Exploring the burden of the primary family caregivers of schizophrenia patients in Taiwan
Aim: The purpose of the present study was to investigate the burden of the primary family caregivers of schizophrenia patients and the factors that affect caregiver burden.
Methods: A total of 126 pairs of patients and their primary family caregivers were recruited from the day care and acute wards of two teaching general hospitals. Demographic information was collected. All patients went through an interview with a senior psychiatrist using the Brief Psychiatric Rating Scale (BPRS). The Caregiver Burden Inventory–Brief Version and the Chinese Health Questionnaire (CHQ) were used to assess the caregiver burden and the caregivers' health condition. One-way analysis of variance and Pearson correlation were used to analyze the relationship between demographic factors and caregiver burden. Multiple regression was used to analyze predictors of burden of caregivers. Statistical significance in this study was defined as P ≤ 0.05.
Results: The caregiver burden scores (25.9 ± 10.7; range, 3–61) indicated a moderate burden level. Among the five dimensions of burden, caregiver anxiety (2.13 ± 0.86) was the highest, followed by dependency of the patient (1.85 ± 1.02), feeling shame and guilt (1.56 ± 1.02), and family interference (1.43 ± 0.83). The burden level of stigmata (1.32 ± 1.24) was the lowest. The first three rankings of burden were ‘I worry about his/her safety when he/she is alone’ (2.26 ± 1.15), ‘I worry that she/he will become sick at any time’ (2.23 ± 1.10), and ‘I feel he/she depends on me very much’ (2.02 ± 1.29). Patient satisfaction with medical staff support (P = 0.00), CHQ scores of the caregivers (P = 0.00), and the BPRS scores of the patients (P = 0.01) were significantly associated with the caregiver burden scores.
Conclusions: The physical and mental health condition of the primary family caregiver of schizophrenia patients was the most important factor determining the caregiver burden level. Comprehensive multidisciplinary care of chronic schizophrenia patients is needed, care that supports the needs and improves the psychiatric symptoms of the patients, helps to decrease the caregiver burden level, and facilitates family participation in treatment.
SCHIZOPHRENIA IS A leading contributor to the global burden of disease.1 Currently, there are 9.28‰ of 87 039 chronic psychiatric patients in Taiwan, and 69.38% of them are suffering from moderate to severe mental disorders. One-fourth of these patients with severe mental disorders are dependent on long-term medical care.2 Although the burdens of the disease have been considered, the new trend in psychiatric care focuses on deinstitutionalization and community care.3 The execution of short-term hospital stays has been emphasized, and patients return to the community as soon as their critical condition has improved. A marked impact on community care is expected. With the insufficient community care system, the responsibility of patient care will land on the family.
It was estimated that 50–80% of schizophrenia patients keep close contact with their friends or families in Western countries.4 In Taiwan, 95% of psychiatric patients in communities live with their friends, families, and relatives.5 When the patients returned to the community, families were their main support system and even more important than the medical staff.6,7 In the community, 36.9% of caregivers resigned from their jobs in order to care for the patients, and 57.4% of the patients were unable to stay home alone for more than 3 hours per day.8 The family members suffered from major psychological burdens because of long-term care.4 The family caregivers who lacked support were more vulnerable to psychiatric disease.9,10 Nevertheless, the personal impact of schizophrenia has rarely been investigated.11,12
From the perspectives of the holistic and continuous care of chronic disease, health care not only focuses on the care of the patients, but also the caregivers. Therefore, the purpose of the present study was to understand the perceived burden level of the primary family caregivers, the contents of the burden, and the factors that affected caregiver burden.
Participants were consecutively selected from the day care and acute wards of two teaching general hospitals. Inpatients diagnosed with schizophrenia, according to the criteria of the DSM-IV,13 and their primary caregivers were recruited for the study by convenience sampling. Patients who were younger than 18 years of age, older than 65 years of age, with substance-related disorders, or with an organic mental diagnosis, or who were unable to comprehend the items on the inventories, were excluded. Written informed consent was obtained from all patients and primary caregivers after detailed explanation of the purpose of the study. The research was approved by the Institutional Review Board of Kaohsiung Medical University Hospital.
