Evaluation of caregiver burden, and quality of life in metastatic prostate cancer patients and caregivers; Effects of clinical and socioeconomic factors

Purpose: To investigate the caregiver burden and quality of life (QoL) of patients with metastatic prostate cancer. The relationship between questionnaire results and medical, socioeconomic, disease and treatment related factors of the patients were evaluated. Methods: One hundred twenty one metastatic prostate cancer patients and caregivers were enrolled in this cross-sectional study. Short Form 36 (SF-36) questionnaires and Zarit caregiver burden interviews (ZCBI) were conducted with patients and their caregivers seperately. Patients’ comorbidities, presence of bone and visceral metastases, bone related events, palliative radiotherapy needs, ECOG scores, being under active chemotherapy, socioeconomic factors such as public transportation need, residential factors and caregiver related factors such as comorbidities and kinships were recorded respectively. Results: Patients’ comorbidities, ECOG score, presence and quantity of bone or visceral metastasis, bone related events, receiving palliative radiotherapy and chemotherapy, using public transportation resulted in lower SF-36 and higher ZCBI results (p < 0.001– p = 0.049). Obligation to live in rented houses were correlated with lower SF-36 physical function results (p = 0.043). Condition of the caregiver being from distant relatives resulted in a lower SF-36 score (p = 0.01 - p= 0.043). Moderate to strong negative correlation was detected between ZCBI and all SF-36 results (p < 0.001, r: -646 to -749). In ZCBI, score over 47 resulted with % 69.57 sensitivity and % 65.52 specicity (AUC = 0.671, p = 0.008). Conclusion: QoL was directly related to caregiver burden. as score,


Introduction
Patients and caregivers have some concerns about reducing daily productivity or missed business days due to hospitality and long-term complications of treatment or disease itself. The idea of having a potential or absolute mortal disease, sustained therapies and disease course could be devastating on feelings in the individuals and family members [1]. Therefore, caregivers of cancer patients might feel burden, and as a result patients might seek for poor quality of life conditions. Caregiver burden and quality of life (QoL) have been investigated in various diseases such as chronic kidney disease including patients receiving dialysis and transplant recipients, schizophrenia, Parkinson's disease, in ammatory bowel disease and Alzheimer's disease [2,3,4,5,6].
There is limited knowledge in the literature regarding caregiver burden with QoL in Cancer. Johansen S et al. were determined the factors that effects caregiver burden of cancer patients. There is a relationship between the patient factors such as sleep disturbance, fatigue and cancer symptoms with caregiver burden. Additionally caregiver's factors such as being female, having sleep disturbance and other chronic illnesses signi cantly impacts the burden [7]. In advanced cancer, the severe patient's physical and mental health is associated with a heavier caregiver burden, regardless of caregiving hours [8].
Crowder et al. have been determined that head and neck cancer patients with a long illness age adapt to daily life restrictions. Most of these patients continue to have unmet needs. They experience a period of accepting that some of their needs are not met in the coping process, and this situation is associated with depressive status of the patients [9]. Guan T et al. found that uncertainty in the course of the disease was associated with mental well-being disorders [10].
There are several studies evaluating the quality of life of prostate cancer patients. The survival and quality of life of patients have increased with new generation anti-androgen treatments. A recent metaanalysis revealed that apalutamide and abiraterone acetate provided the best overall survival and enzalutamide provided the best radiological progression-free survival [11]. With advancing treatments, prostate cancer patients who have a long disease age and can reach older ages represent an important segment in the patient pro le. In the rst 24 months of the disease, although tolerance of all treatment modalities is good, each has a different effect on the quality of life [12].
In this study, we aimed to investigate the caregiver burden and quality of life of patients with metastatic prostate cancer. Our purpose is to investigate questionnaires's relationships with treatments and patient related factors. Additionally we aimed to evaluate the correlation of these questionnaires with each other.
