raudino et al., 2014), providing relevant data for predicting the use of these resources.

It would therefore be of interest to study the role of behaviour problems in predicting the use of mental health resources, as it would explore another side of social adaptation and what it contributes to prediction of these resources.

Previous studies have attempted to determine the most relevant behaviour problems in psychotic disorders. Four behavioural factors were found in schizophrenia in a study (Harvey, Curson, Pantelis, Taylor, & Barnes, 1996) that was later replicated and confirmed (Curson, Duke, Harvey, Pantelis, & Barnes, 1999). Those four factors, found using the Social Behaviour Schedule (Wykes & Sturt, 1986), were thought disturbance, social withdrawal, depression behaviour and antisocial behaviour. This new perspective in understanding the different components of psychotic disorders made it possible to study community adaptation of patients with this diagnosis (Cella et al., 2014).

As far as the authors know, there has been no research to date including this variable in mental health resource prediction models, so this is intended as an exploratory study on the subject.

This study attempted to determine the influence that behaviour problems of patients with severe mental illness has on the use of mental health resources. Our hypothesis is the following:

Behaviour problems on patients with severe mental illness are positively associated with a higher use of mental health services.

Method

Subjects

The sample comprised 185 patients who received care at a Community Mental Health Unit (CMHU) and were diagnosed with a severe mental disorder: schizophrenia (F. 20 according to the International Classification of Diseases, 10th edition (ICD-10)), other psychotic spectrum disorders (F.21–F.29 according to ICD-10), bipolar disorder Type 1 (F.31 according to ICD-10) and severe depression disorder (F.32 according to ICD-10).

The sample was selected by convenience sampling in which the patients in a psychoeducation programme given in the CMHU were asked to participate in the evaluation. All of them accepted.

The clinical psychologist or psychiatrist responsible for each patient made the diagnosis using a clinical interview. The criteria for inclusion in the study were that they had to (1) be under treatment in the CMHU at the time of the study, (2) have one of the diagnoses mentioned above, (3) be 18–65 years of age and (4) give their consent to participating in the study. Exclusion criteria were primary diagnosis of substance use disorder or mental retardation.

Instruments and measures

Behaviour Problem Inventory

The scale comprises 14 items which contain the most relevant behaviour problems found in individuals with psychosis. The main purpose of the scale is to evaluate behaviour problems observed in patients with a severe mental disorder quickly and effectively. The items were based on a review of prior instruments, such as the Social Behaviour Schedule (Wykes & Sturt, 1986), the REHAB (Baker & Hall, 1988) and the PC section of the Social Behaviour Assessment Schedule (SBAS) (Platt et al., 1980) combined with the clinical and research experience of the authors themselves. The criterion followed was that each item be insofar as possible an observable behaviour. All answers refer to the patient’s behaviour during the past 3 months.

A total score is found from the sum of the scores on each of the 14 items and those on the three subscales (identified by factor analysis): (1) Underactivity/Social Withdrawal, (2) Active Problems and (3) Lack of Impulse Control. Another two scores are also found: (1) the Moderate Behaviour Problem (MBP) score, equal to the number of items with a score over 2, and (2) Severe Behaviour Problem (SBP) score, equal to the number of items with a score of 3. Higher scores on each scale and index show more problematic behaviour in the patients evaluated. Preliminary data on the scale (Vázquez-Morejón et al., 2005) confirm that scale characteristics are adequate, and the final adaptation of the instrument is now in press. Cronbach’s alpha calculated for the sample in this study is .82. Five measures with this instrument were included in this study, the three main subscales plus the SBP and MBP indices.

Mental health resource use measures

Resource use variables were found using data available in the Andalusian Healthcare System’s computerized records of every patient’s contacts with mental health services. Two variables were included: one related to outpatient visits, that is, the total number of CMHU interventions. The CMHU is a community outpatient resource where there is a variety of mental health professionals (psychiatrists, clinical psychologists, social workers and nurses) specialized in all types of mental health disorders. The other was hospital admissions in the Inpatient Unit during the follow-up period. The rest of the resources (the Mental Health Therapeutic Community and the Mental Health Rehabilitation Unit) were excluded due to the small number of patients in the sample using them during the follow-up period. None of the patients were referred to the Mental Health Therapeutic Community and only six patients in the sample were referred to the Mental Health Rehabilitation Unit. Nevertheless, patients who use these kinds of resources were not excluded of the sample.

