- Email: [email protected]
The importance of cross-sectional remission in schizophrenia for long-term outcome: A clinical prospective study
Schizophrenia Research 115 (2009) 67–73
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
The importance of cross-sectional remission in schizophrenia for long-term outcome: A clinical prospective study☆ Lars Helldin a,c, John M. Kane b,⁎, Fredrik Hjärthag c, Torsten Norlander c a b c
The NU Health Care, Department of Psychiatry, SE-461 85 Trollhättan, Sweden The Zucker Hillside Hospital and the Albert Einstein College of Medicine, Glen Oaks, NY 11004, USA Department of Psychology, Karlstad University, SE-651 88 Karlstad, Sweden
a r t i c l e
i n f o
Article history: Received 13 March 2009 Received in revised form 29 June 2009 Accepted 6 July 2009 Available online 8 August 2009 Keywords: Remission Patients with schizophrenia Long-term outcome Resource utilization
a b s t r a c t Introduction: This study examines the relationship between having achieved cross-sectional remission and the need for future psychiatric and nursing home care. The study is a prospective long-term follow-up of patients with schizophrenia. Materials and methods: Cross-sectional remission was defined by applying the Positive and Negative Syndrome Scale (PANSS) criteria requiring that none of the eight core positive and negative symptom items are scored greater than mild. Patients are followed-up by yearly clinical examinations and medical record review. Information on consumption of healthcare resources and residency status were also gathered. Visits to mental health professionals, number and duration of inpatient psychiatric or nursing home admissions were also recorded. The patients are enrolled in a 12 year prospective study, the Clinical Long-term Investigation of Psychosis in Sweden (the CLIPS study). This report covers the first seven years. Results: Those patients who achieved cross-sectional remission at baseline had a lower total consumption of healthcare services than those who were not in remission. The latter group displayed higher values for all measured variables. Discussion: Our results show that cross-sectional remission is likely to be an important goal to achieve in order to reduce future treatment needs. Patients in remission live a more independent life and have better preconditions for functioning in society. © 2009 Published by Elsevier B.V.
1. Introduction Although antipsychotic drugs have been available for over 50 years, clinician goals for treatment are varied. Liberman et al. (2002a) proposed criteria for recovery in 2002, but these have not been widely implemented. Recently an attempt was made to define remission of schizophrenia based on the control of symptoms over a continuous extended time period. Two expert groups, one in the USA and one in Europe, worked towards a consensus which was published in 2005 and 2006 (Andreasen et al., 2005; van Os ☆ This study has been approved by the Ethical Research Committee at the University of Gothenburg, Sweden. ⁎ Corresponding author. E-mail address: [email protected] (J.M. Kane). 0920-9964/$ – see front matter © 2009 Published by Elsevier B.V. doi:10.1016/j.schres.2009.07.004
et al., 2006a). Earlier studies have investigated the relationship between cognitive ability and remission (Helldin et al., 2006), as well as the significance of remission for the individual's functional outcome (Helldin et al., 2007) and wellbeing (Lambert et al., 2006). van Os et al. (2006b) also reported that changes in healthcare organisation in the Netherlands has led to a higher likelihood of remission, demonstrating that the concept is also sensitive to changes in treatment practices over time. The remission criteria are based on eight items from PANSS, representing core symptoms in schizophrenia, and require that none of these items are rated greater than mild in severity. The cut-off limit was set at mild following expert consideration and allows the patient to have some symptoms, but not of such severity to impact day-to-day functioning. The full remission criteria require maintaining this threshold of symptom severity
68
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
continuously for six months. Some confusion has arisen surrounding the concept of remission as many studies reported are based only on cross-sectional severity. An expert meeting in Paris in December 2006 established that in cases where only symptom control is taken into account, this should be defined as ‘cross-sectional remission’ and that ‘remission’ should be reserved for studies that fulfil the criteria on both symptom control and duration. Follow-up studies have shown that patients with greater cognitive ability have a higher likelihood of achieving remission (Kopelowicz et al., 2005; Helldin et al., 2006). Patients who were in remission had a higher functional level in general, were functioning better in society (Helldin et al., 2007), their quality of life was higher and their illness burden milder (Helldin et al., 2008). To date, most studies have been based solely on symptom control and relatively brief, if any, follow up. There are very few long-term studies. One challenge in conducting long-term studies is the problem of ensuring that the patients have remained in remission continuously between different time points of assessment. The PANSS-scale assessment is based on symptoms present during the preceding week. Ideally, in order to be certain that the patient does not deviate from cross-sectional remission would require 26 weekly follow-up examinations every year. This is of course is unlikely in routine clinical practice. Instead the evaluation has to rely on reports from the patients themselves, relatives and caregivers, in combination with information from the medical records about requirements for changes in medication, symptom exacerbation, relapse or hospitalisation. The current report is based on long-term outcome data for a group of patients who have been followed since the year 2000. Our goal was to determine if patients who were examined during a non-acute period of their illness (representative of their highest level of symptom control and performance) and found to be in cross-sectional remission, would have a better long-term prognosis in terms of needing less care, i.e. number of visits to a mental health professional, number and duration of hospitalizations or admission to a nursing home, compared to patients not in cross-sectional remission. The hypothesis was that patients, who achieved cross-sectional remission, would have better outcome in terms of care needs. 2. Materials and methods 2.1. Participants The original purpose of the Clinical Long-term Investigation of Psychosis in Sweden (the CLIPS study) was to look at patients' function and adaptation to society. One of the objects was to investigate their cognitive ability. To be included in the study, patients had to be identified to be in a phase of their illness, where they should have been free from a relapse for at least the last six months. Otherwise it was expected that their cognitive ability could be impaired and not representative for their best performance. In fact, patients therefore were more likely to be free from a psychotic episode closer to one year rather than six months. One research-nurse identified which patients were relevant to include in the study by interviewing their case-managers and
investigating medical records at the different outpatient settings. Of the about 800 available patients, 550 were identified to match the criteria and were then invited to participate in the study. 300 patients accepted and were then tested. When their diagnoses had been controlled for with the DSM-IV decision-trees, 269 patients were found to be diagnosed with schizophrenia, schizoaffective disorder or delusional disorder. Also co-morbidity such as mental retardation, autistic disorder or dementia was checked for, excluding patients with double diagnoses. 242 patients (140 male and 102 female) were then remaining and completed all the instruments that were administrated in the study between 2000 and 2004. Of these patients, 30 individuals were diagnosed with delusional disorder, 10 with disorganized schizophrenia, 1 with catatonic schizophrenia, 79 with paranoid schizophrenia, 14 with residual schizophrenia, 50 with schizoaffective disorder and the remaining 58 individuals with undifferentiated schizophrenia. Of all patients included only 6 intermittently used narcotic substances (4 in remission and 2 in non-remission). 93 patients met the criteria for crosssectional remission while 149 did not. The patients in the study have then been followed up with annual assessments since 2005. Their consumption of healthcare and nursing home services has been recorded from their first day of assessment until 15 May 2008. Of the original group, 191 patients remained available for analysis in 2008. 21 patients had died, 7 of them were in remission (7.5% of the total number of patients in remission) while 14 were not (9.3% of the total number of patients not in remission). Data on the other patients was missing either because the patients had moved from the area or either because they had decided that they no longer wanted to participate in the study. There was no difference between those in remission and those not in terms of subsequent study refusal. In the remaining sample 72 were female and 119 were male. Fisher's Exact Test found no differences regarding gender between the remission and the non-remission groups, p = 0.21. The mean age at baseline for patients in remission was 46.9 years (standard deviation 11.6) and 47.6 years for the patients not in remission (standard deviation 12.6). Independent Samples T-test was not significant, p = 0.70. The observation time for patients who were in remission was 67.2 months (standard deviation 13.4) and 68.7 months for those who were not in remission (standard deviation 13.5). There were no differences between groups (Pearson Chi-Square Test was not significant, p = 0.38). 2.2. Design The patients have been followed-up since their first study evaluation with respect to contact with psychiatric specialist outpatient care, admission to psychiatric inpatient care and residential care in nursing homes. The number of visits to doctors and other caregivers (mainly nurses) as outpatients was recorded as number of visits, the amount of institutional care consumption as number of inpatient admissions, and the number of care days and days in sheltered care. In order to describe the total consumption of care, each outpatient visit has been re-calculated in hours, with each visit being assumed to have an average duration of 1 h. The number of
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
inpatient care admissions and the number of 24-hour sheltered care periods have also been re-calculated to give numbers of hours. The number of inpatient admissions has been re-calculated to give the equivalent number of hours as the preceding visit to the doctor, where the decision to admit overnight was taken, was not reported as an independent activity in our hospital statistics. One-hour per doctor visit in conjunction with an acute episode and admission to hospital is probably an underestimate, but provides a conservative bias. Remitted and non-remitted subjects were then compared with respect to their consumption of outpatient visits, number of inpatient admissions as well as the number of 24-hour care periods, the number of 24-hour periods in sheltered care and finally as a sum of their total care needs during the time period. The information about patient care interventions were extracted from their medical records and information about sheltered care was obtained via Strauss–Carpenter Scale evaluations. To increase the understanding of patients' status at baseline, PANSS-scores including cut-off levels on the eight core items for symptomatic remission is presented (Table 1). Descriptive data are also presented for positive, negative, general and total symptoms on PANSS (Table 2) for all patients possible to follow up over time. Also, to control for the impact of remission stability, three groups of remission instead of two were created and compared. The first group was patients in cross-sectional remission at baseline and without any admission to hospital during follow-up. The second group consisted of patients in cross-sectional remission at baseline but with admissions to hospital during follow-up. Finally, the third group were those not in cross-sectional remission at baseline. The three groups were compared on the different outcome variables with One-way ANOVAs with Bonferroni post hoc tests. Our intention with this step was to control for the stability of remission and if this way to present the phenomena would give more substantial information.