One hundred and sixty research questionnaires were distributed, with a response rate of 78.8% (126 returned questionnaires). The main reason for the unreturned questionnaires was refusal of caregivers in 21 cases (61.8%). Data were incompletely entered in 10 cases (29.4%) and medical evaluation incompletely in three cases (8.8%). We finally recruited 126 pairs of patients and their primary family caregivers and collected data between April and October of 2003.
Demographic information sheet
The demographic information was collected on a sheet designed for the purpose of this study. Information collected from patients included their age, gender, marital status, education, occupation and religious activities. The patients were also requested to rate the support they received from family, friends and medical staff. Information collected regarding the diagnosis included age of onset, duration of illness, side-effects of medication, and history of self-injuring behavior.
Demographic information collected from the primary caregivers included age, gender, marital status, occupation, religious activities, relationship to the patient, amount of time spent with the patient per day, and the duration of providing care. Family information collected included level of financial sufficiency, and the number of family members cohabiting.
Brief Psychiatric Rating Scale
The Chinese version of the Brief Psychiatric Rating Scale (BPRS) was revised to fit the Taiwanese culture by Chang et al.14 The scale consists of 18 items; the first 16 were rated on a 7-point Likert scale of 0–6, and items 17 and 18 used a 5-point Likert scale of 0–4. Higher scores implied greater symptom severity. The interrater reliability of the Chinese BPRS was 0.71–0.83.14,15
Caregiver Burden Inventory–brief version
The Caregiver Burden Inventory–brief version (CBI) included 18 items measuring feelings and problems encountered by caregivers while taking care of a psychiatric patient. Five dimensions of caregiver burden were measured, seven items on family disruption, two items on stigma, seven items on feelings of guilt, three items on caregiver anxiety, two items on the dependency of the patient, and two unscored items. Responses were on a 5-point Likert scale, and the addition of all scores resulted in the total burden score. Higher scores represented higher levels of burden. The result could be categorized into four levels of burden. The internal consistency of the inventory was 0.88, and between 0.65 and 0.95 for individual dimensions. The test–retest reliability of the inventory was 0.90, and the correlation between the dimensions was 0.45. The inventory demonstrated good internal construct validity.16
Chinese Health Questionnaire
The Chinese Health Questionnaire (CHQ) is a self-report inventory designed by Cheng and Williams.17 The participants are asked to rate their self-perceived health condition within the past 2 weeks. The inventory consists of 12 items using a 4-point Likert scale. The cut-off point was 3/4 for the psychological health of the individual: those scoring <4 had a normal psychological health condition, and those scoring ≥4 were at risk of a mental diagnosis. Previous studies have shown that the inventory is suitable for the screening of minor psychiatric disorders in the community. The internal consistency of the CHQ was indicated by alpha coefficients of 0.79.17
The researchers read the questions and circled the responses for participants who were illiterate or had sight problems. After the demographic information was collected, the patients went through an interview with a senior psychiatrist using the BPRS. The same psychiatrist interviewed all participants in both hospitals. The primary family caregivers had to fill out the CBI and the CHQ.
We used one-way analysis of variance to assess the correlation between caregiver burden scores and demographic factors, with respect to continuous variables such as age and duration of illness and a Pearson correlation to examine the relationship between demographic factors, with respect to discrete variables such as gender, family support and caregiver burden. Multiple regression was used to analyze predictors of burden of caregivers. Analysis was carried out using SPSS 10.0 for Windows (SPSS Inc., Chicago, IL, USA). Statistical significance was defined as P ≤ 0.05.
Demographic characteristics of the 126 patients were as follows: mean age 36 ± 9.5 years; age at onset, 25 years, with 11-year duration of illness; and BPRS score of 12.9 ± 7.0. Of the 126 patients, 79 (62.7%) were male, 100 (79.4%) were single, 85 (67.5%) were junior high school graduates or above, 108 (85.7%) were unemployed or not enrolled in a rehabilitation program, 96 (76.2%) did not take part in regular religious activities, 105 (83.3%) were satisfied with their family's support, 83 (65.9%) were satisfied with their friends' support, 110 (87.3%) were satisfied with the medical staff's support, 73 (57.9%) felt that their daily lives were affected by the side-effects of medication, and 43 (34.1%) had a history of self-injuring behavior (Table 1).
