Quality of Life and its associated factors amongst Patients with Brain Tumors at a Tertiary Care Hospital in Pakistan: An Analytical Cross-Sectional Study

This study aimed to measure the QoL in patients with primary brain (PBTs), and assess its association with mental health and A cross-sectional survey was conducted among primary brain tumor patients. QoL was measured using European Organization for Research and Treatment of Cancer Quality of Life Questionnaire. Resilience was assessed by Wagnild and Young's Resilience Scale, mental health outcomes using Hospital Anxiety and Depression Scale, and social support using the Enriched Social Support Instrument.


Introduction
Though malignant primary brain tumors (PBTs) are responsible for 2.7% of cancer deaths globally [1], both the incidence and mortality rate of PBTs are lower in developing countries [2]. In Pakistan, a South Asian lower-middle-income country, malignant PBTs comprise around 3.6-3.9% of all malignancies, with a higher incidence and mortality rate in males [3,4]. Patients with PBTs often suffer from clinically signi cant psychological distress [5], the levels of which are higher than that experienced by patients suffering from most other types of malignancies [6].
Quality of life (QoL) is an increasingly important outcome in clinical neuro-oncology that holistically encompasses functionality and wellness in the psychosocial, emotional, and physical domains [7]. Most patients with PBTs face distress across these domains, due to factors such as physical disability and symptomatology, dis gurement, sensorimotor de cits, losses of individual freedoms, employment, and income, and social stigma [6]. Apart from the clinical and physical domains, existing literature suggests that QoL is majorly affected by psychosocial factors, including mental health, resilience, and social support. Mental health outcomes, particularly depression, correlate strongly with poorer QoL in patients with PBTs [8]. Resilience, which enables patients to maintain stable functionality throughout the disease course, may help protect against adverse mental health outcomes and improve QoL in patients with PBTs [9,10]. Similarly, social support is also associated with better QoL [11,12].
Although QoL of patients with PBTs has been a subject of global interest since the late 20 th century [13], there is a notable lack research from Pakistan to date [10]. Findings related to QoL, and its associated psychosocial factors, are likely to differ in Pakistan, due to a unique economic, social, and cultural landscape [10]. As survivorship of patients with PBTs increases as a function of rapidly evolving therapies, QoL is being increasingly measured as an additional endpoint alongside traditional clinical outcomes in disease management and clinical trials [14]. There is an acutely increasing need to assess the QoL experienced by patients with PBT in Pakistan using widely accepted standardized tools. The EORTC QLQ-C30 (European Organization for Research and Treatment of Cancer Quality of Life Questionnaire) and its brain tumor-speci c module EORTC QLQ-BN20 (EORTC QLQ-Brain Neoplasms 20) have proved to be brief, reliable, and valid assessment measures in this regard [15,16]. Thus, this study aimed to measure the quality of life in patients with PBT in Pakistan, and assess its association with mental health outcomes, resilience, and social support.

Study Design and Setting
This cross-sectional study was conducted at the Aga Khan University Hospital (AKUH), from November 2019 to May 2020. The AKUH is a Joint Commission International Accreditation (JCIA-accredited) notfor-pro t private sector hospital, in Karachi, Pakistan. AKUH serves as a major quaternary care hospital in Karachi and the surrounding regions in the province of Sindh. The protocol of this study has been approved by the institutional review board at AKUH (Reference Number: 5154-Sur-ERC-17) and is published by the authors with reporting in accordance with the STROBE statement [10].

Study Participants
We recruited all adult (≥18 years) patients treated for brain cancer at AKUH, provided that they gave informed consent and were currently ≥ 4 weeks post-initiation of treatment. Furthermore, we only considered patients who had been living in Pakistan for at least the past 3 months, as we wished to investigate the relationship of social support in the context of Pakistan with QoL. Patients residing abroad could possibly be accustomed to sociocultural environments that differ to those experienced by patients in Pakistan, which may in turn affect their social support. In addition, patients with any known psychiatric history or on psychiatric medications, or with debilitating concurrent illnesses such as kidney failure or stroke, were excluded. However, patients were not excluded if they had hypertension, type 2 diabetes mellitus or cardiovascular disease, As the prevalence of these co-morbids is relatively high in Pakistan, including in patients with brain cancer [17], excluding patients on the basis of these comorbids would have led to a non-representative sample and hindered minimum sample size achievability.

