The impact of chronic diseases on the family environment can have a decisive influence on the course and prognosis of these conditions, as disease evolution has been found to be influenced by both family-related risk and protective factors. The presence of chronic physical or mental disorders is also a significant contributor to family burden, which can be understood as the emotional and financial stresses experienced by the families of chronically ill patients.
Most studies of family burden among mental health clients limit their focus to common mental disorders, such as schizophrenia, and point to the fact that all areas of family functioning are affected by the presence of the disease.[4, 5] However, recent studies, such as a review of caregiver burden in anxiety disorders, have also addressed the impact of obsessive–compulsive disorder (OCD) and post-traumatic stress disorder (PTSD) on patient families. A study of family burden in OCD revealed similar results to those obtained for patients with schizophrenia, and found that family support may have significant repercussions for the prognosis and treatment outcomes of OCD. Another recent study on family accommodation and OCD reported a high prevalence of accommodating behaviors in patient families. Studies of PTSD generally involve the families of war veterans, and tend to find that the wives and children of patients with this disorder tend to report the highest levels of burden, as well as high rates of hostility and domestic violence. However, the previously cited review did not discuss any studies on the correlation between family burden and panic disorder (PD).
The characteristic symptoms of PD, especially anticipatory anxiety, the avoidance of certain places and situations due to the fear of experiencing panic attacks, and the loss of autonomy, lead to increased dependence on family members. PD is also a chronic illness whose symptoms have a negative influence on psychosocial functioning.
Comparative studies of the family burden and family environment profiles reported by the caregivers of patients with clinical diseases versus caretakers of individuals with psychiatric disorders are still rare in the literature. A recent study compared burden levels between the caregivers of patients with mental disorders (schizophrenia, mood disorders and substance abuse) with those of a group of patients with clinical diseases (diabetes mellitus, arthritis and lung disease). The study found that the former group reported higher levels of subjective burden. Similar results were found in another study involving patients with somatic diseases and individuals with mental disorders and somatic comorbidities. However, both studies are limited due to a lack of representativeness of the caregiver population, and emphasize the need for further research comparing the levels of burden reported by individuals involved in the care of family members diagnosed with different illnesses.
Studies involving the families of patients with cardiovascular, metabolic, neoplastic, inflammatory and degenerative diseases have found that these illnesses have negative effects on the quality of life of family caregivers,[1, 15] and tend to increase reported burden levels and generate conflicts in the family environment. Other studies have also found that the family caretakers of patients with chronic diseases are more likely to present with mood disorders, such as anxiety and depression, and report higher levels of stress than the general population.[16, 17]
The present study aimed to assess whether the family environment and levels of burden reported by the family members of patients with a PD differ from those observed in the family members of patients with chronic clinical diseases. The correlation between family burden and environment and the sociodemographic and clinical characteristics of patients with PD were also assessed.
- Top of page
The sample comprised a group of 67 relatives of patients with PD, and a control group of 66 relatives of patients with clinical diseases. The symptom severity of the 67 patients with PD whose family members participated in the former group was also assessed. The sociodemographic characteristics of patients' relatives are described in Table 1. Groups did not differ as to age and marital status. However, among the family members of patients with PD, there was a significant predominance of men, higher prevalence of parental ties and higher levels of education than among members of the control group. A significant predominance of caregivers who were the children of the patients in question (son or daughter) was also observed among individuals with clinical diseases.
Table 1. Comparison of sociodemographic characteristics between the relatives of patients with panic disorder and those of subjects with clinical diseases
|Characteristics||Total||Clinical disease||Panic disorder||P|
|n = 133||n = 66 (49%)||n = 67 (51%)|
|Sexb|| || || || |
|Male||43 (32)||12 (28)||31 (72)||0.001a|
|Female||90 (68)||54 (60)||36 (40)|
|Age (years)c||47.10 (13.48)||49.30 (13.23)||44.93 (13.47)||0.061|
|Education (years)c||10.13 (4.35)||8.18 (3.67)||12.04 (4.13)||<0.001a|
|Marital statusb|| || || || |
|With partnerd||95 (71)||47 (49)||48 (51)||<0.999|
|Without partnere||38 (29)||19 (50)||19 (50)|
|Caregiver kinshipb|| || || || |
|Spouse||70 (53)||30 (43)||40 (57)||0.002a|
|Parents||14 (11)||3 (21)−||11 (79)+|
|Children||33 (25)||25 (76)+||8 (6)−|
|Othersf||16 (12)||8 (50)||8 (50)|
The most common diagnoses in the control group were systemic arterial hypertension (SAH), which was reported by 50 (75%) participants, followed by heart disease (n = 41; 62%) and diabetes mellitus type 2 (DM2) (n = 24; 36%). On average, patients reported 3.1 (SD = 1.21) clinical comorbidities and high rates of polypharmacy, with patients taking a median of six (4–7.75) medications every day.