2.3. Instruments 2.3.1. Positive and negative syndrome scale—PANSS Subjects were interviewed using an adapted Swedish translation of the Structured Clinical Interview—Positive and Negative Syndrome Scale (SCI-PANSS) (Kay et al., 1987; Kay and Opler, 1987; Kay et al., 1988; Lindstrom et al., 1994).
69
2.3.2. Strauss–Carpenter (extended version) This scale (Strauss and Carpenter, 1972), with a total of five items, includes domains of living conditions, social contacts, working ability, number of days of sheltered care and hospital admission during the preceding 12 months. The first three questions each offer four response alternatives, whereas the responses to the final two questions represent the exact number of days spent in sheltered care and in psychiatric wards respectively. The reports are extracted directly from patient interviews, patient medical records and reports from relatives and caregivers. 2.4. Statistics The Fischer's Exact Test was utilized to examine gender distribution between the two groups, the Independent Samples T-Test to compare age at inclusion and the Pearson Chi-Square Test to examine the time of participation in the study. The potential differences between the groups in terms of care interventions and consumption were all analysed using the Independent Samples T-Test. Correlations (Pearson) were carried out for sums of PANSS positive symptoms, PANSS negative symptoms, PANSS general symptoms, PANSS total and the different outcome variables. Finally, after constructing three levels of symptom severity out of remission stability, i.e. “in remission at baseline and not further hospitalized”, “in remission at baseline but later hospitalized” and “not in remission at baseline”, the outcome variables were compared with One-way ANOVAs (Bonferroni post hoc) to detect differences between the three groups. 3. Procedure The study was carried out at the Department of Psychiatry, NU Health Care, during the period of November 2000 to May 2008. Patients were recruited through their case manager at the outpatient sites where they received treatment. Each patient was required to complete an “Informed Consent” agreement form before he/she could be included in the study. The study was approved by the Ethical Research Committee at the University of Gothenburg and was carried out in accordance with the latest version of the Helsinki Declaration. All participation was voluntary and patients could withdraw from the study whenever they wished, without having to provide a reason. Patients agreeing to participate in the
Table 1 Means, standard deviations and cut-off percentages for the eight remission items from PANSS at baseline. Core PANSS item for remission
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
Both groups together (n = 191)
Cut-off level for score 3 (n = 191)
M
SD
M
SD
M
SD
%
‘P1 Delusions’ ‘P2 Conceptual disorganization’ ‘P3 Hallucinatory behavior’ ‘N1 Blunted affect’ ‘N4 Passive/apathetic and social withdrawal’ ‘N6 Lack of spontaneity and flow of conversation’ ‘G5 Mannerisms and posturing’ ‘G9 Unusual thought content’
1.36 1.33 1.25 1.61 2.07 1.43 1.19 1.22
0.72 0.69 0.55 0.74 0.89 0.78 0.52 0.56
2.51 2.04 2.37 2.52 3.30 2.24 1.39 1.92
1.66 1.33 1.62 1.25 1.32 1.34 0.80 1.15
2.08 1.77 1.95 2.18 2.83 1.93 1.31 1.66
1.48 1.18 1.42 1.17 1.32 1.22 0.72 1.03
79.6 89.0 79.0 84.8 72.6 86.9 98.4 92.2
70
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
Table 2 Means and standard deviations for the PANSS domains at baseline. PANSS domain
Positive symptoms Negative symptoms General symptoms PANSS total sum
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
Both groups together (n = 191)
M
SD
M
SD
M
SD
9.12 12.58 27.18 48.80
2.52 3.68 6.45 9.97
14.01 18.44 33.48 65.65
5.33 4.70 6.98 12.78
12.17 16.22 31.12 59.29
5.06 5.19 7.43 14.34
investigation underwent tests and interviews that were not normally a part of their treatment. Patients were interviewed and their medical records were examined.
4.3. Total need of care The total consumption of the different care interventions; doctor visits, visits to other outpatient practitioners, the number of inpatient admissions, number of 24-hour care periods and 24-hour periods in nursing homes, (where the two latter categories are re-calculated as number of hours), shows an average consumption during the follow-up period. For the group of patients not in cross-sectional remission, the average need of care measured in days was higher compared to the patients in the cross-sectional remission group (Table 3).
4. Results 4.1. Psychiatric treatment Contact with specialised psychiatric outpatient care and the extent of inpatient care from baseline assessment up to 15th May 2008 is detailed in Table 3. This indicates the number of contacts with doctors and with other outpatient care personnel, the number of inpatient admissions and 24-hour care periods. During more than five and a half years of follow-up, 25 of the patients (35%) who entered into remission were admitted as an inpatient on one or more occasions, in comparison to 46 of the non-remitted subjects (46%). On average those in the former group were admitted as inpatients 0.8 times per patient compared to 1.4 times for the non-remission group. For each and every variable the care contacts are more numerous and longer, respectively, for the group not in remission, but these differences were not significant.
4.4. Correlations between PANSS domains and outcome variables The correlations (Pearson´s r) between PANSS domains and outcome variables (Table 4) showed a pattern similar to the above presented t-tests between patients in crosssectional remission and patients not in cross-sectional remission on the same outcome variables. The only difference in pattern was that days of hospitalisation correlated significantly with general symptoms and total score for the PANSS. No significant correlations occurred for positive or negative symptoms in this area.
4.2. Need of nursing homes
4.5. Influence of admissions added to the remission-concept
10 patients (14%) of those in the remission group have needed sheltered residential care during the follow-up period, compared to 34 (29%) of those not in remission; Pearson Chi-Square Test p = 0.02. The number of days in nursing home care for patients in the cross-sectional remission group was lower compared to the number of days for the patients not in cross-sectional remission (Table 3).
Patients grouped into three different groups based on remission stability (1: remission–no following admission, 2: remission–followed by admission, 3: no remission) were compared on the outcome variables (Table 5). Number of admissions and days at hospital turned by definition out to be different between the first group and the two others. Also, Bonferroni post hoc tests showed that
Table 3 Means, standard deviations and independent t-test results for both groups (cross-sectional remission/no cross-sectional remission) on the health care outcome variables, measured from the baseline assessment up to the 15th May 2008. Health care outcome variable
Psychiatric treatment
Days in nursing homes Total need of care (hours) a
Significant on the 5% level.