Sampling Technique
A minimum sample size of 250 was calculated using the one population mean formula, a standard deviation of 20, a 5% level of signi cance with precision of 2.5, and by in ating by 10% to account for incomplete responses. Nonprobability consecutive sampling was employed to recruit patients. Members of the research team screened all patients with brain cancer who were presenting to the neurosurgery or oncology clinics at AKUH for their scheduled appointment. If deemed eligible, patients were thoroughly briefed regarding the study and then requested for informed consent for participation. If patients agreed, they were administered the survey questionnaire by the data collector.

Data Collection Tools
The nal questionnaire consisted of the following components: Sociodemographic and Clinical Characteristics: Participants' sociodemographic details were recorded, with monthly household income reported in both Pakistani rupees (PKR) and United States dollars (US$). Patients' role in their family, with regards to decision-making, was also recorded. Clinical data included tumor type, and type of surgical intervention and adjuvant therapy, if any.
EORTC QLQ-C30: This 30-item QOL measure for patients with cancer. It comprises ve functional scales (physical, role, cognitive, emotional, and social), three symptom scales (fatigue, pain, and nausea and vomiting), a global QOL scale, and single items for measurement of other symptoms frequently experienced by cancer patients, in addition to the perceived nancial impact of the disease and treatment [15]. All items are scored using a 4-point Likert scale (1: 'not at all'; to 4: 'very much'), except for two items in the global QOL scale which instead employ modi ed 7-point linear analog scales. The functioning and global QOL subscales are scored ranging from 0 to 100, where higher scores imply favorable conditions. However, while symptom subscales are also scored ranging from 0 to 100, higher scores in these subscales imply greater symptoms i.e., unfavorable conditions. EORTC QLQ-BN20: This 20-item QOL measure is speci cally for patients with primary brain neoplasms [16], and comprises four domains relevant to the disease (future uncertainty, visual disorder, motor dysfunction, and communication de cit), in addition to seven single items. All items are scored using a 4-point Likert scale (1: 'not at all'; to 4: 'very much') and are then linearly converted to a 0-100 scale, where a higher score implies unfavorable conditions. The translation and in-depth validation of the EORTC QLQ-C30 and BN20 in Urdu has been performed by the authors in accordance with the EORTC guidelines. Psychosocial Characteristics: The Hospital Anxiety and Depression Scale (HADS), a 14-item tool using a 4-point ordinal scale, was used to measure patients' depression and anxiety [10]. Patients' social support was evaluated using the Enriched Social Support Instrument (ESSI) [10]. For depression and anxiety, the score obtained on HADS, was categorized as follows: 0-7 (Normal); 8-10 (Mild Depression/Anxiety); and 11-21 (Symptomatic Depression/ Anxiety). A score ≤ 18 on the ESSI was classi ed as Low Social Support, and > 18 as High Social Support. The Urdu translations of both the HADS [22] and ESSI (CVI for relevance and clarity of 0.95 and 0.97 respectively; and Cronbach's alpha 0.8246) [23] were used.
The nal Urdu questionnaire was pretested on 10% of the sample size to elucidate any ambiguities. However, no major changes were needed on the basis of the pretest.