Procedure

After giving their informed consent, the Behaviour Problems Inventory was administered to the patients’ main caregiver. Later, each patient whose relative had filled in the questionnaire was followed up for 36 months after evaluation. Each contact with the mental health services was collected by means of its identity number in the computerized records of the Andalusian Healthcare System as described under section ‘Instruments and measures’.

Statistical analysis

Statistical analysis was done using the SPSS v.22 statistical package:

A descriptive analysis of the type and amount of resources needed by the patients during the 36-month follow-up period.

Logistic regression analysis was used to identify significant predictors of a more intensive use of resources, as the data collected did not fulfil the requirements for linear regression. The two dependent variables were therefore dichotomized to fit to the logistic regression criteria. The measure of the hospital admission variable was whether there had been admission after assessment. Appointments at the CMHU were reclassified according to sample distribution into one group under the 50th percentile with average care of fewer than one contact per month and another group above that percentile with more than one contact per month. This decision was made on the basis of statistical (the variable did not meet the basic assumptions of normality and homoscedasticity for linear regression analysis) and clinical criteria by which more than one contact per month could show a high level of assistance in the Andalusian Healthcare System. Two analyses were made per resource, the first using the SBP and MBP scores and the other with each of the Behaviour Problem Inventory subscales.

Results

Descriptive analysis

The average age of the sample was 37.63 years (SD = 9.71; range = 17–56); 114 of whom were males (61.6%) and 77 females (38.4%). Most participants were single (N = 143, 77.3%), 26 were married (14.1%) and the rest separated or widowed (9.6%). A total of 102 participants were diagnosed with schizophrenia (55.1%), 53 with other psychotic disorders (28.7%), 25 with bipolar affective disorder (13.5%) and 5 with severe depressive disorder (2.7%).

Table 1 shows the descriptive analysis of sociodemographic data.

Table 1.

Descriptive analysis of sociodemographic data of the patients who participated in the study (n = 185).

Variables	n	%
Sex
Women	114	38.4
Men	71	61.6
Age
<26	18	9.7
26–35	51	27.6
36–45	58	31.4
46–55	31	16.8
56–65	27	14.6
Marital status
Single	143	77.3
Married	26	14.1
Separated	13	7
Widowed	3	1.6
Education
Incomplete primary school	25	13.4
Primary school	45	24.3
Secondary/grade school	65	35.1
College/university	37	2
Unknown	13	7
Diagnosis
Schizophrenia	102	55.1
Other psychotic disorder	53	28.7
Bipolar disorder	25	13.5
Severe depressive disorder	5	27

Table 2 shows the results of the descriptive analysis of resource use during the 36-month follow-up period. All the patients were in contact with the CMHU during the 3 years of follow-up while only a small percentage had any hospital admission during the first, second or third year.

Table 2.

Descriptive analysis of the use of mental health resources during the 36-month follow-up period (n = 185).

Type of resource	No. of patients who made use of the unit	% of patients who made use of the unit	M of contacts	SD of contacts
1st year CMHU	185	100	17.44	16.35
2nd year CMHU	185	100	14.89	17.15
3rd year CMHU	185	100	13.67	17.14
1st year IU	13	7	1.54	1.13
2nd year IU	12	6.5	1.33	.65
3rd year IU	11	5.9	1.64	1.03

CMHU: Community Mental Health Unit, IU: inpatient unit, M: mean; SD: standard deviation.

Table 3 shows the results of the descriptive analysis of the various Behaviour Problem Inventory subscales and item by item.

Table 3.

Descriptive analysis of data on behaviour problems collected from the Behaviour Problem Inventory (n = 185).

	M	SD
Moderate Behaviour Problems score	3.69	3.07
Severe Behaviour Problems score	1.32	2.17
Underactivity/social withdrawal subscale	5.6	3.84
Active problems subscale	4.08	3.72
Lack of impulse control subscale	0.95	1.85

M: mean; SD: standard deviation.