Outpatient visits to doctor Outpatient visits to other health care staff Number of admissions Number of hospital days
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
M
SD
M
SD
7.53 114.85 0.78 42.37 102.31 3506.7
3.91 122.63 1.62 102.78 357.59 8770.1
8.70 139.83 1.42 51.24 305.19 8616.4
6.06 185.73 3.14 121.27 546.94 13,644.0
t-value
p-value
− 1.62 − 1.12 − 1.86 − 0.54 − 3.10 − 3.13
0.107 0.265 0.064 0.592 0.002 a 0.002 a
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
71
Table 4 Correlations (Pearson's r) between the PANSS domains at baseline and the different health care outcome variables for the total sample (n = 191), p-values are presented in parentheses below.
PANSS Positive symptoms PANSS Negative symptoms PANSS General symptoms PANSS Total sum a b
Outpatient visits to doctor
Outpatient visits to other health care staff
Number of admissions
Number of hospital days
Days in nursing homes
Total need of care (hours)
0.122 (0.092) − 0.030 (0.679) 0.047 (0.523) 0.054 (0.462)
0.171 a 0.120 0.083 0.160 a
0.152 a (0.035) 0.030 (0.685) 0.104 (0.156) 0.112 (0.125)
0.073 (0.315) 0.141 (0.053) 0.021 (0.771) 0.092 (0.213)
0.166 a (0.021) 0.249 b (0.001) 0.242 b (0.001) 0.242 b (0.001)
0.175 a 0.280 b 0.246 b 0.259 b
(0.018) (0.099) (0.259) (0.029)
(0.016) (0.001) (0.001) (0.001)
Significant on the 5% level. Significant on the 1% level.
patients in group 2 ‘remission followed by admission’ had significantly more hospital days compared to them in group 3 ‘no remission’. Regarding the number of days in nursing homes, the no remission group had significantly needed more days compared to the patients in group 2 ‘remission–followed by admission’. Finally, according to the post hoc tests, the patients in the first group ‘remission–no admission’ had a significantly lower amount of hours in total need of care compared to the patients in the third group ‘no remission’. No other differences on the outcome variables were found. 5. Discussion With an average follow-up of more than five and a half years, these data suggest that not achieving cross-sectional remission is connected to an increased need for sheltered living, 29% (no cross-sectional remission) compared to 14% (cross-sectional remission) have been admitted to nursing home or other kinds of 24 h sheltered livings and the duration of staying were also longer. To investigate the patients' need of health care we analysed patients' use of outpatient care, i.e. number of visits to psychiatrist and other health personnel, number of admissions to hospital and number of days at hospital. The results did not indicate any significant differences even though all assessed areas showed higher scores for patients not in remission. We assumed that patients who are living in some sort of sheltered environment, due to the increased level of support compared to patients living on their own, would have a higher threshold for admission o hospital. To weight the different kinds of treatments and support, a final dimension for total need of care was constructed. Given that it is complicated to
compare a single visit at an outpatient setting with 24 h of support in a nursing home, and that day to day care in hospitals and nursing homes are more similar but differs in costs. We decided to describe this variable as total hours of given support no matter which institution or staff category gave the support. It showed that patients in remission had a lower need of the combination of treatment and housing support. To summarize, the value of achieving cross-sectional remission may be discussed. It is obvious that patients have a greater need for sheltered living but their need for psychiatric treatment compared to patients not in remission is still an open question. There were no major differences in the outcome patterns when the concept of cross-sectional remission was used, compared to when the patients' PANSS-scores were used as the basis for investigating the long-term outcomes. However, from a clinically point of view it is probably easier for the psychiatrist and the patient to understand, relate and aim for a threshold cut-off level, like that of remission. This supports the value of using the concept of remission in daily care giving. The analysis with patients divided into three groups based on remission stability, did not yield any new information compared to using the two original groups of cross-sectional remission and not cross-sectional remission. The only new difference found was that the remission patients not admitted to hospital during follow up had less inpatient treatment. As the remission groups were constructed based on not being admitted to hospital this difference was expected. However, future studies with more precise ways of measuring the remission stability could contribute significantly to the knowledge within this question, so this question should be further addressed in the future.
Table 5 Descriptive statistics and results from one-way ANOVAs on the health care outcome variables when the sample is divided into the three remission groups. Health care outcome variable
Psychiatric treatment
Outpatient visits to doctor Outpatient visits to other health care staff Number of admissions Number of hospital days
Days in nursing homes Total need of care (hours) a b
Significant on the 5% level. Significant on the 1% level.
Group in cross-sectional remission at baseline and no later admissions (n = 46)
Group in cross-sectional remission at baseline and admissions followed (n = 26)
Group not in cross-sectional remission at baseline (n = 119)
One-way ANOVA
M
M
M
F
SD
SD
SD
p
6.91 102.47
3.90 102.26
8.96 131.74
4.42 149.33
8.63 141.33
5.99 186.96
2.01 0.93
0.137 0.396
0 0 125.74 3083.48
0 0 411.47 9974.04
2.26 119.96 53.93 4223.08
2.12 142.13 216.83 5954.80
1.40 50.73 310.41 8667.28
3.14 121.66 550.16 13691.41
7.64 10.11 4.48 4.17
b0.001 b b0.001 b 0.013 a 0.017 a
72
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
In the short term however, a possible conclusion is that meeting the criteria of symptomatic remission indicates a better prognosis and is, therefore, a condition that is important to achieve when treating patients with schizophrenia. Especially important is this since remission could be linked to patients living a more independent life and experiencing a higher degree of adaptation to society. The study also highlights the challenges in determining a continuous six month duration of remission in routine clinical practice, as would be required to meet the full Andreasen criteria. Achieving remission is important because the presence of psychotic and negative symptoms reduces already compromised cognitive abilities which in turn can further negatively affect both judgement and ability. The ability of patients to accept and benefit from their treatments can become increasingly impaired. Our previous study showed that patients who achieve remission have a higher level of awareness of their illness, both in terms of experience and understanding of symptoms (Helldin et al., 2008). An increase in symptoms can impair insight and reduce awareness of the illness leading to more difficulty in seeking and accepting necessary treatment with an accompanying increased risk of relapse. Both of these areas subsequently contribute in turn to the quality of the patient's compliance (Smith et al., 1999). Low or reduced neurocognitive abilities mean that even a patient who is motivated to take their medication often forgets to do so (Green, 1996; Jeste et al., 2003). Previous studies demonstrate that even a relatively marginal reduction in adherence can increase the risk of relapse and, if the antipsychotic medication is omitted completely, the risk of relapse is very high (Weiden et al., 2004). This study also indirectly highlights the importance of attempting to achieve remission in those patients who with current treatment paradigms fail to do so. While monitoring the patients' treatment from a remission perspective is gradually developing into a clinical practice, more resources must also be steered towards patients with low remission potential. Initially, these patients need to be identified through symptomatic and neuropsychological examinations at an early stage of the illness, followed by targeted efforts to enhance the likelihood of remission (Helldin et al., 2006). In addition to assessment measures, the choice of drug and drug formulation (Barnes and Curson, 1994; Fleischhacker et al., 2003) should also be considered. Long-acting injectable formulations are one approach to improving compliance (O'Donnell et al., 2003), and others, including community orientated treatment programmes with case management models (Malla et al., 2002) should be considered. Psychoeducational training, including activity-stimulating checklists, improves the possibilities for patients and close relatives to receive successful care (Medalia et al., 2002; Liberman et al., 2002b; McFarlane et al., 1995). Finally, there are also data suggesting that patients' cognitive abilities can be improved by systemised training. Both the ability to solve social problems (Kern et al., 2005) and executive function (Bell et al., 2001; Kern et al., 2002) can be improved. There are several potentially effective methods for cognitive training (Twamley et al., 2003). In summary, this study was a prospective follow-up of patients with and without cross-sectional remission. We have found that the requirement for nursing home care and shel-
tered living are significantly higher for the group of patients who not were in cross-sectional remission while the total needs of healthcare interventions were lower among patients being in cross-sectional remission. We have also found that short of achieving the full remission criteria (including duration) the cross-sectional severity criteria for remission is an important and highly relevant treatment goal. When this straightforward, dual categorisation of patients that is crosssectional remission, were compared with the more extensive PANSS-scores, or if admissions to hospital are added to being in remission, no different patterns in outcomes could be seen. This observation may serve as a validation of the remissionconcept and an argument for the use of this thresholdmethod, that remission represents, offering patients and psychiatrists a solid goal for the treatment. A limitation of the study was the large variances in measures as an expression for a high variability inside the two groups. This may be relevant given the heterogeneity and lack of differences in the context of a remitted/non-remitted classification. Another shortcoming is that it has not been possible to follow the remission status of the patients with adequate frequency in this type of naturalistic study to determine the duration of remission. Strategies should be tested for monitoring time in remission for those patients seen infrequently. However, it is likely that many patients are seen by some health care provider at least every one to two months and adaptation of the PANSS items to assess longer intervals than one week would be appropriate in that context. Role of funding source This study was sponsored by Janssen-Cilag, Sweden with unrestricted grants. Contributors There were no other contributors than the department of psychiatry, NU Health Care Hospital, Trollhättan Sweden. Conflicts of interest There were no conflicts of interest. Acknowledgements This study was supported by unrestricted grants from Janssen-Cilag, Sollentuna, Sweden. We gratefully acknowledge the excellent technical assistance of Britt-Marie Hansson, Ruth Johansson and Maivor Olsson.
