Sleep Quality, Cancer-Related Fatigue, and Health-Related Quality of Life Among Omanis Hospitalized Patients with Cancer: A Cross-Sectional Study.

Purpose: This study aimed to examine the difference in HRQoL by participants’ characteristics and to investigate the determinants of health-related quality of life HRQoL among Omani hospitalized patients with cancer. Methods: This cross-sectional study was conducted in two oncology centers in Oman. Omani hospitalized patients with breast, thyroid, colorectal, stomach, and prostate cancer were recruited using convenience sampling. Participants completed Arabic versions of self-reported questionnaires: Pittsburgh Sleep Quality Index, Brief Fatigue Inventory scales and Functional Assessment of Cancer Therapy. Descriptive and inferential statistics were performed. T-test, ANOVA and multiple linear regression analysis was utilized to determine predictors of HRQOL. Results: in total, 275 participants were recruited (Mean age= 52 years). About 64 % of the participates reported poor sleep and 18.5% reported severe cancer-related fatigue. Sleep quality, cancer-related fatigue, age, and prostate and thyroid cancer were signicant predictors of HRQoL (F (5, 269) = 26.26, p < 0.000) and they explained 33% of the variances in the HRQoL (R 2 = .328). Conclusions: This study highlights the impact of sleep quality and caner-related fatigue on the HRQOL among Omanis hospitalized with cancer. Thus, sleep quality and cancer-related fatigue should be assessed routinely during hospitalization of oncology patients with special attention patients’ age.


Introduction
Globally, over the past century, there has been a dramatic increase in cancer incidence and mortality, making it the second leading cause of death after cardiovascular disease [1]. Cancer mortality expanded by 25% from 1990, with an anticipated frequency of 23.6 million cases annually by 2030 [2]. The global cancer burden estimated to have risen to 18.1 million new cases and 9.6 million deaths in 2018 [3].
Cancer has been reported that one in six women and one in ve men will develop cancer during their lifespan [4]. In the Eastern Mediterranean region (including the Middle East), about 555,318 cancer cases were reported in 2012 [5]. The number of cancer cases is estimated to double in nearly 20 years, meaning that the Eastern Mediterranean is expected to rank the highest relative increase in the world [5]. In Oman, 30% of the mortality rate is attributed to neoplasm; the crude cancer incidence rate among Omanis was 63.91 per 100,000 for men and 74.88 per 100,000 for women [6]. Furthermore, for solid cancer, breast cancer is the most common type of cancer among Omani female (24.48%), followed by thyroid (15.47%) and colorectal (7.51 %). Among Omani males, prostate cancer was the most common solid type (11.08%), followed by colorectal (10.95%), and stomach (6.00%) [6].
Universally, cancer burden and its management have been linked to consequences such as nancial burden, increased morbidity, impaired quality of life, and premature death [7]. Cancer impacts on patients' physiological, psychological, and social status have been the subject of concern lately [8]. Sleep quality and cancer-related fatigue are two signi cant physical consequences experienced by adult patients with Page 3/18 cancer affecting their quality of life and disabling daily functioning, memory, and concentration [8].
Health-related quality of life (HRQoL) is commonly used to observe the in uence of health status on the general quality of life and is used as a primary outcome measure in studies evaluating the success and effectiveness of cancer treatment [9]. Poor sleep quality is known to be a dominant concern in about 60% of patients with cancer [10] with a potential effect on patients' HRQoL, which has been coupled with a decrease in functional performance and psychological status of patients with cancer [11]. Like poor sleep quality, severe cancer-related fatigue is commonly found among 25% to 99% of patients with cancer [12].
Cancer-related fatigue is a common side-effect of cancer and its treatment that is regarded as a risk factor leading to a reduction in cancer survival and HRQoL [12] and treatment suspension [13].
Similar to other Middle Eastern countries, patients' characteristics, sleep quality, and cancer-related fatigue and their in uence on HRQoL among patients with cancer have not received adequate attention in Oman. The aims of this study were to examine 1) difference in HRQoL by participants'' demographical and clinical characteristics, sleep quality, and cancer-related fatigue, and 2) determinants of HRQoL among hospitalized Omani patients with cancer.