Table 1. Demographic factors that affected caregiver burden scores (discrete variables)
|Patients|| || |
| Gender|| ||0.72 (0.40)|
| Male (79)||25.25 (10.37)|| |
| Female (47)||26.91 (11.14)|| |
| Marital status|| ||0.02 (0.88)|
| No regular partner (100)||25.80 (10.44)|| |
| Having a regular partner (26)||26.15 (11.66)|| |
| Education (years)|| ||0.03 (0.86)|
| ≤9 (41)||26.12 (9.84)|| |
| >9 (85)||25.75 (11.08)|| |
| Occupation|| ||0.88 (0.35)|
| Unemployed (18)||28.06 (10.45)|| |
| Employed (108)||25.51 (10.69)|| |
| Religious activities|| ||0.16 (0.69)|
| None or not frequent (96)||26.08 (11.09)|| |
| Regular (30)||25.20 (9.26)|| |
| Family support|| ||2.33 (0.13)|
| Dissatisfied (21)||29.10 (14.19)|| |
| Satisfied (105)||25.23 (9.76)|| |
| Support from friends|| ||0.33 (0.57)|
| Dissatisfied (43)||25.12 (12.54)|| |
| Satisfied (83)||26.27 (9.59)|| |
| Medical staff support|| ||8.39 (*0.00)|
| Dissatisfied (16)||32.88 (14.44)|| |
| Satisfied (110)||24.85 (9.66)|| |
| Side-effects of medication|| ||0.02 (0.90)|
| No (53)||25.74 (11.05)|| |
| Yes (73)||25.97 (10.43)|| |
| History of self-injuring behavior|| ||0.25 (0.62)|
| No (83)||25.53 (11.33)|| |
| Yes (43)||26.53 (9.30)|| |
|Caregivers|| || |
| Gender|| ||0.80 (0.37)|
| Male (48)||26.96 (10.48)|| |
| Female (78)||25.21 (10.77)|| |
| Marital status|| ||0.89 (0.35)|
| No regular partner (41)||24.59 (10.87)|| |
| Having a regular partner (85)||26.49 (10.55)|| |
| Education (years)|| ||0.35 (0.55)|
| ≤9 (67)||26.40 (11.28)|| |
| >9 (59)||25.27 (9.96)|| |
| Occupation|| ||0.43 (0.51)|
| Unemployed (69)||25.30 (10.16)|| |
| Employed (57)||26.56 (11.27)|| |
| Religious activities|| ||0.28 (0.60)|
| None or not frequent (90)||25.56 (11.00)|| |
| Regular (36)||26.67 (9.84)|| |
| Financial sufficiency|| ||1.31 (0.25)|
| Not enough (46)||27.30 (11.71)|| |
| Enough (80)||25.05 (9.98)|| |
| Relationship to the patient|| ||0.85 (0.43)|
| Parents (84)||26.63 (10.30)|| |
| Spouse (11)||26.18 (15.69)|| |
| Siblings or relatives (31)||23.71 (9.54)|| |
| CHQ scores|| ||9.98 (*0.00)|
| <4 (91)||24.08 (9.93)|| |
| ≥4 (35)||30.54 (11.18)|| |
Demographic characteristics of the 126 primary family caregivers were as follows: mean age 55 years; CHQ scores 2.71 ± 3.20, with an 8-year duration of providing care. On average, they spent 9 h with the patient per day and had two to three cohabiting family members.
Of the 126 caregivers, 78 (61.9%) were female, 85 (67.4%) were married, 67 (53.2%) were junior high school graduate or above, 69 (54.8%) were unemployed, 90 (71.4%) did not have regular religious activities, 80 (63.5%) were financially sufficient, 84 (66.7%) were the parents themselves, and 91 (72.2%) were healthy mentally and physically (CHQ < 4).
Burden level of the primary family caregivers and the contents of the burden
The burden level and the contents of the burden of the primary family caregivers are listed in Table 2. The caregiver burden score of 25.9 ± 10.7 (range, 3–61) indicated a moderate burden level. A total of 21.4% of the caregivers had a moderate to severe burden level (CBI scores 33–44), and 3.2% had a critical burden level with CBI scores >45. Among the five dimensions of burden, burden level of caregiver anxiety (2.13 ± 0.86) was the highest, followed by dependency of the patient (1.85 ± 1.02), feeling shame and guilt (1.56 ± 1.02), and family interference (1.43 ± 0.83). The burden level of stigmata (1.32 ± 1.24) was the lowest. We used the paired t-test to compare the mean scores of five subscales of caregiver burdens. The results indicated that the mean score of caregiver anxiety was higher than that of family interference (t = 10.872, P < 0.001), stigmata (t = 7.223, P < 0.001), feeling shame and guilt (t = 5.512, P < 0.001), and dependency of the patient (t = 2.883, P < 0.01). Thus, the mean score of caregiver anxiety was the highest among the five subscales of caregiver burden.