References Andreasen, N.C., Carpenter, W., Kane, J.M., Lasser, R.A., Marder, S.R., Weinberger, D.R., 2005. Remission in schizophrenia: proposed criteria and rationale for consensus. The Am. J. Psychiatry. 162 (3), 441–449. Barnes, T.R., Curson, D.A., 1994. Long-term depot antipsychotics. A risk– benefit assessment. Drug Safety. 10 (6), 464–479. Bell, M., Bryson, G., Greig, T., Corcoran, C., Wexler, B.E., 2001. Neurocognitive enhancement therapy with work therapy: effects on neuropsychological test performance. Arch. Gen. Psychiatry. 58 (8), 763–768. Fleischhacker, W.W., Eerdekens, M., Karcher, K., Remington, G., Llorca, P.M., Chrzanowski, W., Martin, S., Gefvert, O., 2003. Treatment of schizophrenia with long-acting injectable risperidone: a 12-month open-label trial of the first long-acting second-generation antipsychotic. J. Clin. Psychiatry. 64 (10), 1250–1257. Green, M.F., 1996. What are the functional consequences of neurocognitive deficits in schizophrenia? Am. J. Psychiatry. 153, 321–330. Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2006. Remission and cognitive ability in a cohort of patients with schizophrenia. Psychiatric Res. 40, 738–745. Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2007. Remission in prognosis of functional outcome: a new dimension in the treatment of patients with psychotic disorder. Schizophr Res. 93, 160–168.
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73 Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2008. Experience of quality of life and attitude to care and treatment in patients with schizophrenia: role of cross-sectional remission. Int. J. Psychiatr. Clinical Pract. 1–8. Jeste, S.D., Patterson, T.L., Palmer, B.W., Dolder, C.R., Goldman, S., Jeste, D.V., 2003. Cognitive predictors of medication adherence among middle-aged and older outpatients with schizophrenia. Schizophr. Res. 63 (1–2), 49–58. Kay, S.R., Fitzbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261–267. Kay, S.R., Opler, L.A., 1987. The positive–negative dimension in schizophrenia: its validity and significance. Psychiatr. Dev. 2, 79–103. Kay, S.R., Opler, L.A., Lindenmayer, J.P., 1988. Reliability and validity of the positive and negative syndrome scale for schizophrenia. Psychatr. Res. 23, 99–110. Kern, R.S., Liberman, R.P., Kopelowicz, A., Mintz, J., Green, M.F., 2002. Applications of errorless learning for improving work performance in persons with schizophrenia. Am. J. Psychiatry. 159 (11), 1921–1926. Kern, R.S., Green, M.F., Mitchell, S., Kopelowicz, A., Mintz, J., Liberman, R.P., 2005. Extensions of errorless learning for social problem-solving deficits in schizophrenia. Am. J. Psychiatry. 162 (3), 513–519. Kopelowicz, A., Liberman, R.P., Ventura, J., Zarate, R., Mintz, J., 2005. Neurocognitive correlates of recovery from schizophrenia. Psychol. Med. 35 (8), 1165–1173. Lambert, M., Schimmelmann, B.G., Naber, D., Schacht, A., Karow, A., Wagner, T., Czekalla, J., 2006. Prediction of remission as a combination of symptomatic and functional remission and adequate subjective well-being in 2960 patients with schizophrenia. J. Clin. Psychiatry. 67, 1690–1697. Liberman, R.P., Kopelowicz, A., Venture, J., Gutkind, D., 2002a. Operational criteria and factors related to recovery from schizophrenia. Int. Rev. Psychiatry. 14, 256–272. Liberman, R.P., Glynn, S., Blair, K.E., Ross, D., Marder, S.R., 2002b. In vivo amplified skills training: promoting generalization of independent living skills for clients with schizophrenia. Psychiatry. 65 (2), 137–155.
73
Lindstrom, E., Wieselgren, I.M., von Knorring, L., 1994. Interrater reliability of the structured clinical interview for the positive and negative syndrome scale for schizophrenia. Acta. Psychiatr. Scand. 89 (3), 192–195. Malla, A.K., Norman, R.M., Manchanda, R., McLean, T.S., Harricharan, R., Cortese, L., Townsend, L.A., Scholten, D.J., 2002. Status of patients with first-episode psychosis after one year of phase-specific communityoriented treatment. Psychiatric Services. 53 (4), 458–463. McFarlane, W.R., Link, B., Dushay, R., Marchal, J., Crilly, J., 1995. Psychoeducational multiple family groups: four-year relapse outcome in schizophrenia. Family Process. 34 (2), 127–144. Medalia, A., Revheim, N., Casey, M., 2002. Remediation of problem-solving skills in schizophrenia: evidence of a persistent effect. Schizophr. Res. 57, 165–171. O'Donnell, C., Donohoe, G., Sharkey, L., Owens, N., Migone, M., Harries, R., Kinsella, A., Larkin, C., O'Callaghan, E., 2003. Compliance therapy: a randomised controlled trial in schizophrenia. British Medical Jour. 11 327 (7419), 834. Smith, T.E., Hull, J.W., Goodman, M., Heayat-Harris, A., Willson, D.F., Israel, L.M., Munich, R.L., 1999. The relative influences of symptoms, insight, and neurocognition on social adjustment in schizophrenia and schizoaffective disorder. J. Nerv. Ment. Dis. 187 (2), 102–108. Strauss, J.S., Carpenter, W.T., 1972. The prediction of outcome in schizophrenia I: characteristics of outcome. Arch. Gen. Psychiatry. 27, 739–746. Twamley, E.W., Jeste, D.V., Bellack, A.S., 2003. A review of cognitive training in schizophrenia. Schizophr. Bull. 29 (2), 359–382. van Os, J., Burns, T., Cavallaro, R., Leucht, S., Peuskens, J., Helldin, L., Bernardo, M., Arango, C., Fleischhacker, W., Lachaux, B., Kane, J.M., 2006a. Standardized remission criteria in schizophrenia. Acta. Psychiatr. Scand. 91–95. van Os, J., Drukker, M., Ã. Campo, J., Meijer, J., Bak, M., Delespaul, P., 2006b. Validation of remission criteria for schizophrenia. Am. J. Psychiatry.163 (11), 2000–2002. Weiden, P.J., Kozma, C., Grogg, A., Locklear, J., 2004. Partial compliance and risk of rehospitalization among California Medicaid patients with schizophrenia. Psychiatr. Serv. 55 (8), 886–891.