Methods
A descriptive correlational cross-sectional was carried out in two oncology centers in Oman, which are considered the country's main oncology centers that provide management to all patients with cancer. The study was conducted from June to October 2019.
We used a non-probability convenience sampling to recruit patients with cancer. The sample size was determined based on the numbers of independent variables included in multivariate regression analysis, which was recommended to be 10-20 participants per variable [14]. As a rule of thumb, we counted 15 participants per variable (total of 18 variables). Accordingly, the required sample size was 270 participants. The total participants included in the study were 275.
Participants were included if they were: Omani aged ≥ 18 years, hospitalized, diagnosed with cancer (i.e., breast, thyroid, colorectal, stomach, prostate cancer), aware of cancer diagnosis, understand and read Arabic, and willing to participate in the study. Participants who were critically ill, unable to complete the survey, and had cognitive impairment were excluded from the study.
Participants admitted in the oncology, and surgical units were screened for eligibility using a patient list obtained through electronic medical records. Eligible participants were initially contacted by the principal investigator (PI) to inform them about the study purpose and got an informed written consent form. Voluntary participation and refusal to participate would not affect their treatment was assured. Participants self-completed the questionnaire. The clinical data were obtained through medical records by the PI. Data were collected from two main oncology centers, representing 11 governorates of Oman, from June to October 2019. Permission to use all questionnaires described hereinafter was obtained from the primary developer. The study was approved by the Ethics and Research Committees of both centers: Sultan Qaboos University and Ministry of Health (CON/EA/26/2019, SRC#46/2019), respectively. The study was performed in line with the principles of the declaration of Helsinki.
All participants completed the Pittsburgh Sleep Quality Index (PSQI), the Brief Fatigue Inventory (BFI), and Functional Assessment of Cancer Therapy (FACT-G) questionnaires. In this study, sleep quality was assessed by the Pittsburgh Sleep Quality Index (PSQI) [15]. PSQI is a self-report measure with evidence of good internal homogeneity, test-retest reliability, validity, sensitivity (89.6%), speci city (86.5%), and Cronbach's alpha of 0.83 [15]. The PSQI distinguishes "poor" from "good" sleep that has been used in the cancer population widely [16] and the Arab cancer population, speci cally [17]. PSQI is rated on a 0-3 Likert-like scale, whereby 0 (no di culty) to 3 (severe di culty). In a clinical cancer population, a cutoff score of ≥ 8 suggests a poor sleep, while a score of ≤ 7 indicates a good sleep quality [15,18]. The PSQI Arabic translated version showed internal consistency of 0.74 in the Arabic cancer population [19]. In the current study, Cronbach's alpha of PSQI was 0.63, and we categorized sleep as poor quality (score ≤ 7) or good quality (score ≥ 8).
Cancer-related fatigue was assessed using the Brief Fatigue Inventory scale (BFI), a reliable tool established for the rapid assessment of fatigue severity in clinical screening for patients with cancer [20]. The scale's items are rated numerically from 0 to 10 and consists of nine items; the rst three assess fatigue severity (0= No fatigue and 10= as poor as you can imagine for the rst 3 items), while the remaining six items assess fatigue interference with general activity, mood, walking, work, relation and enjoyment (0= Does do not interfere, 10= completely interfere). The BFI psychometric evaluation supports its internal consistency and constructs validity with Cronbach's alpha of 0.96 [20]. BFI scoring is categorized into two groups: 1) severe fatigue (score of 7-10) and 2) non-severe fatigue (score of 0-6) with a higher score indicating higher fatigue severity. In the present study, fatigue was categorized as non-severe (score of 0-6) and severe (score of 7-10). The Arabic translated version BFI showed good internal consistency and convergent validity with a Cronbach alpha of 0.86 [18]. In the present study, BFI Cronbach's alpha was 0.89.
Health-related quality of life (HRQoL) was measured by using version 4 of the FACT-G that was developed by David Cella and to be used in the oncology population [21]. FACT-G has been administered in several types of cancer, validated with reliability Cronbach's alpha of 0.88, and translated into different languages [22]. FACT-G includes four domains: physical, social and family, emotional, and functional [21]. FACT-G version 4 has 27 items; each answered using a 5-point Likert Scale ranging from 0 (Not at all) to 4 (Very much). Total scale and subscale scores are summed to yield 108 points, with the higher scores indicating better HRQoL [22]. The current study showed a Cronbach's alpha of the FACT-G of 0.84 for the total score.