With the results we hoped to achieve, our aim was to provide a systematic quality of life analysis and caregiver burden monitoring. We started our study with the foresight that it can contribute to the formation of a guide in terms of rational and effective distribution of healthcare services whose economic burden is growing day by day.

Materials And Methods
Design, Setting and Patients: One hundred twenty one metastatic prostate cancer patients and primary caregivers were enrolled in this descriptive cross-sectional study from October 2019 to October 2020.
Patients aged 50-91 years who have had the illness for three months or more, followed-up in Gazi Yasargil Training and Research Hospital. Additionally, the primary caregivers of the patients who gave consent for participation were enrolled in the study. The diagnosis of metastatic prostate cancer was con rmed with Galium-68 prostate speci c membrane antigen positron emulsion tomography.
All patients and their primary caregivers (spouse or children) were evaluated in the same meeting room at the Medical Oncology Department of Gazi Yasargil Training and Research Hospital. The evaluation was performed by a medical oncology specialist.
Inclusion Criteria: Patients with metastatic prostate cancer who had at least one bone, lymph node or visceral metastasis who had undergone androgen deprivation or chemotherapy treatments for at least 3 months were included.
Exclusion Criteria: Patients who had an additional uncontrolled chronic disease such as severe hypertension, dis-regulated diabetes mellitus, any hearth disease, any type of ischemic cerebrovascular disease and any type of baseline orthopedic failure that causes mobilization defect, were excluded from the study. The exclusion criterion for hypertension was blood pressure above 170/100 mm / hg in at least 2 consecutive measurements despite using antihypertensive drugs. Diabetic patients with an HbA1c level above 7.5 or using insulin therapy or using more than one oral anti-diabetic agents were excluded.
Patients with an HbA1c level below 7.5 with metformin treatment or single oral anti-diabetic therapy were included in the study. The exclusion criteria of diabetic patients were evaluated according to the relevant age group recommendations of the American diabetes association [13].
Data Collection and Questionnaires: Two self-administered surveys were completed by caregivers as follows: ZCBI Caregiver Burden Interview (ZCBI): This questionnaire was applied to caregivers whom were patients' relatives. ZCBI was created by ZCBI et al. [15] in 1980 as a caregiver self-report measure. The revised version of this scale consists of 22 items about physical health, psychological well-being, nancial status and interpersonal relationships to evaluate caregiver burden and the impact of the patient's disease on a caregiver's life [15]. Ozlu et al.
[16] developed the Turkish-validated ZCBI. On investigating the validity and reliability of the Turkish-language version, three items (numbers 1, 4 and 16) were removed. Hence, the Turkish-language version consists of 19 items. The scale was developed to be self-administered; however, it can also be administered by an interviewer. In the present study, the interviewer read each item aloud and asked the respondent to state their answer. Each item is scored from 1 to 5, 1 = never, 2 = rarely, 3 = sometimes, 4 = quite frequently, and 5 = nearly always. Caregiver burden is evaluated on the basis of the total score obtained from the sum of the responses. The total scores were calculated to be between 19 and 95 points, and higher scores indicating higher caregiver burden. A score of 21 or less indicates no burden, 22-46 indicates mild burden, 47-55 means moderate burden and 56 or more indicates severe burden.

Strati cation Factors
Patient related clinical factors: Age, comorbidities, ECOG score, presence of bone metastatsis, presence of bone-related event and organ metastasis were determined. The bone-related event was de ned as the need for opioid use due to severe bone pain or as an incurred bone fracture.The number of bone metastases was also strati ed as 0, 1-3 and >4 metastases. The reason for this strati cation method was that the cut off value for the distinction between oligometastatic disease and diffuse metastatic disease had been generally considered to be between 3 and 5 metastases by most authors [17,18].
Treatment related factors: All patients received androgen deprivation therapy. The treatment related factors were chemotherapy which was accepted as active receiving in terms of cross sectional data collecting. History of chemotherapy within six months was included as active chemotherapy too. In addition, whether the patients received palliative radiotherapy, prostatic region radiotherapy and prostatectomy was also planned to examine.