Statistical Analysis
The data was analyzed using STATA version 15. Continuous variables were reported as mean and standard deviation (SD) or median and range, while categorical variables were reported as frequency and percentages. The means and standard deviations of the QoL scales were calculated according to the EORTC QLQ-C30 manual. General linear model (GLM) multivariate analysis of variance (MANOVA) was used to determine the association between independent factors and the QLQ-C30 and BN20 scales. The GLM-MANOVA approach is used to test the association between interrelated dependent variables (QoL scales) and independent variables. The independent variables considered were demographics (age, gender, monthly income, and working status), comorbids (HTN, T2DM, and CVD), use of addictive substances (tobacco and alcohol), family history of cancer, tumor and treatment related factors (type of tumor, type of surgical intervention, and adjuvant therapy), social support, resilience and depression, and anxiety. Variables with a p-value < 0.20 on the one-factor models were entered into the multi-factor model.
Linear regression was used to determine the factors associated with global QoL, and unadjusted and adjusted beta coe cients, standard error (SE), and 95% con dence interval (CI) were reported. All plausible interactions were assessed, including household monthly income and social support, while also considering all aforementioned independent variables. Variables with a p-value < 0.25 on univariate models were included in the multivariable linear regression model. A p-value < 0.05 was considered signi cant for all analyses.

Study Participants
Our study included a total of 250 patients, with median age of 42 years (range 33-54 years) and 169/250 (67.6%) males. The vast majority (91.6%) had received a formal education. Urdu was the most common mother-tongue (30.8%), with Sindhi (18.8%) and Punjabi (14.8%) being others. Most patients (82%) were married, and around half of them (54%) lived in extended families. About one-third (30.8%) of patients reported being their family's head/chief decision-makers. However, half (50.8%) of patients were not currently working, and two-thirds (66.8%) had non-working spouses. The median monthly household income was PKR 60000 (30000-100000)/USD 374 (187-623). The patients' sociodemographic characteristics are shown in Table 1 of the previous publication by the authors based on analysis of the same sample [24].

Participants' Medical History and Tumor Characteristics
Hypertension (HTN: 24%) and type 2 diabetes mellitus (T2DM: 14.8%) were the most common comorbids, while 6% of patients currently smoked and 2.8% currently used smokeless tobacco. The commonest brain tumors were glioma (46.8%), meningioma (21.2%), and schwannoma (4.8%). Most patients underwent only a biopsy (78%), while 4.4% underwent total resection only. Adjuvant therapy was received 39.2% of patients. Around half (55.2%) of patients were still receiving treatment for their brain tumor, while the rest had completed treatment (Table 1). A family history of brain or other cancers were reported by 4.8% and 16% of patients, respectively, and only 4% of patients reported a family history of depression.

Resilience, Social Support, and Mental Health
Most patients reported high social support (97.6%) and normal depression (90%) and anxiety (91.6%) scores. The mean resilience was 83.71 ± 9.32, which is categorized as high ( Table 2).

Quality of Life
The mean global QoL score was 75.73. The values of the ve functioning scales ranged from 81.2 (role functioning) to 87.93 (social functioning). In the symptom scales, pain had the highest score (29.7), followed by appetite loss (26.13), insomnia (25.06), and nancial di culties (21.06). The two worst symptoms on the BN-20 symptom scale were headache (21.46) and weakness (18.40). These are detailed in Table 3.

Variables associated with QoL outcomes
In the rst step of the GLM-MANOVA, the association between the independent variables (detailed in Methods: Statistical Analysis) and the dependent variables (15 scales of QLQ-C30) was investigated (one-factor model). On the one-factor model, 17 variables were associated with the overall outcome of QLQ-C30 with p < 0.20. These 17 variables were entered into the multifactor model analysis. Variables signi cantly associated on the multifactor model were age, type of intervention, adjuvant therapy, use of feeding tube, use of a urine catheter, resilience, depression, anxiety and social support (all p < 0.05). In the one-factor model with QLQ-BN20 as the outcome, 17 variables were associated with the overall outcome on QLQ-BN20 at p < 0.20. In the multifactor model, role in family, smoking, smokeless tobacco use, family history of cancer other than brain, resilience, depression, anxiety and social support remained signi cantly associated (p < 0.05). Results of the GLM-MANOVA are shown in Table 4.