Data analyses
Statistical Package for the Social Sciences (SPSS) Version 24 was used for data entry and analysis. In this study, data were tested for normality by using Histogram, Plots, Skewness, and Kurtosis where all data were normally distributed. For the study variables description, continuous variables presented with mean and standard deviation, while categorical variables were presented with frequencies and percentages. The categorical association was compared by using the Chi-square test, and continuous association was compared by using the T-test, Pearson correlation, and One-Way Analysis of Variance (ANOVA). To examined the determinants of HRQoL, a multiple linear regression analysis was used.
Multiple linear regression assumptions (i.e., normal distribution, linearity, and multicollinearity) were examined and not violated.

Sample demographical and clinical characteristics
A total of 275 participants were included in this study, with a mean age of 52 years (SD= 14.1). The majority were female (65.1%), had chronic diseases (56%), and had a family history of cancer (37%). The mean total score of FACT-G was 73.0 (SD= 16.2) above the midpoint of 54. Among HRQoL domains, the social and family well-being scored the highest mean (Mean= 21.1, SD= 5.3), and physical well-being scored the lowest mean (Mean=15.6, SD= 17.0). The majority of participants had breast cancer (41.1%), were at stage IV (45.1%), and in the rst year of cancer diagnosis (70%). Out of 275 participants, 64.4% were anemic and on chemotherapy (43.3%). The majority (64.4%) reported poor sleep quality, and 18.5% reported non-severe cancer-related fatigue (Table 1).
Health-related quality of life by patient's demographical and clinical characteristics, sleep quality, and cancer-related fatigue We used Pearson correlation, t-test, and ANOVA analysis to examine the difference in HRQoL by patients' characteristics, sleep quality, and cancer-related fatigue. The ndings showed that age was signi cantly and positively related to HRQoL (r = 0.25, p<0.001) re ecting as age increased, HRQoL is getting better. In  Table 2).

Determinants of health-related quality of life
A multiple linear regression model was used to examine the determinants of HRQoL. Variables that in uenced HRQoL signi cantly (i.e., income, educational status, age, family history of cancer, chronic disease, sleep quality, cancer-related fatigue, cancer stage, cancer type, years of disease diagnosis, anemia status, and chemotherapy status; table 2) were only entered into the regression model. The multiple linear regression model showed that age, sleep quality, cancer-related fatigue, and cancer type were signi cant determinants of HRQoL. The model was signi cant, (F (5, 269) = 26.26, p < 0.000) and it had explained 33% of the variance in HRQoL (R 2 = .328). The model showed that HRQoL is worse in patients with advanced age (B = -0.266, p= 0.000) and patients who had poor sleep quality (B = -11.387, p= 0.000) and severe cancer-related fatigue (B = -6.639, p= 0.003). Furthermore, the model indicated that HRQoL is signi cantly better in patients with prostate and thyroid cancer (B =8.466, p= 0.001; B = 6.225, p= 0.047), respectively. (Table 3).