Mental health resource use predictors

Table 4 shows the MBP and the SBP scores and the Behavioural Problems subscales as predictors of the various mental health resources during 3 years. It may be observed in the table that the SBP score was a good predictor of appointments at the CMHU during the first year (p = .017) and hospital admissions once or more during the first year (p = .010) and the second year (p = .054), while the rest of resources were not significant. Similarly, the MBP score predicted appointments at the CMHU during the first year (p = .010), second year (p = .035) and third year (p = .031). It also predicted hospital admissions during the first year (p = .022), the second year (p = .024) and the third year (p = .014). The variance explained by the model using the Nagelkerke measure represents from 3.2% to 8.8%, which is a small proportion of little or moderate variance. The most variance explained is in predicting hospitalization during the first year (8.8%).

Table 4.

Severe Behaviour Problem (SBP) and Moderate Behaviour Problem (MBP) scores and Behaviour Problems subscales as a predictor of other types of resources (n = 185).

	OR	95% CI	p
1st year appointments at CMHU
SBP (R² Nagelkerke = .040)	1.185	1.023–1.373	.017
MBP (R² Nagelkerke = .048)	1.136	1.029–1.254	.010
2nd year appointments at CMHU
SBP (R² Nagelkerke = .015)	1.104	.964–1.265	.149
MBP (R² Nagelkerke = .032)	1.108	1.006–1.221	.035
3rd year appointments at CMHU
SBP (R² Nagelkerke = .007)	1.070	.934–1.225	.330
MBP (R² Nagelkerke = .034)	1.113	1.009–1.227	.031
1st year hospital admissions
SBP (R² Nagelkerke = .088)	1.297	1.077–1.561	.010
MBP (R² Nagelkerke = .070)	1.216	1.032–1.434	.022
2nd year hospital admissions
SBP (R² Nagelkerke = .052)	1.232	1.013–1.499	.054
MBP (R² Nagelkerke = .071)	1.220	1.030–1.447	.024
3rd year hospital admissions
SBP (R² Nagelkerke = .002)	1.047	0.810–1.353	.735
1st year appointments at the CMHU		1.009–1.089	.014
Underactivity/social withdrawal subscale (R² Nagelkerke = .069)	1.132	1.045–1.226	.002
2nd year appointments at the CMHU
Underactivity/social withdrawal subscale (R² Nagelkerke = .031)	1.085	1.004–1.173	.036
1st year hospital admissions
Underactivity/social withdrawal subscale (R² Nagelkerke = .078)	1.191	1.032–1.374	.016
2nd year hospital admissions
Active problems subscale (R² Nagelkerke = .071)	1.176	1.025–1.349	.024

CMHU: Community Mental Health Unit; SBP: severe behaviour problems; MBP: moderate behaviour problems; OR: odds ratio; CI: confidence interval.

The Underactivity/Social Withdrawal Subscale predicted appointments at a CMHU in both the first (p = .002) and second year (p = .036). This Underactivity/Social Withdrawal Subscale also satisfactorily predicted hospital admissions during the first year (p = .016). Hospital admissions in the second year, however, were predicted by the Active Problems Subscale (p = .024).

The variance explained by the model using the Nagelkerke measure represents from 3.1% to 7.8%, which is a small proportion of little or moderate variance, again highest in hospitalization during the first year where the variance explained is 7.8%.

Discussion

This study was intended to examine the role of behaviour problems in predicting the use of mental health resources in the population with severe mental illness. The data found in this study generally confirm the hypothesis of our research, underlining the importance of behaviour problems as predictors of these resources.

In the first place, it may be observed how the general indicators of behaviour problems, such as the MBP and SBP indices, were predictors of most of the mental health resources during the 36-month follow-up.

Specifically, the MBP score was more effective in predicting most of the mental health resources, with higher explained variance. The only exception to this was hospital admissions during the first 12 months of follow-up, in which the variance explained was higher for the SBP score.