Contents lists available at ScienceDirect
Schizophrenia Research j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m / l o c a t e / s c h r e s
The importance of cross-sectional remission in schizophrenia for long-term outcome: A clinical prospective study☆ Lars Helldin a,c, John M. Kane b,⁎, Fredrik Hjärthag c, Torsten Norlander c a b c
The NU Health Care, Department of Psychiatry, SE-461 85 Trollhättan, Sweden The Zucker Hillside Hospital and the Albert Einstein College of Medicine, Glen Oaks, NY 11004, USA Department of Psychology, Karlstad University, SE-651 88 Karlstad, Sweden
a r t i c l e
i n f o
Article history: Received 13 March 2009 Received in revised form 29 June 2009 Accepted 6 July 2009 Available online 8 August 2009 Keywords: Remission Patients with schizophrenia Long-term outcome Resource utilization
a b s t r a c t Introduction: This study examines the relationship between having achieved cross-sectional remission and the need for future psychiatric and nursing home care. The study is a prospective long-term follow-up of patients with schizophrenia. Materials and methods: Cross-sectional remission was defined by applying the Positive and Negative Syndrome Scale (PANSS) criteria requiring that none of the eight core positive and negative symptom items are scored greater than mild. Patients are followed-up by yearly clinical examinations and medical record review. Information on consumption of healthcare resources and residency status were also gathered. Visits to mental health professionals, number and duration of inpatient psychiatric or nursing home admissions were also recorded. The patients are enrolled in a 12 year prospective study, the Clinical Long-term Investigation of Psychosis in Sweden (the CLIPS study). This report covers the first seven years. Results: Those patients who achieved cross-sectional remission at baseline had a lower total consumption of healthcare services than those who were not in remission. The latter group displayed higher values for all measured variables. Discussion: Our results show that cross-sectional remission is likely to be an important goal to achieve in order to reduce future treatment needs. Patients in remission live a more independent life and have better preconditions for functioning in society. © 2009 Published by Elsevier B.V.
1. Introduction Although antipsychotic drugs have been available for over 50 years, clinician goals for treatment are varied. Liberman et al. (2002a) proposed criteria for recovery in 2002, but these have not been widely implemented. Recently an attempt was made to define remission of schizophrenia based on the control of symptoms over a continuous extended time period. Two expert groups, one in the USA and one in Europe, worked towards a consensus which was published in 2005 and 2006 (Andreasen et al., 2005; van Os ☆ This study has been approved by the Ethical Research Committee at the University of Gothenburg, Sweden. ⁎ Corresponding author. E-mail address: [email protected] (J.M. Kane). 0920-9964/$ – see front matter © 2009 Published by Elsevier B.V. doi:10.1016/j.schres.2009.07.004
et al., 2006a). Earlier studies have investigated the relationship between cognitive ability and remission (Helldin et al., 2006), as well as the significance of remission for the individual's functional outcome (Helldin et al., 2007) and wellbeing (Lambert et al., 2006). van Os et al. (2006b) also reported that changes in healthcare organisation in the Netherlands has led to a higher likelihood of remission, demonstrating that the concept is also sensitive to changes in treatment practices over time. The remission criteria are based on eight items from PANSS, representing core symptoms in schizophrenia, and require that none of these items are rated greater than mild in severity. The cut-off limit was set at mild following expert consideration and allows the patient to have some symptoms, but not of such severity to impact day-to-day functioning. The full remission criteria require maintaining this threshold of symptom severity
68
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
continuously for six months. Some confusion has arisen surrounding the concept of remission as many studies reported are based only on cross-sectional severity. An expert meeting in Paris in December 2006 established that in cases where only symptom control is taken into account, this should be defined as ‘cross-sectional remission’ and that ‘remission’ should be reserved for studies that fulfil the criteria on both symptom control and duration. Follow-up studies have shown that patients with greater cognitive ability have a higher likelihood of achieving remission (Kopelowicz et al., 2005; Helldin et al., 2006). Patients who were in remission had a higher functional level in general, were functioning better in society (Helldin et al., 2007), their quality of life was higher and their illness burden milder (Helldin et al., 2008). To date, most studies have been based solely on symptom control and relatively brief, if any, follow up. There are very few long-term studies. One challenge in conducting long-term studies is the problem of ensuring that the patients have remained in remission continuously between different time points of assessment. The PANSS-scale assessment is based on symptoms present during the preceding week. Ideally, in order to be certain that the patient does not deviate from cross-sectional remission would require 26 weekly follow-up examinations every year. This is of course is unlikely in routine clinical practice. Instead the evaluation has to rely on reports from the patients themselves, relatives and caregivers, in combination with information from the medical records about requirements for changes in medication, symptom exacerbation, relapse or hospitalisation. The current report is based on long-term outcome data for a group of patients who have been followed since the year 2000. Our goal was to determine if patients who were examined during a non-acute period of their illness (representative of their highest level of symptom control and performance) and found to be in cross-sectional remission, would have a better long-term prognosis in terms of needing less care, i.e. number of visits to a mental health professional, number and duration of hospitalizations or admission to a nursing home, compared to patients not in cross-sectional remission. The hypothesis was that patients, who achieved cross-sectional remission, would have better outcome in terms of care needs. 2. Materials and methods 2.1. Participants The original purpose of the Clinical Long-term Investigation of Psychosis in Sweden (the CLIPS study) was to look at patients' function and adaptation to society. One of the objects was to investigate their cognitive ability. To be included in the study, patients had to be identified to be in a phase of their illness, where they should have been free from a relapse for at least the last six months. Otherwise it was expected that their cognitive ability could be impaired and not representative for their best performance. In fact, patients therefore were more likely to be free from a psychotic episode closer to one year rather than six months. One research-nurse identified which patients were relevant to include in the study by interviewing their case-managers and
investigating medical records at the different outpatient settings. Of the about 800 available patients, 550 were identified to match the criteria and were then invited to participate in the study. 300 patients accepted and were then tested. When their diagnoses had been controlled for with the DSM-IV decision-trees, 269 patients were found to be diagnosed with schizophrenia, schizoaffective disorder or delusional disorder. Also co-morbidity such as mental retardation, autistic disorder or dementia was checked for, excluding patients with double diagnoses. 242 patients (140 male and 102 female) were then remaining and completed all the instruments that were administrated in the study between 2000 and 2004. Of these patients, 30 individuals were diagnosed with delusional disorder, 10 with disorganized schizophrenia, 1 with catatonic schizophrenia, 79 with paranoid schizophrenia, 14 with residual schizophrenia, 50 with schizoaffective disorder and the remaining 58 individuals with undifferentiated schizophrenia. Of all patients included only 6 intermittently used narcotic substances (4 in remission and 2 in non-remission). 93 patients met the criteria for crosssectional remission while 149 did not. The patients in the study have then been followed up with annual assessments since 2005. Their consumption of healthcare and nursing home services has been recorded from their first day of assessment until 15 May 2008. Of the original group, 191 patients remained available for analysis in 2008. 21 patients had died, 7 of them were in remission (7.5% of the total number of patients in remission) while 14 were not (9.3% of the total number of patients not in remission). Data on the other patients was missing either because the patients had moved from the area or either because they had decided that they no longer wanted to participate in the study. There was no difference between those in remission and those not in terms of subsequent study refusal. In the remaining sample 72 were female and 119 were male. Fisher's Exact Test found no differences regarding gender between the remission and the non-remission groups, p = 0.21. The mean age at baseline for patients in remission was 46.9 years (standard deviation 11.6) and 47.6 years for the patients not in remission (standard deviation 12.6). Independent Samples T-test was not significant, p = 0.70. The observation time for patients who were in remission was 67.2 months (standard deviation 13.4) and 68.7 months for those who were not in remission (standard deviation 13.5). There were no differences between groups (Pearson Chi-Square Test was not significant, p = 0.38). 2.2. Design The patients have been followed-up since their first study evaluation with respect to contact with psychiatric specialist outpatient care, admission to psychiatric inpatient care and residential care in nursing homes. The number of visits to doctors and other caregivers (mainly nurses) as outpatients was recorded as number of visits, the amount of institutional care consumption as number of inpatient admissions, and the number of care days and days in sheltered care. In order to describe the total consumption of care, each outpatient visit has been re-calculated in hours, with each visit being assumed to have an average duration of 1 h. The number of
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
inpatient care admissions and the number of 24-hour sheltered care periods have also been re-calculated to give numbers of hours. The number of inpatient admissions has been re-calculated to give the equivalent number of hours as the preceding visit to the doctor, where the decision to admit overnight was taken, was not reported as an independent activity in our hospital statistics. One-hour per doctor visit in conjunction with an acute episode and admission to hospital is probably an underestimate, but provides a conservative bias. Remitted and non-remitted subjects were then compared with respect to their consumption of outpatient visits, number of inpatient admissions as well as the number of 24-hour care periods, the number of 24-hour periods in sheltered care and finally as a sum of their total care needs during the time period. The information about patient care interventions were extracted from their medical records and information about sheltered care was obtained via Strauss–Carpenter Scale evaluations. To increase the understanding of patients' status at baseline, PANSS-scores including cut-off levels on the eight core items for symptomatic remission is presented (Table 1). Descriptive data are also presented for positive, negative, general and total symptoms on PANSS (Table 2) for all patients possible to follow up over time. Also, to control for the impact of remission stability, three groups of remission instead of two were created and compared. The first group was patients in cross-sectional remission at baseline and without any admission to hospital during follow-up. The second group consisted of patients in cross-sectional remission at baseline but with admissions to hospital during follow-up. Finally, the third group were those not in cross-sectional remission at baseline. The three groups were compared on the different outcome variables with One-way ANOVAs with Bonferroni post hoc tests. Our intention with this step was to control for the stability of remission and if this way to present the phenomena would give more substantial information.