Discussion
This study was conducted in Oman. It included 275 participants diagnosed with breast, thyroid, colorectal, stomach, or prostate cancer. The main aim of the study was to examine the determinants of HRQoL in hospitalized patients with cancer in Oman and to investigate if HRQoL varied by patients' demographical and clinical variables, cancer-related fatigue, and sleep quality. Findings of the study showed that HRQoL differed signi cantly by patients' demographical and clinical characteristics, sleep quality, and cancer-related fatigue. Furthermore, results showed that age, poor sleep quality, severe cancer-related fatigue, and being diagnosed with thyroid or prostate cancer were signi cant determinants of HRQoL.
Our ndings showed that 64% of Omani patients with cancer reported poor sleep quality, which is similar to a Lebanon study that showed poor sleep quality was high among patients who had non-metastatic breast cancer during active treatment [17]. It is also consistent with ndings of a systematic review that reported about 30-70% of newly diagnosed patients with breast, thyroid, colorectal, stomach, and prostate cancer had poor sleep quality [23]. Various factors suggested the growth of poor sleep quality among patients with cancer, of which, cancer treatments modalities, the frequent awakening of patients in midnight and early morning for a nursing procedure or therapy administration, breath discomfort, hospital environmental noise, pain, and worries about illness and treatment are some examples in uencing sleeping patterns [17]. In Oman, these factors could also explain why the majority of patients reported poor sleep quality taking into consideration that about 82% of patients were receiving chemotherapy. This study established baseline data about sleep quality in patients with cancer in Oman. Our nding highlights that health care professionals should pay attention to patients' sleep patterns and quality and identify speci c factors that can be managed to improve patients' sleep quality and, subsequently, improve their health-related quality of life.
The mean total score of HRQoL was 73.0 (SD=16.2), which was above the midpoint of 54. Physical wellbeing, which involves issues related to pain, nausea, low energy, feeling ill, treatment's side effects, and trouble meeting family's needs, showed the lowest score among HRQoL domains, and physical well-being seemed to pose the most signi cant challenge for patients with cancer in Oman. A similar result was found in Saudi Arabia, a middle eastern Arabic country, for patients with solid cancer [24]. A Low physical well-being score could be related to the fact that the majority of the Omani patients included in this study were in stage III and IV, were anemic, and had poor sleep quality, all of which have been found to in uence the overall HRQoL in current study signi cantly. Additionally, in this study, social and family well-being domains (feeling like being close to friends, family and friend's support, family cancer acceptance, feeling close and getting enough support and love from the partner) were reported the highest score compared to the other three domains. This nding is parallel to a previous Omani study among oncology population that found Omani patients said that their relationships with husbands and family members became more substantial and more supportive after diagnosis with cancer [25], which could explain our ndings considering that 73.8% of patients were married. This nding also supports the notion that social support can improve distress and alleviate coping skills among Omani women challenged with cancer and its management [26]. This nding denotes that healthcare providers should improve patients' physical wellbeing and correlate this domain to patients' sleep quality.
In this study, we found that poor sleep quality, severe cancer-related fatigue, age, and type of cancer (prostate and thyroid) signi cantly determine HRQoL and explained about 33% of the variance in HRQoL. This study had found that those participants with poor sleep quality had decreased HRQoL in general. This nding is in line with the literature that showed sleep quality as one of the main predictors of HRQoL; for instance, Ha and colleague (2019) reported that sleep quality was an important determinant of HRQoL in patients with lung cancer; and that poor sleep quality, dyspnea, fatigue and depression were related to lower HRQoL, and together accounted for 85% of variances in HRQoL [27]. Likewise, a Nigerian study among women with breast cancer with lower sleep quality showed a signi cant reduction in HRQoL [28]. In the Arab world, a study in Palestine reported that sleep quality was one of the HRQoL determinants and, together with the pain, were responsible for 42% of the variance in HRQoL [29]. The positive and signi cant relationship between poor sleep quality and low HRQoL could be related to the negative impact of poor sleep on the patients' immune systems, making them more susceptible to infection and illness [30]. Another explanation could be linked to the fact that poor sleep tempts changes in the cognitive performance of the patients with cancer, which, as a result, in uence their HRQoL [31]. Besides, Poor sleep quality is connected with poor physical well-being, such as gastrointestinal dysfunction that aggravates sickness [32]. The current study extends the observations of the relationship between sleep quality and HRQoL from Western societies to the developing world, speci cally the Arabs Omani population, and enlightens future focus to promote screening for and improving sleep quality to enhance HRQoL community with cancer.