Economic and educational factors: Patient educational status, personal vehicle presence or public transport usage of family, place of residence (city center, village), residence status (own house -rent) were determined Caregiver-related factors: Caregiver age, the degree of proximity of caregivers such as spouse, child, sibling or other relatives, caregiver education and caregiver comorbidities were identi ed as caregiverrelated factors.

Statistical Analysis and Ethics
Statistical analysis of the data was performed using Statistical Package for Social Sciences (SPSS) software version 15 (SPSS Inc., Chicago, IL, USA). First, we examined caregivers' and patients' characteristics. The caregiver and patient characteristics are presented as the number and percentage.
Data are presented as the mean ± standard deviation (SD) or median (minimum-maximum) according to the distribution of the data. Normality of the distribution of continuous variables was determined using the Kolmogorov-Smirnov test. The mean differences between two independent groups were compared using the independent samples t-test for normally distributed variables and Mann-Whitney U test for non-normally distributed variables. The differences among three independent groups were compared When we looked at the economic and educational factors, we observed that the educational status did not have a direct effect on the quality of life from illiterate to university graduates. It was observed that those with private means of transport had statistically signi cantly higher SF-36 and lower ZCBI caregiver burden results than those who had to use public transport (p = 0.009 -p = 0.048). However, it was also observed that the difference in SF-36 GH, SF and M results did not reach statistical signi cance. When the patients whose place of residence was in the city center were compared with the patients residing in the village, it was observed that those in the village had lower SF-36 and higher ZCBI scores. In order to investigate the degree of kinship related to the caregiver; the situations of the caregiver being spouse, children, siblings and other relatives were compared. In SF-36 PF, BP and M results, it was observed that the condition of the caregiver being from distant relatives constituted a lower quality of life score (p = 0.01 -p = 0.043). It was observed that the situation of siblings being the caregivers were associated with better SF-36 BP and GH results than children as caregivers (p = 0.025 -0.026). It was observed that whether the caregiver is female or male did not make a statistically signi cant difference. It was also observed that caregiver's own comorbidities did not signi cantly affect the results.
When the factors affecting the ZCBI score were examined; the patients' ECOG ndings were found to have a 6.631 times (β= 6.631), the caregiver age were found to have 0.384 times (β= 0.384), and SF-36 GH scores were found to have 0.645 times (β= 0.645) positive effect (p <0.001 -p = 0.042). It was determined that the transportation factor had a negative correlation of -8.373 times (= -8.373) and SF-36 M scores had a negative correlation of -0.443 times (β = -0.443) (p = 0.038) (table 5).
When the correlation between SF-36 results and ZCBI caregiver burden results was examined; Spearman correlation analysis showed moderate to strong negative correlation between ZCBI and all SF-36 results (p <0.001, r: -646 to -749) (table 6).
The distribution of ZCBI score by age is shown in Figure 1. With the increase in the age of the patient and the age of the caregiver, the ZCBI scores also increased. Roc curve analysis was performed on the results of the ZCBI caregiver burden questionnaire and the cut off age of 47 resulted in the sensitivity of % 69.57 and the speci city as % 65.52 (AUC = 0.671, P = 0.008) ( gure 2).

Discussion
In our study, the caregiver burden and the patients' results of the SF-36 questionnaires in metastatic prostate cancer patients were examined in terms of the clinical and socioeconomic conditions of the patients. To our knowledge, this study is the rst to comparatively evaluate the quality of life, diseaserelated clinical factors, caregiver characteristics and socioeconomic status in prostate cancer patients as a whole.
The presence of comorbidities, high ECOG score, the number of bone metastases and the presence of bone related complications were signi cant patient-related factors that negatively affected the survey results. In a study examining anxiety and quality of life in prostate cancer patients, it was revealed that the quality of life was negatively affected in patients with high degree of anxiety related to cancer [19]. In the quality of life assessment performed by Vieira et al. to assess pain in patients with bone metastasis; it was determined that most of the female patients had metastatic breast cancer and the male patients had metastatic prostate cancer. It was observed that male patients had to cope with more severe pain at an older age and their quality of life was found to be lower (Vieira C).