These results seem to be in agreement with the nature of the Inpatient Unit itself, which is mainly for occasional admission of patients whose situation is more severe, such as decompensation or attempted suicide. Another possible explanation has been argued by previous studies which have found similar results (Moreno-Kustner et al., 2011) associating the lowest levels of social adaptation, which in this study could be likened to more behaviour problems, with less adherence to mental health services.

In addition to the overall behaviour problem subscales, it would seem relevant to study specifically what types of behaviour problems could predict each of the mental health resources studied, because the behaviour problems a patient with severe mental illness has could vary considerably (Wykes & Sturt, 1986).

This new analysis shows how the Underactivity/Social Withdrawal Subscale is the best predictor of more appointments at the CMHU during the first 2 years and hospital admissions during the first year. Nevertheless, the Active Problems Subscale seems to predict hospital admissions better during the second year of follow-up. None of the subscales were found to be able to predict contacts with mental health resources during the third year.

The consistency shown by the Underactivity/Social Withdrawal Subscale seems coherent with previous studies, which have attempted to determine predictors of the course of psychotic disorders. Several authors have found the importance of negative symptoms for predicting the course of illness (Breier, Schreiber, Dyer, & Pickar, 1991; Sipos, Harrison, Gunnell, Amin, & Singh, 2001) and have also demonstrated how social isolation, which would be related to the Underactivity/Social Withdrawal Subscale, is also an important predictor of the course of illness (Harvey, Jeffreys, McNaught, Blizard, & King, 2007) and the use of mental health resources (Raudino et al., 2014).

The only case in which the Underactivity/Social Withdrawal Subscale was not the best predictor was for hospital admissions during the second year. Again in the case of use of the Inpatient Unit, it seems that the Active Problems Subscale, which could correspond to positive symptomatology in psychosis, was a better predictor, coinciding with the need for hospitalization for decompensation.

In conclusion, it seems that behaviour problems, and especially the underactivity/social withdrawal area, are good predictors of the need for mental health resources of patients with mental disorders. The variance explained is low or moderate, but these results are in line with those that show that the variables predicting resource use are many and varied, from sociodemographic variables (Donisi et al., 2013; Jin et al., 2003; Mann et al., 2014; Usall et al., 2012) to clinical (Moreno-Kustner et al., 2011).

One of the limitations of this study is precisely the inclusion of a single variable and not others such as sociodemographic or clinical. However, the role of those variables has been well studied before, so this study would only have found results similar to those observed previously.

Another possible limitation of the study is that the Behaviour Problem Inventory was filled out only by the main caregiver. Although main caregivers have turned out to be generally good informants (Sabbag et al., 2011), it is true that it would be of interest to check the information provided by other informants, for example, the patient’s clinical referrer or the patient, since the information could vary.

In future studies, other health resources could be included as variables, such as primary attention appointments or house calls, which have been included in other studies (Raudino et al., 2014), for a more complete image of all the resources that each patient could require based on their behaviour problems.

Finally, it is worth mentioning one strong point of this research, which is the 3-year-long follow-up of our sample. Most studies on use of resources are cross-studies or have a shorter follow-up (Donisi et al., 2013; Moreno-Kustner et al., 2011; Raudino et al., 2014) so the inclusion of data for 36 months after evaluation may be considered confirmation of the long-term consistency of the results.

This study has shown the need to increase the number of variables included in current mental health resource use prediction models for patients with severe mental illness and the importance of also including behaviour problems as a variable in such models. Extending mental health resource use prediction models could facilitate more effective health service planning. Furthermore, in the future, it would be of great interest to develop evaluation instruments, such as the Behaviour Problem Inventory, for short, easy application, enabling fast, effective planning of the resources any patient with a severe mental disorder in particular may need.

Footnotes

Acknowledgements

We are grateful to all the professionals working at the Community Mental Health Unit Guadalquivir.

Funding

The author(s) received no financial support for the research, authorship and/or publication of this article.

References

American Psychiatric Association. (2013). Diagnostic and statistical manual of mental disorders (5th ed.). Arlington, VA: American Psychiatric Publishing.