In the PD group, of the 10 patients (15%) who presented with clinical comorbidities, seven (10%) had SAD, three (4%) had been diagnosed with hypothyroidism, two (3%) with DM2 and two (3%) with asthma. Five (7%) patients had more than one clinical comorbidity in addition to PD. The most commonly reported psychiatric comorbidities (Axis I) in this group were: agoraphobia (n = 61; 91%); mood disorders (current and past depression or dysthymia) (n = 34; 51%); and other anxiety disorders (generalized anxiety disorder, social anxiety disorder or OCD) (n = 32; 48%). A total of 22 (33%) patients met criteria for more than one comorbid disorder, excluding agoraphobia. All patients received medication, with 48 (72%) taking selective serotonin reuptake inhibitors (SSRI), seven (10%) receiving tricyclic antidepressants (TCA), 36 (54%) taking benzodiazepines, and 32 (48%) taking more than one type of medication.
Between-group comparisons of family burden and environment
The present results indicated that the family members of patients with PD reported higher levels of subjective burden associated with daily patient assistance, although no between-group differences were observed in other subjective burden subscales. The control group reported higher levels of objective burden on the daily patient care, impact on caregiver's life, and general objectivity scales (Table 2).
Table 2. Comparison between the family burden (FIBS-BR) and family environment (FES) reported by the families of patients with panic disorder and clinical diseases
| ||Clinical disease||Panic disorder||Pb|
|n = 66 (49%)||n = 67 (51%)|
|Family burden|| || || |
|Objective|| || || |
|Assistance to the patient in daily life||2.24 (0.83)||1.90 (0.81)||0.017a|
|Supervision of problematic behaviors||1.39 (0.51)||1.35 (0.47)||0.601|
|Impact on caregiver's life||1.64 (0.95)||1.34 (0.63)||0.032a|
|Total objective burden||1.80 (0.56)||1.58 (0.55)||0.024a|
|Subjective|| || || |
|Assistance to the patient in daily life||1.43 (0.69)||1.71 (0.71)||0.025a|
|Supervision of problematic behaviors||2.81 (1.05)||2.79 (1.00)||0.938|
|Worries about patient||3.07 (0.90)||3.01 (0.96)||0.686|
|Total subjective burden||2.53 (0.67)||2.59 (0.67)||0.592|
|Family environment|| || || |
|Relationship|| || || |
|Cohesion||7.58 (1.74)||7.25 (1.83)||0.291|
|Conflict||1.62 (1.80)||2.42 (2.07)||0.020a|
|Expressiveness||4.33 (1.45)||4.57 (1.43)||0.347|
|System maintenance|| || || |
|Control||4.32 (1.83)||4.27 (1.95)||0.808|
|Organization||7.25 (1.86)||6.16 (2.33)||0.004a|
|Personal growth|| || || |
|Achievement orientation||6.46 (1.13)||5.94 (1.44)||0.023a|
Analysis of the family environment instruments revealed that the family members of patients with PD reported more family conflicts than the control group, who obtained higher scores in the organization and achievement-orientation scales (Table 2). No significant differences were found in the remaining dimensions of the FES.
A GLM analysis was then performed to investigate the association between family burden, family environment and independent variables. Subscale scores on family burden and family environment instruments, which differed significantly between groups, were considered dependent variables. Patient sex, level of education and caregiver kinship were entered into the model as independent variables.
Higher levels of subjective burden (Table 3) were significantly higher when the caregivers were the children of the patients (P = 0.001). The relatives of patients with PD also obtained lower scores than those of patients with clinical diseases on the ‘assistance in daily life’ (P = 0.023) and ‘impact on caregiver's life’ (P = 0.008) scales of the FIBS, as well as lower total objective burden scores on this scale (P = 0.024).