Likewise, cancer-related fatigue was found as another determinant of HRQoL in Oman, in which those with severe fatigue had a signi cant reduction in their overall HRQoL. Our nding is similar to studies from Nigeria, Greek, and France that were conducted in patients with breast and colorectal cancer and found that cancer-related fatigue was a signi cant determinant of HRQoL [28, 33,34]. In the Arab population, we couldn't identify literature examining the relationship between fatigue and HRQoL. Many reasons could explain the relationship between fatigue and HRQoL; one reason could be attributed to chemotherapy management and its side-effects [35], and this could explain our nding as about 43% of our participants were on chemotherapy. Other reasons could be due to the effect of cancer-related fatigue on patients' self-care behavior as they demonstrated less involvement in taking care of themselves, which as a result, affect their HRQoL [36]. Although cancer-related fatigue experienced by the participants in the current study was as low as 18.5%, nevertheless, cancer-related fatigue remains a disabling symptom, a common determinant of patients' HRQoL, and it may prolong to two years after diagnosis [33,34]. These ndings could inform the future direction to establish strategies to understand cancer-related fatigue and its attributing factors to enhance HRQoL of patients with cancer.
In this study, age was also found to determine HRQoL in the Omani oncology population, indicating that HRQoL is better among younger age. Our ndings were similar to results among cancer population from Sweden and Turkey [37,38], but inconsistent with a study in Yemen, an Arab Country, that reported no relationship [39]. This nding could be because as people get older, they demonstrated lower performance in activities of daily living, lower functional abilities, and more risk of depression, which in consequence, reduce the HRQoL and supporting our nding [37,40]. Our nding underlines the mounting need to focus HRQoL at a younger age and pay speci c attention toward senior patients. In the Omani context of Oman and based on our ndings related to the mean age of our participants (52 years), special attention toward improving HRQoL among older patients with cancer should be accentuated.
Thyroid and prostate cancer found to determine HRQoL in hospitalized Omani, and that those with thyroid and prostate cancer reported better HRQoL compared to other types of cancer. Our ndings were similar to a study from Iran [41] but inconsistent with studies from Turkey and France who reported no association [42,43]. The high survival rate of prostate and thyroid cancer compared to another type of cancer could explain the link between these types of cancer and HRQoL compared to other cancer such as stomach (70%) and colorectal (80%) [44].
This study should be considered within the following limitation; rst, use of a cross-sectional design limits the ability to establish causal relationships. Second, we use a convenient nonprobability sample, which could impede the generalizability of the ndings. However, a heterogeneous sample from the largest two tertiary health settings in Oman could minimize this threat. Third, Pittsburgh Sleep Quality Index (PSQI) instrument had below the satisfactory level of internal reliability (Cronbach's alpha = .67), which is similar to another recent Omani study that used PSQI among myocardial infarction patients that reported a Cronbach's α of 0.64 [45]. However, having an adequate sample size could provide robust statistical power and con dence in the results.
Sleep quality, cancer-related fatigue, and HRQoL deserve adequate medical attention regarding supportive care and routine assessment. Prompt recognition, observation, and documentation of these variables will guide the clinical teams to develop interventions to improve fatigue and sleep quality and their impact on the health-related quality of life. Holistic multidisciplinary cancer care is necessary to implement by clinicians inside the hospital and out of the hospital to improve all aspects of HRQoL of oncology population. On the other hand, an individualized approach should be maintained considering unique patients' characteristics such as age, cancer type, stage, and treatment modalities. Policymakers should design the oncology centers to empower patients' physical, functional, social, and emotional well-being. Sleep quality, cancer-related fatigue, and HRQoL can be regarded as quality indicators for patients with cancer. Finally, future studies should examine 1) strategies to improve sleep quality and cancer-related fatigue across the cancer population in general and the Arab community in speci c, 2) the same variables among patients with non-solid cancer like leukemia and 3) determinants of each subscale as this was not the scope of the current study.

Conclusion
Health-related quality of life is a multidimensional construct that necessitates further understanding. Sleep quality, cancer-related fatigue, and patients' clinical and demographical characteristics could impose critical determination of overall HRQoL.

Declarations Funding
No funding was received for conducting this study

Con icts of interests
The author(s) declared no potential con ict of interest with respect to the research, authorship, and/or publication of this article.

Ethics approval
The study was approved by the Ethics and Research Committees of both centers: Sultan Qaboos University and Ministry of Health (CON/EA/26/2019, SRC#46/2019), respectively. The study was performed in line with the principles of the declaration of Helsinki and was reported based on STROBE reporting guidelines for cross-sectional studies.

Consent to participate
Written informed consent was obtained for each participant according to national and institutional guidelines.

NA Data Availability
The data set used and/or analyzed during the current study are available from the corresponding author on reasonable request.

Supplementary Files
This is a list of supplementary les associated with this preprint. Click to download.