Andrews

Issakidis

Sanderson

Corry

Lapsley

(2004). Utilising survey data to inform public policy: Comparison of the cost-effectiveness of treatment of ten mental disorders. British Journal of Psychiatry, 184, 526–533. doi:10.1192/bjp.184.6.526.

Baker

Hall

J. N.

(1988). REHAB: a new assessment instrument for chronic psychiatric patients. Schizophrenia Bulletin, 14, 97.

Bellido-Zanin

Pérez-San-Gregorio

M. Á.

Martín-Rodríguez

Vázquez-Morejón

A. J.

(2015). Social functioning as a predictor of the use of mental health resources in patients with severe mental disorder. Psychiatry Research, 230, 189–193.

Birchwood

Smith

Cochrane

Wetton

Copestake

(1990). The social functioning scale – The development and validation of a new scale of social-adjustment for use in family intervention programs with schizophrenic-patients. British Journal of Psychiatry, 157, 853–859. doi:10.1192/bjp.157.6.853

Breier

Schreiber

J. L.

Dyer

Pickar

(1991). National institute of mental health longitudinal study of chronic schizophrenia – Prognosis and predictors of outcome. Archives of General Psychiatry, 48, 239–246.

Carpenter

W. T.

JR Strauss

J. S.

(1991). The Prediction of Outcome in Schizophrenia IV: Eleven-Year Follow-Up of the Washington IPSS Cohort. The Journal of Nervous and Mental Disease, 179, 517-525.

Carr

V. J.

Johnston

P. J.

Lewin

T. J.

Rajkumar

Carter

G. L.

Issakidis

(2003). Patterns of service use among persons with schizophrenia and other psychotic disorders. Psychiatric Services, 54, 226–235. doi:10.1176/appi.ps.54.2.226.

Cella

Stratta

Chahal

Huddy

Reeder

Wykes

(2014). Measuring community functioning in schizophrenia with the Social Behaviour Schedule. Schizophrenia Research, 153, 220–224. doi:10.1016/j.schres.2013.12.016

10.

Curson

D. A.

Duke

P. J.

Harvey

C. A.

Pantelis

Barnes

T. R. E.

(1999). Four behavioural syndromes of schizophrenia: A replication in a second inner-London epidemiological sample. Schizophrenia Research, 37, 165–176. doi:10.1016/s0920-9964(98)00151-0

11.

Donisi

Tedeschi

Percudani

Fiorillo

Confalonieri

De Rosa

… Amaddeo

(2013). Prediction of community mental health service utilization by individual and ecological level socio-economic factors. Psychiatry Research, 209, 691–698. doi:10.1016/j.psychres.2013.02.031

12.

Harvey

C. A.

Curson

D. A.

Pantelis

Taylor

Barnes

T. R. E.

(1996). Four behavioural syndromes of schizophrenia. British Journal of Psychiatry, 168, 562–570. doi:10.1192/bjp.168.5.562

13.

Harvey

C. A.

Jeffreys

S. E.

McNaught

A. S.

Blizard

R. A.

King

M. B.

(2007). The Camden Schizophrenia Surveys III: Five-year outcome of a sample of individuals from a prevalence survey and the importance of social relationships. International Journal of Social Psychiatry, 53, 340–356. doi:10.1177/0020764006074529

14.

Have

M. T.

Oldehinkel

Vollebergh

Ormel

(2003). Does educational background explain inequalities in care service use for mental health problems in the Dutch general population? Acta Psychiatrica Scandinavica, 107, 178–187.

15.

Jin

Folsom

D. P.

Lindamer

Bailey

Hawthorne

Garcia

Jeste

D. V.

(2003). Patterns of public mental health service use by age in patients with schizophrenia. American Journal of Geriatric Psychiatry, 11, 525–533. doi:10.1176/appi.ajgp.11.5.525

16.

Kilian

Matschinger

Becker

Angermeyer

M. C.

(2003). A longitudinal analysis of the impact of social and clinical characteristics on the costs of schizophrenia treatment. Acta Psychiatrica Scandinavica, 107, 351–360. doi:10.1034/j.1600-0447.2003.00072.x

17.

Knapp

Mangalore

Simon

(2004). The global costs of schizophrenia. Schizophrenia Bulletin, 30, 279–293.