Table 3. Multivariate analyses of family burden, controlling for sex, age and education†
|Assistance to the patient in daily life||Impact on caregiver's life||Total objective burden||Assistance to the patient in daily life|
|Mean (SE)||P||Mean (SE)||P||Mean (SE)||P||Mean (SE)||P|
|Group||Panic disorder||1.92a (0.13)*||0.023*||1.42a (0.12)*||0.008||1.64a (0.09)*||0.024*||1.64 (0.11)||0.884|
|Clinical disease||2.36b (0.15)*||1.93b (0.15)*||1.94b (0.11)*||1.61 (0.13)|
|Parent||Spouse||1.89 (0.96)||0.165||1.35a (0.95)*||0.017||1.57a (0.07)*||0.021*||1.53a (0.08)*||0.001*|
|Others||2.09 (0.20)||1.57ab (0.20)||1.66b (0.14)*||1.30a (0.17)*|
|Children||2.12 (0.17)||1.60ab (0.17)||1.82ab (0.11)||1.37a (0.16)*|
|Parents||2.47 (0.26)||2.20b (0.26)*||2.10ab (0.18)||2.31b (0.22)*|
Caregiver kinship was also associated with scores on the ‘conflict’ dimension of the FES (Table 4), with significant differences being identified between the scores of adult children and spouse caregivers on this scale (P = 0.045), regardless of group. Scores on the ‘organization’ dimension were lower for the family members of patients with PD than for individuals in the control group (P = 0.036), with the scores of spouse caregivers differing significantly from those of other caregivers (P = 0.034).
Table 4. Multivariate analysis of family environment, controlling for sex, age and education†
|Family environment||Relationship||System maintenance||Personal growth|
|Mean (SE)||P||Mean (SE)||P||Mean (SE)||P|
|Group||Panic disorder||2.76 (0.30)||0.258||5.76a (0.33)*||0.036*||5.87 (0.19)||0.160|
|Clinical disease||2.23 (0.37)||6.83b (0.41)*||6.32 (0.27)|
|Parent||Spouse||1.74a (0.23)*||0.045*||7.03a (0.26)*||0.038*||6.32 (0.15)||0.248|
|Others||2.35ab (0.48)||5.51b (0.53)*||5.74 (0.31)|
|Children||2.81ab (0.40)||6.01ab (0.44)||6.42 (0.26)|
|Parents||3.06b (0.63)*||6.63ab (0.69)||5.91 (0.47)|
Family burden, family environment and PD severity
The mean (SD, range) PDSS score in the patient group was 12.48 (SD = 6.98; 0–27). A significant positive correlation was found between scores on all objective burden scales and the intensity of PD symptoms (Table 5). Subjective burden presented meaningful positive correlations with worries about the patient and with total burden scores. No meaningful correlations were found between the severity of PD and family environment scores (coefficients ranged from r = –0.020 to 0.042).
Table 5. Correlation between the severity of PD (PDSS) and family burden
|Family burden (FIBS-BR)b||Objective||Subjective|
|Assistance to the patient in daily life||0.361a||0.003a||0.119||0.367|
|Supervision of = problematic behaviors||0.352a||0.003a||0.057||0.718|
|Impact on caregiver's life||0.424a||<0.001a||—||—|
|Worries about patient||—||—||0.494a||<0.001a|
A GLM analysis was then used to identify the independent variables associated with the family burden levels reported by the relatives of patients with PD. All sociodemographic characteristics that differed significantly between the two groups of caregivers (two-tailed P < 0.05) were entered as independent predictors in the model (sex, level of education and caregiver kinship), as were patients' PDSS scores. Scores on FIBS-BR subscales, which were significantly correlated with PDSS scores, were entered as dependent variables, as shown in Table 3.
Higher levels of objective burden associated with ‘assistance in daily life’ (PDSS – b [95% confidence interval (CI)] = 0.04 [0.02–0.07]; P < 0.001) and ‘impact on caregiver's life’ (PDSS – b [95%CI] = 0.04 [0.02–0.06]; P < 0.001), as well as greater overall objective burden scores (PDSS – b [95%CI] = 0.03 [0.0–0.05]; P < 0.001) were found to be related to the severity of PD symptoms. In addition to being explained by more severe PD (PDSS – b [95%CI] = 0.02 [0.01–0.03]; P = 0.001), scores on the ‘supervision of patient problematic behaviors’ scale were also associated with the presence of caregivers who were the children of the patient (b [95%CI] = 0.56 [0.17–0.95]; P = 0.004).