It was observed that the results of the questionnaire were negatively affected by the treatment-related factors, chemotherapy and palliative radiotherapy. Uemura et al. Showed that the use of new generation antiandrogens and an symptom control focused active patient follow-up procedures are associated with higher quality of life and lower care burden in bone metastatic prostate cancer patients [20]. Since our study was crossectional, whether the patients had received active chemotherapy in the last 6 months or less (including the time while data was collected) was evaluated. Consequently, higher caregiver burden and lower quality of life can be observed in cases experiencing chemotherapy side effects. However, it is clear that chemotherapy improves the quality of life in eligible patients in castration resistant metastatic prostate cancer [21]. There is still controversy about the timing of chemotherapy with new generation antiandrogens in sequential treatment algorithms. This is not a focus of our study.
Among socioeconomic factors, it has been observed that the obligation to use public transport, being a resident in the village and being a tenant negatively affected the survey results at various levels. Previous studies described that the caregivers of patients with chronic diseases are exposed to nancial, psychological and physical burden [3,4,6]. Caregivers might have to quit working, reduce working time, and loss income due to hospital visits or hospitalizations [22]. Soen et al. evaluated the caregiver burden of families of patients with osteoporotic fractures. Sixty-one percent of caregiver family members developed nancial problems such as not being able to continue their profession as before and they experienced economic stress [23].
Although there was no strong relationship between caregiver-related factors in our study, the condition that the caregiver was someone other than a spouse, sibling or child could be interpreted as a negative factor. Caregivers without social support can feel isolated and alone in the care giving process. It could have resulted in higher levels of stress and family con icts, further exacerbating this condition [5]. Faronbi  Our study looks to provide critical insight into how to relieve the economic burden laid upon advanced cancer patients, their caregivers and the healthcare system. There are few studies that shed light on the more systematic transfer of health budget resources to the needs of patients and their families. In their study in which they evaluated 832 bone metastatic patients diagnosed with castration resistant prostate cancer, McKay et al. detected symptomatic skeletal events in 207 patients during the cohort. The relationship between bone events, which mostly result in emergency admission with poor quality of life, care burden and economic burden has been shown. They highlighted the importance of palliative approaches and the rational use of cancer therapies. [27]. In their studies addressing the subject in a similar context, Uemura et al. showed that the symptomatic bone event is associated with increased healthcare costs and analgesic use [20]. At this point, the support needs of routine follow-up patients can be measured with systematic measurement methods. We believe that a more accurate management of health economics' orientation towards patient service will be possible with this method. The method in which ZCBI caregiver burden and SF-36 are evaluated together has proven to be useful for measuring the burden of the family in coping with the disease. Healthcare services in our country are primarily provided by the state. We think that it is critical to direct both the state and private health services correctly. Considering the fact that patient demands are effective in addition to objective medical examination results, active use and monitoring of scales in routine polyclinic functioning, as we did in this study, may be bene cial. It can be rational and e cient to recruit employees who are specialized in this matter.
The most important limitation of our study is that it is cross-sectional. A prospective study following a long course of treatment could more objectively evaluate the effect of therapeutic factors. In addition, the fact that it is a single center study can be considered as a limiting factor in terms of not being able to examine different geographical and cultural habitats.
In conclusion; it was found that the quality of life in patients with metastatic prostate cancer was directly related to caregiver burden. Factors such as patient comorbidities, bone metastasis, ECOG score, development of bone incident and associated palliative radiotherapy requirement, presence of visceral metastasis, being under active chemotherapy, low socioeconomic status, and the condition of caregiver not being a spouse, sibling or extra-foster relatives negatively affect both the quality of life and caregiver burden. If a social assistance system can be implemented in which socioeconomic conditions and patient-related factors can be monitored more systematically, objective and cost-effective bene ts related to the care of these patients and the maintenance of their quality of life can be achieved.