First Adaptation of a Family-Based ACT Model in Mainland China: A Pilot Project

This study was carried out from 2011 to 2012 by the Second Xiangya Hospital in Changsha, the capital city of Hunan province in south central China. Changsha, which is the 23rd largest city in China, has a population of seven million. The new ACT team in Changsha was supported by the Mount Sinai Hospital ACT team of Toronto, Canada, which consulted in the grant application process, provided three staff training sessions in Changsha lasting two or three days each, hosted a visit by Changsha’s team leader to Toronto, provided ongoing clinical consultation via e-mails and phone conferences, and participated in evaluating ACT team fidelity by using the Dartmouth Assertive Community Treatment Scale (DACTS) (7).

The team consists of a team leader (psychiatrist), another part-time psychiatrist, three part-time psychiatric nurses, two full-time clinical psychologists, three full-time social workers, and a part-time employment specialist. The study admission criteria for patients were age 18 or older, DSM-IV diagnosis of schizophrenia, significant functional impairments, and a high need for psychiatric services. The exclusion criteria were refusal to participate in the study or a primary diagnosis that was not a psychotic disorder. The inclusion criteria for family caregivers were age 18 or older and living with and caring for the patient (the only family member with a mental illness) for more than three months. [Further details on inclusion and exclusion criteria are available in an online data supplement to this report.]

From a list of about 500 patients with schizophrenia registered in their local community health offices in two (Tianxin and Furong) of the six districts of Changsha, 180 patients and families were identified who met the study criteria. Of these, the first 31 eligible patients and their families who agreed to participate were assigned to either the intervention group (N=15) or control group (N=16). Assignment was nonrandom and based on the date of study enrollment and space available. No patient dropped out or refused treatment after entering the study.

The intervention group received family-based ACT services that included the following: two or three community or home visits per week to deliver usual ACT care (for example, clinical assessment, crisis intervention, psychosocial assistance, supportive counseling, family support, and functional assistance), and two-hour psychoeducation sessions every two weeks for 24 weeks (total of 12 sessions over six months), with the well-validated McFarlane family psychoeducation model (10) and the mutual support group program for patients and caregivers developed in Hong Kong (11) as foundations. The sessions included components of team rapport building, education on illness and coping skills, self-care, problem-solving skills, creation of an optimal social environment for recovery, and an empathy workshop. Culturally adapted content included how to address mental illness–related stigma in China, how to access local resources, the value of traditional medicine, culturally informed communication skills, and an explanatory model of illness. The control group received a range of standard community services, ranging from an outpatient visit to a psychiatric clinic every one to three months (the Free-Medicine Service of the Ministry of Civil Affairs) to an enriched version of community mental health services (the “686 project” model of the Ministry of Health) that included case manager home visits to the patient up to once a month. All participants provided written, informed consent. The study was approved by the research ethics committee of the Central South University.

Data collection occurred at baseline, three months, and immediately after the six-month intervention. For patients, the measures included the Positive and Negative Syndrome Scale (PANSS) for severity of psychosis, Social Disability Screening Schedule (SDSS), Personal and Social Performance Scale (PSP), and Brief UCSD Performance-Based Skills Assessment (UPSA-B) to assess the level of social, personal, and everyday functioning. For family caregivers, the measures included the Symptom Checklist–90 (SCL-90) for general psychopathology, Family Burden Interview Schedule (FBIS), and Social Support Questionnaire. The evaluators were the psychiatrists and clinical psychologists on the ACT team. The differences between the two groups in before-and-after intervention differences in measurements were compared by using two independent-samples, nonparametric tests (Mann-Whitney U Test), employing SPSS, version 18.0.

Of the 31 patients, 20 were male; the mean±SD age of the sample was 31.7±6.9. No significant difference between the intervention and control groups was noted for any demographic, socioeconomic, or clinical variable. In addition, no significant between-group difference was found in caregiver age, biological relationship between caregiver and patient, number of family members, and family economic conditions. [Data on characteristics of the two groups are presented in a table in the online supplement.] No significant between-group difference was noted at baseline on any of the psychosocial rating scales for patients and family caregivers (Table 1).

The DACTS scores indicated average team fidelity (score of 3.7 out of 5) [for details, see table in the online supplement]. Because substance abuse and addiction are uncommon among patients with schizophrenia living in mainland Chinese communities, the modified DACTS score was 4.1 out of 5 after items pertaining to the team’s capacity for addressing addictions issues (H9, S7, S8, and S9) were excluded, which suggests good model fidelity (7).