Scores on the ‘worries about patient’ scale and overall subjective burden levels were also predicted by more severe PD (PDSS – b [CI 95%] = 0.06 [0.03–0.09]; P < 0.001 – PDSS – b [95%CI] = 0.03 [0.01–0.05]; P = 0.002, respectively).
- Top of page
The present results have shown that the family members of patients with PD report higher levels of subjective burden than the relatives of patients with clinical diseases. These findings support those of recent studies that have compared the burden levels reported by the families of patients with mental disorders with those of subjects with clinical diseases, and found higher levels of subjective burden in the former. However, the burden scores in our study were lower than those found in the validation study of the scale used to assess this variable, which involved patients with more severe mental disorders, such as schizophrenia. The fact that the current sample was composed of family members of patients with non-psychotic mental disorders who were receiving outpatient treatment might have attenuated family burden levels.
On the other hand, objective burden levels were lower in the PD group than among control participants. According to previous studies, family re-adaptation is necessary in all cases of chronic illness, regardless of the diagnosis. The medical orientations and treatment associated with these conditions often result in changes in the family routine. For example, chronic illnesses often require continuous medication use, changes in the patient's diet and other similar restrictions, especially when patients require regular assistance to perform daily life activities.
The between-group differences found between sociodemographic characteristics might be due to characteristics of the diseases studied. The fact that most caregivers of patients with PD were male could be related to the fact that women are two to three times more likely to be affected by this disorder, panic attacks or agoraphobia than men, and that the caretaker role is often assigned to their spouse. On the other hand, the predominance of female caregivers among patients with clinical disease was in agreement with other studies.[1, 27] According to some authors, daughters or wives tend to take on the primary caregiver role in these cases due to the social and cultural expectations of women.
A multivariate analysis showed that the highest levels of subjective burden were reported by caregivers who were also the patient's children, regardless of the group to which they belonged. A similar result was found in a recent study comparing burden levels between the family caregivers (parents, spouses and children) of patients with schizophrenia and mood disorders. This study evidenced that burden levels were influenced by caregiver kinship in addition to the kind of disease.
The results of family environment scores revealed that PD was associated with worse family ‘organization.’ This finding corroborates those of a separate study with 97 children with chronic diseases and their respective family members, which found that lower anxiety levels were associated with better family functioning. However, other studies have also found that the presence of clinical diseases in a family could result in greater cohesion between family members, which made activity planning easier. Thus, it is possible that ‘organization’ deficits may be consistently and specifically associated with the families of patients with anxiety disorders.
The multivariate analysis performed on patients in the PD group revealed that the severity of the condition was an independent predictor of family burden. It is possible that patients with more intense panic attacks or who have a greater fear of new attacks have more entrenched avoidance behaviors, and consequently, grow more dependent on their families even for routine activities, such as, for example, riding the bus, going to the bank, to the supermarket, or to the shopping mall. Moreover, the financial difficulties often reported by patients with mental disorders also lead to increased family dependency.
Studies of the impact of family-focused interventions in different disorders have pointed out that psychoeducational interventions were effective in reducing family burden and in improving the symptoms of patients with clinical conditions, such as DM2, cardiovascular diseases and cancer, and with mental disorders, such as depression and schizophrenia.
Therefore, the assessment of family burden can be a useful tool in designing family-focused therapeutic strategies, which contribute both to the course and prognosis of the disease, as well as to patient quality of life. To our knowledge, this is the first study to assess the impact of PD on patient families, and our results underscore the importance of including the family in the patient's treatment.
Some limitations of this study must be considered, such as the heterogeneity of our sample and the fact that family burden and family environment were only assessed at one point in time. The scarcity of similar studies in the literature also limited our discussion of the influence of patient sociodemographic characteristics on our results. The presence of mental disorders and the severity of the clinical conditions presented by patients in the control were also not assessed, although both of these factors could influence burden levels. The lack of studies about family burden, family environment and PD in the literature also limits our ability to compare our results to those of other studies. Lastly, we did not assess the influence of clinical comorbidities in the control group on family burden and family environment.