Table 2. Burden level and content of burden on main family caregiver
| 02. Interference in daily work and rest due to taking care of him/her||1.84||1.08||0.25||−0.69||5|
| 03. Cannot handle both family and career||1.51||1.13||0.49||−0.41||10|
| 06. Interference in leisure activities||1.72||1.20||0.24||−0.80||6|
| 08. Income has been reduced because of caring for him/her||1.31||1.34||0.75||−0.59||14|
| 09. Family members argue because of taking care of him/her||1.36||1.14||0.44||−0.78||13|
| 11. Social life has been affected||1.50||1.28||0.47||−0.77||11|
| 16. Marriage has been affected||0.77||1.06||1.34||1.12||16|
| 13. Makes family members feel a sense of shame||1.37||1.29||0.67||−0.56||12|
| 14. Makes me feel a sense of shame||1.26||1.29||0.81||−0.39||15|
|Feeling shame and guilt||1.56||1.02||0.60||−0.05||3|
| 01. I cannot do enough for him/her||1.58||1.20||0.50||−0.64||8|
| 04. What I do is not good enough||1.55||1.11||0.58||−0.10||9|
| 05. His/her behavior and the state of his/her illness make me nervous and afraid||1.91||1.14||0.11||−0.82||4|
| 07. I worry that he/she will become sick at any time||2.23||1.10||−0.25||−0.47||2|
| 15. I worry about his/her safety when he/she is alone||2.26||1.15||−0.05||−0.60||1|
|Dependency of the patient||1.85||1.02||−0.02||−0.74||2|
| 10. I feel he/she depends on me very much||2.02||1.29||0.00||−1.01||3|
| 12. There are some things he/she can do, but still wants me to help||1.67||1.10||0.17||−0.71||7|
|Others – rhetorical questions|
|17. Caring for him/her makes me feel useful||1.89||1.13||0.09||−0.81|| |
|18. Caring for him/her makes me more accepting of this situation||2.00||1.23||0.03||−0.91|| |
|Caregiver burden scores||25.87||10.65||0.33||0.15|| |
The first three rankings of burden were ‘I worry about his/her safety when he/she is alone’ (2.26 ± 1.15), ‘I worry that he/she will become sick at any time’ (2.23 ± 1.10), and ‘I feel he/she depends on me very much’ (2.02 ± 1.29).
Factors that affected caregiver burden and predictors of caregiver burden
The correlation between demographic factors and caregiver burden scores are shown in Tables 1,3. Patient satisfaction with medical staff support (P = 0.00), CHQ scores of the caregivers (P = 0.00), and BPRS scores of the patients (P = 0.01) were significantly associated with the caregiver burden scores.
Table 3. Demographic factors that affected caregiver burden score (continuous variables)
|Patients|| || |
| Age||−0.18|| *0.05|
| Age of onset||−0.03||0.70|
| Duration of illness (years)||−0.14||0.11|
| BPRS scores||0.24|| **0.01|
|Caregivers|| || |
| Duration of providing care (years)||−0.15||0.09|
| Amount of time spent with the patient per day (h)||−0.01||0.94|
| No. family members cohabiting||−0.13||0.13|
It was found that the predictors of the caregivers burden scores were caregiver CHQ score (P = 0.00), patient satisfaction with medical staff support (P = 0.01), and the BPRS scores (P = 0.04; Table 4). This explained 15% of the total variances.
Table 4. Multiple regression analysis of caregiver burden scale predictors
|Caregiver burden scores|| || || || ||0.15||8.34||<0.000|
| CHQ scores||6.26||0.26||3.191||0.00|| || || |
| Satisfaction with medical staff support||−7.09||−0.22||−2.641||0.01|| || || |
| BPRS||0.27||0.17||2.067||0.04|| || || |
| Constant||26.88|| ||8.229||0.00|| || || |
Current medical policy encourages short-term hospital stay and promotes community care for patients with schizophrenia. Family members are the main support system and shoulder the responsibility for patient care in the community. The personal impact of this chronic disease needs to be emphasized.