18.

Lee

Laiewski

Choi

S. H.

(2014). Racial-ethnic variation in U.S. mental health service use among Latino and Asian non-U.S. citizens. Psychiatric Services, 65, 68–74. doi:10.1176/appi.ps.201200430

19.

Lindamer

L. A.

Bailey

Hawthorne

Folsom

D. P.

Gilmer

T. P.

Garcia

… Jeste

D. V.

(2003). Gender differences in characteristics and service use of public mental health patients with schizophrenia. Psychiatric Services, 54, 1407–1409. doi:10.1176/appi.ps.54.10.1407

20.

Mann

Fisher

H. L.

Johnson

(2014). A systematic review of ethnic variations in hospital admission and compulsory detention in first-episode psychosis. Journal of Mental Health, 23, 205–211. doi:10.3109/09638237.2014.910641

21.

Moreno-Kustner

Mayoral

Rivas

Angona

Requena

Garcia-Herrera

J. M.

… Bellon

J. A.

(2011). Factors associated with use of community mental health services by schizophrenia patients using multilevel analysis. BMC Health Services Research, 11, 257. doi:10.1186/1472-6963-11-257

22.

Ngamini Ngui

Perreault

Fleury

M. J.

Caron

. (2012). A multi-level study of the determinants of mental health service utilization. Revue d’Epidemiologie et de Sante Publique, 60, 85–93. doi:10.1016/j.respe.2011.09.007.

23.

Platt

Weyman

Hirsch

Hewett

(1980). The Social Behaviour Assessment Schedule (SBAS): rationale, contents, scoring and reliability of a new interview schedule. Social Psychiatry and Psychiatric Epidemiology, 15, 43–55.

24.

Raudino

Carr

V. J.

Bush

Saw

Burgess

Morgan

V. A.

(2014). Patterns of service utilisation in psychosis: Findings of the 2010 Australian National Survey of Psychosis. Australian and New Zealand Journal of Psychiatry, 48, 341–351. doi:10.1177/0004867413511996

25.

Sabbag

Twamley

E. M.

Vella

Heaton

R. K.

Patterson

T. L.

Harvey

P. D.

(2011). Assessing everyday functioning in schizophrenia: Not all informants seem equally informative. Schizophrenia Research, 131, 250–255. doi:10.1016/j.schres.2011.05.003

26.

Sipos

Harrison

Gunnell

Amin

Singh

S. P.

(2001). Patterns and predictors of hospitalisation in first-episode psychosis: Prospective cohort study. The British Journal of Psychiatry, 178, 518–523.

27.

Somaiya

Grover

Avasthi

Chakrabarti

(2014). Changes in cost of treating schizophrenia: Comparison of two studies done a decade apart. Psychiatry Research, 215, 547–553. doi:10.1016/j.psychres.2014.01.013

28.

Tello

J. E.

Mazzi

Tansella

Bonizzato

Jones

Amaddeo

(2005). Does socioeconomic status affect the use of community-based psychiatric services? A south Verona case register study. Acta Psychiatrica Scandinavica, 112, 215–223.

29.

Usall

Iniesta

Ochoa

Villalta

Suarez

Pezzimenti

… Millan

(2012). Gender differences in the use of mental health services in a Barcelona area. European Journal of Psychiatry, 26, 30–40.

30.

Vázquez-Morejón

A. J.

Jiménez-García-Bóveda

Dorado-García

Conde-Díaz

Mármol-Pérez

Yates-Sosa

Verástegui-Benito

(November, 2005). Problemas de Conducta y Psicosis: Datos preliminares en un estudio llevado a cabo en el Hospital Universitario Virgen del Rocio [Behavioral Problems and Psychosis: Preliminary data in a study carried out in the Virgen del Rocio University Hospital]. Paper presented at the I Congreso Andaluz de Salud Mental Comunitaria, Granada, Spain.

31.

Wykes

Sturt

(1986). The measurement of social behavior in psychiatric patients: An assessment of the reliability and validity of the SBS schedule. British Journal of Psychiatry, 148, 1–11. doi:10.1192/bjp.148.1.1