In terms of clinical outcomes, no one in the intervention group experienced rehospitalization, whereas two (13%) of the 16 patients in the control group did (hospital stays of 74 and 109 days). Also, no one in the intervention group experienced a relapse, whereas five patients in the control group did. Relapse was defined by any one of the following: psychiatric hospitalization, an increase of 25% in total PANSS score, or significant self-injury or suicidal or violent behavior.

As shown in Table 1, significant between-group differences were noted after the six-month intervention in the total PANSS score (p<.001), PANSS negative subscale score (p=.004), PANSS positive subscale score (p<.001), and PANSS general psychopathology subscale score (p<.001); SDSS total score (p=.001); and PSP total score (p=.001), PSP personal and social relationships subscale score (p<.001), and PSP self-care subscale score (p=.01). Similar differences were also noted at the three-month data collection [see table in online supplement]. No significant difference between groups was found in the UPSA-B scores. Four patients in the intervention group obtained employment, compared with one patient in the control group; however, this difference was not significant. Among family caregivers, no significant between-group differences were noted in the total score or in any of the subscale scores of the SCL-90 and the FBIS (Table 1).

The attendance rates at the family psychoeducation sessions for patients and family members in the intervention group were as follows: eight of 15 patients and families (53%) completed at least ten of the 12 sessions, five of 15 (33%) completed six to nine sessions, and two (13%) completed two to five sessions (that is, 86% had attended at least half the sessions).

The main message and achievement of this pilot study is the preliminary finding that it may be feasible to implement the Western evidence-based ACT model in mainland China, where resource limitations, the culture of caretaking, and mental health legislation vary a great deal from those in the West (12). Findings also suggest that ACT may contribute to reduced readmission days, lower severity of psychopathology, and improved level of social functioning for patients.

The study’s results are interesting in light of the fact that the control group received free psychiatric care every one to three months from the Ministry of Civil Affairs or from a relatively new and substantial service provided by the Ministry of Health (the so-called 686 Project) (13). As noted by Bond and colleagues (2), the original study of ACT in the United States found strong evidence of ACT’s effectiveness in improving participants’ outcomes; however, the evidence was less robust for several ACT programs that provided only basic ACT services and demonstrated lower fidelity to the ACT model. In addition, studies in the United Kingdom by Burns and colleagues (4) did not find the same strong evidence of effectiveness, which may reflect the fact that ACT was compared with a set of standard community services that had been enriched. Thus the positive outcomes in China are meaningful and may be associated with the team’s high fidelity to the ACT model, the team’s frequent contacts with patients and families, optimization of health care resources, early recognition of warning signs of relapse, and the individualized intervention (2,3,12).

The preliminary finding of improved social functioning of patients in the intervention group may be associated with the ACT model’s frequent contacts and individualized treatment plans for achieving rehabilitative goals through employment, independent living skills, social skills, and crisis coping skills. Lower psychopathology as measured by PANSS scores may also have contributed to better social functioning by lowering interference from symptoms. The positive changes in both social functioning and psychopathology are likely interactive. In this study, strong family involvement and a relative lack of substance abuse issues are also beneficial factors, echoing a study of ACT in Japan (6).

In addition, the introduction of ACT’s principles of respectful, nonjudgmental, patient-centered interactions may have contributed to patients’ improved social and clinical functioning. ACT may also help reduce patients’ sense of stigma.

Contrary to our hypothesis, the study did not find significant differences between caregiver groups in measures of burden and support. Possible explanations include insufficient sensitivity of the study instruments, the inadequacy of a family psychoeducation approach to alleviate burden and provide needed support, the short study duration, and the substantial degree of entrenched burden on families over many years. Also, the families themselves may have resisted change for a variety of reasons, such as a healthy suspicion that care providers who are not related to the patient can care enough and make meaningful differences for their loved ones, a sense that the team’s assistance was interfering with family members’ roles and intruding on or displacing family members, and uncertainty about the sustainability of the ACT team and reluctance to change over the short term (12,14). Further study would be worthwhile.

Future research could examine more systemic issues exposed by the pilot study, such as how to fund resource-intensive ACT services in mainland China; how the lack of general community resources affects patients’ clinical outcomes; and how China’s new mental health laws affect families, police, neighbors, employers, and clinicians (15).

The study was limited by its nonrandomized design, its small sample, short duration, the relatively new ACT team, and potential bias introduced by the fact that all the evaluations were completed by team members and no interrater reliability measures were done.

These preliminary results indicate that it may be feasible to implement ACT in mainland China and that implementation may contribute to reducing hospitalization and psychopathology and improving social functioning of patients. Further study is required to ensure that ACT addresses culturally relevant goals of helping Chinese families, who care for the vast majority of patients in the community.