2.3. Instruments 2.3.1. Positive and negative syndrome scale—PANSS Subjects were interviewed using an adapted Swedish translation of the Structured Clinical Interview—Positive and Negative Syndrome Scale (SCI-PANSS) (Kay et al., 1987; Kay and Opler, 1987; Kay et al., 1988; Lindstrom et al., 1994).
69
2.3.2. Strauss–Carpenter (extended version) This scale (Strauss and Carpenter, 1972), with a total of five items, includes domains of living conditions, social contacts, working ability, number of days of sheltered care and hospital admission during the preceding 12 months. The first three questions each offer four response alternatives, whereas the responses to the final two questions represent the exact number of days spent in sheltered care and in psychiatric wards respectively. The reports are extracted directly from patient interviews, patient medical records and reports from relatives and caregivers. 2.4. Statistics The Fischer's Exact Test was utilized to examine gender distribution between the two groups, the Independent Samples T-Test to compare age at inclusion and the Pearson Chi-Square Test to examine the time of participation in the study. The potential differences between the groups in terms of care interventions and consumption were all analysed using the Independent Samples T-Test. Correlations (Pearson) were carried out for sums of PANSS positive symptoms, PANSS negative symptoms, PANSS general symptoms, PANSS total and the different outcome variables. Finally, after constructing three levels of symptom severity out of remission stability, i.e. “in remission at baseline and not further hospitalized”, “in remission at baseline but later hospitalized” and “not in remission at baseline”, the outcome variables were compared with One-way ANOVAs (Bonferroni post hoc) to detect differences between the three groups. 3. Procedure The study was carried out at the Department of Psychiatry, NU Health Care, during the period of November 2000 to May 2008. Patients were recruited through their case manager at the outpatient sites where they received treatment. Each patient was required to complete an “Informed Consent” agreement form before he/she could be included in the study. The study was approved by the Ethical Research Committee at the University of Gothenburg and was carried out in accordance with the latest version of the Helsinki Declaration. All participation was voluntary and patients could withdraw from the study whenever they wished, without having to provide a reason. Patients agreeing to participate in the
Table 1 Means, standard deviations and cut-off percentages for the eight remission items from PANSS at baseline. Core PANSS item for remission
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
Both groups together (n = 191)
Cut-off level for score 3 (n = 191)
M
SD
M
SD
M
SD
%
‘P1 Delusions’ ‘P2 Conceptual disorganization’ ‘P3 Hallucinatory behavior’ ‘N1 Blunted affect’ ‘N4 Passive/apathetic and social withdrawal’ ‘N6 Lack of spontaneity and flow of conversation’ ‘G5 Mannerisms and posturing’ ‘G9 Unusual thought content’
1.36 1.33 1.25 1.61 2.07 1.43 1.19 1.22
0.72 0.69 0.55 0.74 0.89 0.78 0.52 0.56
2.51 2.04 2.37 2.52 3.30 2.24 1.39 1.92
1.66 1.33 1.62 1.25 1.32 1.34 0.80 1.15
2.08 1.77 1.95 2.18 2.83 1.93 1.31 1.66
1.48 1.18 1.42 1.17 1.32 1.22 0.72 1.03
79.6 89.0 79.0 84.8 72.6 86.9 98.4 92.2
70
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
Table 2 Means and standard deviations for the PANSS domains at baseline. PANSS domain
Positive symptoms Negative symptoms General symptoms PANSS total sum
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
Both groups together (n = 191)
M
SD
M
SD
M
SD
9.12 12.58 27.18 48.80
2.52 3.68 6.45 9.97
14.01 18.44 33.48 65.65
5.33 4.70 6.98 12.78
12.17 16.22 31.12 59.29
5.06 5.19 7.43 14.34
investigation underwent tests and interviews that were not normally a part of their treatment. Patients were interviewed and their medical records were examined.
4.3. Total need of care The total consumption of the different care interventions; doctor visits, visits to other outpatient practitioners, the number of inpatient admissions, number of 24-hour care periods and 24-hour periods in nursing homes, (where the two latter categories are re-calculated as number of hours), shows an average consumption during the follow-up period. For the group of patients not in cross-sectional remission, the average need of care measured in days was higher compared to the patients in the cross-sectional remission group (Table 3).
4. Results 4.1. Psychiatric treatment Contact with specialised psychiatric outpatient care and the extent of inpatient care from baseline assessment up to 15th May 2008 is detailed in Table 3. This indicates the number of contacts with doctors and with other outpatient care personnel, the number of inpatient admissions and 24-hour care periods. During more than five and a half years of follow-up, 25 of the patients (35%) who entered into remission were admitted as an inpatient on one or more occasions, in comparison to 46 of the non-remitted subjects (46%). On average those in the former group were admitted as inpatients 0.8 times per patient compared to 1.4 times for the non-remission group. For each and every variable the care contacts are more numerous and longer, respectively, for the group not in remission, but these differences were not significant.
4.4. Correlations between PANSS domains and outcome variables The correlations (Pearson´s r) between PANSS domains and outcome variables (Table 4) showed a pattern similar to the above presented t-tests between patients in crosssectional remission and patients not in cross-sectional remission on the same outcome variables. The only difference in pattern was that days of hospitalisation correlated significantly with general symptoms and total score for the PANSS. No significant correlations occurred for positive or negative symptoms in this area.
4.2. Need of nursing homes
4.5. Influence of admissions added to the remission-concept
10 patients (14%) of those in the remission group have needed sheltered residential care during the follow-up period, compared to 34 (29%) of those not in remission; Pearson Chi-Square Test p = 0.02. The number of days in nursing home care for patients in the cross-sectional remission group was lower compared to the number of days for the patients not in cross-sectional remission (Table 3).
Patients grouped into three different groups based on remission stability (1: remission–no following admission, 2: remission–followed by admission, 3: no remission) were compared on the outcome variables (Table 5). Number of admissions and days at hospital turned by definition out to be different between the first group and the two others. Also, Bonferroni post hoc tests showed that
Table 3 Means, standard deviations and independent t-test results for both groups (cross-sectional remission/no cross-sectional remission) on the health care outcome variables, measured from the baseline assessment up to the 15th May 2008. Health care outcome variable
Psychiatric treatment
Days in nursing homes Total need of care (hours) a
Significant on the 5% level.