In the present study the primary family caregivers of the schizophrenia patients had a moderate burden level. This finding was consistent with that of previous studies.5,18 The range of caregiver burden scores (3–61), however, reflected the diverse conditions of the main family caregivers: 21.4% of the caregivers had a moderate–severe burden level (CBI scores 33–44), and 3.2% had a critical burden level with CBI scores >45. The caregiver burden level was greater than expected, so it is important for medical staff to understand and support their needs.
We found that the dimensions of caregiver anxiety and the dependency of the patient were the two highest ranked in the CBI. This result was relevant to the first three highly scored items in the CBI: ‘I worry about his/her safety when he/she is alone’, ‘I worry that he/she will become sick at any time’, and ‘I feel he/she depends on me very much’, reflecting that caregiver burden was affected by both positive and negative symptom domains of schizophrenia. The positive symptoms usually attracted clinical attention and created stress for family caregivers.19 The more severe the positive symptoms, the higher the caregiver burden level.20 In contrast, a prolonged course of negative symptoms would cause a long-term burden for the caregivers.
According to the present results, the caregiver burden level can be predicted by the health condition of the caregivers; those with a worse health condition had a higher burden level.5,18 The burdens of daily life, social shame, and negative emotion worsened the caregiver's mental condition.21 The primary caregiver's mental stress was significantly affected by the patient's psychiatric symptoms and could be decreased by improving the patient's symptoms.22 Comprehensive health services should emphasize the personal impact and help reduce the caregiver burden.
Patient satisfaction with support from medical staff was another factor predicting the level of caregiver burden, and indicated that the medical staff was the main source of social support for the patients and the caregivers. Schizophrenia patients have a significant decline in social functioning, with symptoms such as social withdrawal, queer behaviors, negligence of personal hygiene, inadequate interests and energy, and communication disorders,23 and have difficulty in establishing intimacy and emotional bonds with others. The interaction between the patients and medical staff improves the psychosocial functioning of the patients and helps alleviate caregiver burden.
The severity of the patients' symptoms was an important relevant variable in terms of impact on caregiver burden. This finding was similar to those of Song,5 Lin,21 and Jungbauer and Angermeyer.24 The more serious the symptoms of the psychiatric patients, the higher the caregiver burden. Solomon and Draine discovered that most schizophrenia patients had the following positive symptoms: abnormal behavior, retardation, hallucination, phonism etc. These symptoms would attract people's attention and judgments, and engender stigma and family interference.19 The patients' symptom was a burden source for their families.25 The patients' behavior and the degree of the caregivers' burden had a significant interrelation.5 Currently, Taiwan's community psychiatric rehabilitation services include the following: community rehabilitation centers, recovery homes, out-reach care, shelter workshop, emergency treatment, and professional employment consultation services. Mutual referral or the simultaneous provision of services was provided based on the individual needs of the patient.26 The level of freedom in the social life of Asian psychiatric patients is not only limited by physiological and psychological obstacles, but most seriously by societal discrimination and reduced social opportunities. Therefore, the task of community psychiatric care should be to eliminate the misunderstanding of the general population toward psychiatric patients and decrease the stigmatizing attitude. In addition, in Taiwanese society, home and family give a person the most appropriate sense of belonging and a place to return to throughout the life process. Traditionally, psychiatric patients usually live together with their family. Therefore, increasing mutual understanding and harmony among the family members, strengthening the patient's functioning at home, and the family members' understanding of psychiatric symptoms and caring skills are the most important tasks of community psychiatric care to lighten caregiver burden.27 This shows the need for medical personnel to take the initiative in providing education, therapy, and recovery schemes to the patients in order to deal with the problems and needs they may encounter during the course of their illness. In addition, offering the families professional support in the manner of a ‘therapeutic partner’ could help improve the patients' living quality and lessen caregiver burden.
Because caregiver burden may change with the dynamic condition of the patient's illness and the caregiver's mental and physical health and social support, we suggest further qualitative studies to trace the development of caregiver burden in a vertical manner and to understand the changes and factors affecting the primary caregiver burden in different stages.
Conducting family-centered psychoeducative intervention could help promote the management and recovery of the schizophrenia patients' symptoms. In addition, this would further lessen caregiver burden and promote their mental and physical health; in turn, they would able to offer care in a healthier way.