Discussion
Patients with PBTs suffer considerable physical, cognitive, and emotional distress due to their disease course, which negatively impacts their QoL. Our study is the rst to explore factors associated with QoL in patients with PBTs in Pakistan and provide a much-needed perspective on this increasingly important outcome of cancer management within the sociocultural setting of a LMIC.
In our study, the mean global QoL (75.73) and scores on the ve functional subscales (81.20-87.93) as measured by the EORTC QLQ-C30 are higher than a global weighted mean calculated using data from Austria, Germany, France, Turkey, Canada, The Netherlands, Iran and India [25]. In addition, all symptom components on the QLQ-BN20 were lower in our study, barring weakness of legs (mean of 18.40 ± 25.33 in our study vs. weighted mean 17.72 ± 92.99) [25]. It is possible that these ndings of an overall better QoL in our sample are a function of the higher resilience and social support, and lower depression and anxiety, which contribute towards better QoL [21]. The higher social support, in turn, is likely due to the close-knit extended family systems that are a feature of South Asian households (54% of patients in our study lived in extended families). However, our study reported higher scores than the global weighted mean on 5/9 of the symptom scales on the QLQ-C30 (pain, insomnia, appetite loss, diarrhea, and nancial di culties). An interesting theory is that since the population of a developing country like Pakistan is generally accustomed to a lower standard of living and QoL than that of a developed country, the drop in QoL due to an illness may be perceived as less signi cant despite similar or worse symptomatology. Interestingly, the male predominance and median age of 42 years (range 33-54 years) represents the demographic that encompasses the breadwinners of most Pakistani families, which could explain the greater nancial di culties in our sample.
The results of multivariable linear regression in our study revealed comorbid hypertension, use of a urine catheter, lower family income, low social support, and coexisting anxiety and depression to be associated with a poorer global QoL, while higher resilience scores was associated with a better global QoL. These ndings are self-explanatory and supported by previous literature. Hypertension, chronic need for a urinary catheter, and low socioeconomic status are known to be independently associated with poorer quality of life [26-28]. Greater social support and resilience [29] have been found to be associated with better QoL in patients with PBTs [12,30], while poorer mental health, particularly anxiety and depression, is associated with worse QoL [31]. Our ndings are of relevance to the holistic management of patients with PBTs in Pakistan and can aid the development and integration of QoL-improving interventions into the routine care of patients with PBTs. In addition, our results also make the case for periodic screening of patients with PBTs for adverse mental health outcomes, so that these may be identi ed and treated early, and their harmful impact on QoL be negated.
As AKUH is one of the largest quaternary care hospitals in the largest city of Pakistan, the sample recruited in our study was representative of all major ethnicities and economic classes in the country, which boosts the generalizability of our ndings. In addition, the Urdu translations of the tools used in this study were all su ciently validated in a Pakistani population. Importantly, patients identi ed as symptomatic for anxiety or depression were referred to a psychiatrist, whereas those who were mildly anxious or depressed were provided with on-spot counseling by a trained psychologist. Lastly, as this was a cross-sectional study, we were unable to draw conclusions with regards to the temporal relationship between QoL and associated psychological factors such as depression, anxiety, and resilience.

Conclusions
The quality of life of patients with primary brain tumors in Pakistan is a function of clinical factors such as comorbid disease and use of a urinary catheter, social factors such a family income and social support, and psychological factors such as mental illness and resilience. Our ndings may be of use in the development of QoL-improving interventions within the sociocultural setting of Pakistan.

Declarations
Ethics approval and consent to participate: Ethics approval was sorted from Aga University Ethical Review Committee ERC #5154-Sur-ERC-17 and informed written consent was Consent for publication: Consent for publication was obtained from the study participants Availability of data and materials: The datasets used and/or analysed during the current study are available from the corresponding author on reasonable request.
Competing interests: The authors declare that they have no competing interests.
Funding: The work is funded by Aga Khan University's SEED Money grant Award Number PF 89-1016 and the funders had no role in study design and development of proposal Authors' contributions: NZ conceived the study, analyzed data, and critically reviewed the manuscript. RSM drafted, revised, and critically reviewed the manuscript. WZ and IA contributed to data analysis and critically reviewed the manuscript. AH, KA, SSB, and NA overlooked the study and intellectually contributed to the manuscript. AAJ, MSS, RJK, GJ, EB and SAE were the subject experts and contributed to the design of the study. All authors have contributed intellectually to this manuscript and have read and approved the nal manuscript