Outpatient visits to doctor Outpatient visits to other health care staff Number of admissions Number of hospital days
Group in cross-sectional remission (n = 72)
Group not in cross-sectional remission (n = 119)
M
SD
M
SD
7.53 114.85 0.78 42.37 102.31 3506.7
3.91 122.63 1.62 102.78 357.59 8770.1
8.70 139.83 1.42 51.24 305.19 8616.4
6.06 185.73 3.14 121.27 546.94 13,644.0
t-value
p-value
− 1.62 − 1.12 − 1.86 − 0.54 − 3.10 − 3.13
0.107 0.265 0.064 0.592 0.002 a 0.002 a
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
71
Table 4 Correlations (Pearson's r) between the PANSS domains at baseline and the different health care outcome variables for the total sample (n = 191), p-values are presented in parentheses below.
PANSS Positive symptoms PANSS Negative symptoms PANSS General symptoms PANSS Total sum a b
Outpatient visits to doctor
Outpatient visits to other health care staff
Number of admissions
Number of hospital days
Days in nursing homes
Total need of care (hours)
0.122 (0.092) − 0.030 (0.679) 0.047 (0.523) 0.054 (0.462)
0.171 a 0.120 0.083 0.160 a
0.152 a (0.035) 0.030 (0.685) 0.104 (0.156) 0.112 (0.125)
0.073 (0.315) 0.141 (0.053) 0.021 (0.771) 0.092 (0.213)
0.166 a (0.021) 0.249 b (0.001) 0.242 b (0.001) 0.242 b (0.001)
0.175 a 0.280 b 0.246 b 0.259 b
(0.018) (0.099) (0.259) (0.029)
(0.016) (0.001) (0.001) (0.001)
Significant on the 5% level. Significant on the 1% level.
patients in group 2 ‘remission followed by admission’ had significantly more hospital days compared to them in group 3 ‘no remission’. Regarding the number of days in nursing homes, the no remission group had significantly needed more days compared to the patients in group 2 ‘remission–followed by admission’. Finally, according to the post hoc tests, the patients in the first group ‘remission–no admission’ had a significantly lower amount of hours in total need of care compared to the patients in the third group ‘no remission’. No other differences on the outcome variables were found. 5. Discussion With an average follow-up of more than five and a half years, these data suggest that not achieving cross-sectional remission is connected to an increased need for sheltered living, 29% (no cross-sectional remission) compared to 14% (cross-sectional remission) have been admitted to nursing home or other kinds of 24 h sheltered livings and the duration of staying were also longer. To investigate the patients' need of health care we analysed patients' use of outpatient care, i.e. number of visits to psychiatrist and other health personnel, number of admissions to hospital and number of days at hospital. The results did not indicate any significant differences even though all assessed areas showed higher scores for patients not in remission. We assumed that patients who are living in some sort of sheltered environment, due to the increased level of support compared to patients living on their own, would have a higher threshold for admission o hospital. To weight the different kinds of treatments and support, a final dimension for total need of care was constructed. Given that it is complicated to
compare a single visit at an outpatient setting with 24 h of support in a nursing home, and that day to day care in hospitals and nursing homes are more similar but differs in costs. We decided to describe this variable as total hours of given support no matter which institution or staff category gave the support. It showed that patients in remission had a lower need of the combination of treatment and housing support. To summarize, the value of achieving cross-sectional remission may be discussed. It is obvious that patients have a greater need for sheltered living but their need for psychiatric treatment compared to patients not in remission is still an open question. There were no major differences in the outcome patterns when the concept of cross-sectional remission was used, compared to when the patients' PANSS-scores were used as the basis for investigating the long-term outcomes. However, from a clinically point of view it is probably easier for the psychiatrist and the patient to understand, relate and aim for a threshold cut-off level, like that of remission. This supports the value of using the concept of remission in daily care giving. The analysis with patients divided into three groups based on remission stability, did not yield any new information compared to using the two original groups of cross-sectional remission and not cross-sectional remission. The only new difference found was that the remission patients not admitted to hospital during follow up had less inpatient treatment. As the remission groups were constructed based on not being admitted to hospital this difference was expected. However, future studies with more precise ways of measuring the remission stability could contribute significantly to the knowledge within this question, so this question should be further addressed in the future.
Table 5 Descriptive statistics and results from one-way ANOVAs on the health care outcome variables when the sample is divided into the three remission groups. Health care outcome variable
Psychiatric treatment
Outpatient visits to doctor Outpatient visits to other health care staff Number of admissions Number of hospital days
Days in nursing homes Total need of care (hours) a b
Significant on the 5% level. Significant on the 1% level.
Group in cross-sectional remission at baseline and no later admissions (n = 46)
Group in cross-sectional remission at baseline and admissions followed (n = 26)
Group not in cross-sectional remission at baseline (n = 119)
One-way ANOVA
M
M
M
F
SD
SD
SD
p
6.91 102.47
3.90 102.26
8.96 131.74
4.42 149.33
8.63 141.33
5.99 186.96
2.01 0.93
0.137 0.396
0 0 125.74 3083.48
0 0 411.47 9974.04
2.26 119.96 53.93 4223.08
2.12 142.13 216.83 5954.80
1.40 50.73 310.41 8667.28
3.14 121.66 550.16 13691.41
7.64 10.11 4.48 4.17
b0.001 b b0.001 b 0.013 a 0.017 a
72
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73
In the short term however, a possible conclusion is that meeting the criteria of symptomatic remission indicates a better prognosis and is, therefore, a condition that is important to achieve when treating patients with schizophrenia. Especially important is this since remission could be linked to patients living a more independent life and experiencing a higher degree of adaptation to society. The study also highlights the challenges in determining a continuous six month duration of remission in routine clinical practice, as would be required to meet the full Andreasen criteria. Achieving remission is important because the presence of psychotic and negative symptoms reduces already compromised cognitive abilities which in turn can further negatively affect both judgement and ability. The ability of patients to accept and benefit from their treatments can become increasingly impaired. Our previous study showed that patients who achieve remission have a higher level of awareness of their illness, both in terms of experience and understanding of symptoms (Helldin et al., 2008). An increase in symptoms can impair insight and reduce awareness of the illness leading to more difficulty in seeking and accepting necessary treatment with an accompanying increased risk of relapse. Both of these areas subsequently contribute in turn to the quality of the patient's compliance (Smith et al., 1999). Low or reduced neurocognitive abilities mean that even a patient who is motivated to take their medication often forgets to do so (Green, 1996; Jeste et al., 2003). Previous studies demonstrate that even a relatively marginal reduction in adherence can increase the risk of relapse and, if the antipsychotic medication is omitted completely, the risk of relapse is very high (Weiden et al., 2004). This study also indirectly highlights the importance of attempting to achieve remission in those patients who with current treatment paradigms fail to do so. While monitoring the patients' treatment from a remission perspective is gradually developing into a clinical practice, more resources must also be steered towards patients with low remission potential. Initially, these patients need to be identified through symptomatic and neuropsychological examinations at an early stage of the illness, followed by targeted efforts to enhance the likelihood of remission (Helldin et al., 2006). In addition to assessment measures, the choice of drug and drug formulation (Barnes and Curson, 1994; Fleischhacker et al., 2003) should also be considered. Long-acting injectable formulations are one approach to improving compliance (O'Donnell et al., 2003), and others, including community orientated treatment programmes with case management models (Malla et al., 2002) should be considered. Psychoeducational training, including activity-stimulating checklists, improves the possibilities for patients and close relatives to receive successful care (Medalia et al., 2002; Liberman et al., 2002b; McFarlane et al., 1995). Finally, there are also data suggesting that patients' cognitive abilities can be improved by systemised training. Both the ability to solve social problems (Kern et al., 2005) and executive function (Bell et al., 2001; Kern et al., 2002) can be improved. There are several potentially effective methods for cognitive training (Twamley et al., 2003). In summary, this study was a prospective follow-up of patients with and without cross-sectional remission. We have found that the requirement for nursing home care and shel-
tered living are significantly higher for the group of patients who not were in cross-sectional remission while the total needs of healthcare interventions were lower among patients being in cross-sectional remission. We have also found that short of achieving the full remission criteria (including duration) the cross-sectional severity criteria for remission is an important and highly relevant treatment goal. When this straightforward, dual categorisation of patients that is crosssectional remission, were compared with the more extensive PANSS-scores, or if admissions to hospital are added to being in remission, no different patterns in outcomes could be seen. This observation may serve as a validation of the remissionconcept and an argument for the use of this thresholdmethod, that remission represents, offering patients and psychiatrists a solid goal for the treatment. A limitation of the study was the large variances in measures as an expression for a high variability inside the two groups. This may be relevant given the heterogeneity and lack of differences in the context of a remitted/non-remitted classification. Another shortcoming is that it has not been possible to follow the remission status of the patients with adequate frequency in this type of naturalistic study to determine the duration of remission. Strategies should be tested for monitoring time in remission for those patients seen infrequently. However, it is likely that many patients are seen by some health care provider at least every one to two months and adaptation of the PANSS items to assess longer intervals than one week would be appropriate in that context. Role of funding source This study was sponsored by Janssen-Cilag, Sweden with unrestricted grants. Contributors There were no other contributors than the department of psychiatry, NU Health Care Hospital, Trollhättan Sweden. Conflicts of interest There were no conflicts of interest. Acknowledgements This study was supported by unrestricted grants from Janssen-Cilag, Sollentuna, Sweden. We gratefully acknowledge the excellent technical assistance of Britt-Marie Hansson, Ruth Johansson and Maivor Olsson.
References Andreasen, N.C., Carpenter, W., Kane, J.M., Lasser, R.A., Marder, S.R., Weinberger, D.R., 2005. Remission in schizophrenia: proposed criteria and rationale for consensus. The Am. J. Psychiatry. 162 (3), 441–449. Barnes, T.R., Curson, D.A., 1994. Long-term depot antipsychotics. A risk– benefit assessment. Drug Safety. 10 (6), 464–479. Bell, M., Bryson, G., Greig, T., Corcoran, C., Wexler, B.E., 2001. Neurocognitive enhancement therapy with work therapy: effects on neuropsychological test performance. Arch. Gen. Psychiatry. 58 (8), 763–768. Fleischhacker, W.W., Eerdekens, M., Karcher, K., Remington, G., Llorca, P.M., Chrzanowski, W., Martin, S., Gefvert, O., 2003. Treatment of schizophrenia with long-acting injectable risperidone: a 12-month open-label trial of the first long-acting second-generation antipsychotic. J. Clin. Psychiatry. 64 (10), 1250–1257. Green, M.F., 1996. What are the functional consequences of neurocognitive deficits in schizophrenia? Am. J. Psychiatry. 153, 321–330. Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2006. Remission and cognitive ability in a cohort of patients with schizophrenia. Psychiatric Res. 40, 738–745. Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2007. Remission in prognosis of functional outcome: a new dimension in the treatment of patients with psychotic disorder. Schizophr Res. 93, 160–168.
L. Helldin et al. / Schizophrenia Research 115 (2009) 67–73 Helldin, L., Kane, J.M., Karilampi, U., Norlander, T., Archer, T., 2008. Experience of quality of life and attitude to care and treatment in patients with schizophrenia: role of cross-sectional remission. Int. J. Psychiatr. Clinical Pract. 1–8. Jeste, S.D., Patterson, T.L., Palmer, B.W., Dolder, C.R., Goldman, S., Jeste, D.V., 2003. Cognitive predictors of medication adherence among middle-aged and older outpatients with schizophrenia. Schizophr. Res. 63 (1–2), 49–58. Kay, S.R., Fitzbein, A., Opler, L.A., 1987. The positive and negative syndrome scale (PANSS) for schizophrenia. Schizophr. Bull. 13, 261–267. Kay, S.R., Opler, L.A., 1987. The positive–negative dimension in schizophrenia: its validity and significance. Psychiatr. Dev. 2, 79–103. Kay, S.R., Opler, L.A., Lindenmayer, J.P., 1988. Reliability and validity of the positive and negative syndrome scale for schizophrenia. Psychatr. Res. 23, 99–110. Kern, R.S., Liberman, R.P., Kopelowicz, A., Mintz, J., Green, M.F., 2002. Applications of errorless learning for improving work performance in persons with schizophrenia. Am. J. Psychiatry. 159 (11), 1921–1926. Kern, R.S., Green, M.F., Mitchell, S., Kopelowicz, A., Mintz, J., Liberman, R.P., 2005. Extensions of errorless learning for social problem-solving deficits in schizophrenia. Am. J. Psychiatry. 162 (3), 513–519. Kopelowicz, A., Liberman, R.P., Ventura, J., Zarate, R., Mintz, J., 2005. Neurocognitive correlates of recovery from schizophrenia. Psychol. Med. 35 (8), 1165–1173. Lambert, M., Schimmelmann, B.G., Naber, D., Schacht, A., Karow, A., Wagner, T., Czekalla, J., 2006. Prediction of remission as a combination of symptomatic and functional remission and adequate subjective well-being in 2960 patients with schizophrenia. J. Clin. Psychiatry. 67, 1690–1697. Liberman, R.P., Kopelowicz, A., Venture, J., Gutkind, D., 2002a. Operational criteria and factors related to recovery from schizophrenia. Int. Rev. Psychiatry. 14, 256–272. Liberman, R.P., Glynn, S., Blair, K.E., Ross, D., Marder, S.R., 2002b. In vivo amplified skills training: promoting generalization of independent living skills for clients with schizophrenia. Psychiatry. 65 (2), 137–155.
73
Lindstrom, E., Wieselgren, I.M., von Knorring, L., 1994. Interrater reliability of the structured clinical interview for the positive and negative syndrome scale for schizophrenia. Acta. Psychiatr. Scand. 89 (3), 192–195. Malla, A.K., Norman, R.M., Manchanda, R., McLean, T.S., Harricharan, R., Cortese, L., Townsend, L.A., Scholten, D.J., 2002. Status of patients with first-episode psychosis after one year of phase-specific communityoriented treatment. Psychiatric Services. 53 (4), 458–463. McFarlane, W.R., Link, B., Dushay, R., Marchal, J., Crilly, J., 1995. Psychoeducational multiple family groups: four-year relapse outcome in schizophrenia. Family Process. 34 (2), 127–144. Medalia, A., Revheim, N., Casey, M., 2002. Remediation of problem-solving skills in schizophrenia: evidence of a persistent effect. Schizophr. Res. 57, 165–171. O'Donnell, C., Donohoe, G., Sharkey, L., Owens, N., Migone, M., Harries, R., Kinsella, A., Larkin, C., O'Callaghan, E., 2003. Compliance therapy: a randomised controlled trial in schizophrenia. British Medical Jour. 11 327 (7419), 834. Smith, T.E., Hull, J.W., Goodman, M., Heayat-Harris, A., Willson, D.F., Israel, L.M., Munich, R.L., 1999. The relative influences of symptoms, insight, and neurocognition on social adjustment in schizophrenia and schizoaffective disorder. J. Nerv. Ment. Dis. 187 (2), 102–108. Strauss, J.S., Carpenter, W.T., 1972. The prediction of outcome in schizophrenia I: characteristics of outcome. Arch. Gen. Psychiatry. 27, 739–746. Twamley, E.W., Jeste, D.V., Bellack, A.S., 2003. A review of cognitive training in schizophrenia. Schizophr. Bull. 29 (2), 359–382. van Os, J., Burns, T., Cavallaro, R., Leucht, S., Peuskens, J., Helldin, L., Bernardo, M., Arango, C., Fleischhacker, W., Lachaux, B., Kane, J.M., 2006a. Standardized remission criteria in schizophrenia. Acta. Psychiatr. Scand. 91–95. van Os, J., Drukker, M., Ã. Campo, J., Meijer, J., Bak, M., Delespaul, P., 2006b. Validation of remission criteria for schizophrenia. Am. J. Psychiatry.163 (11), 2000–2002. Weiden, P.J., Kozma, C., Grogg, A., Locklear, J., 2004. Partial compliance and risk of rehospitalization among California Medicaid patients with schizophrenia. Psychiatr. Serv. 55 (8), 886–891.