Factors Influencing Palliative Care Utilization Among Patients With Cancer: A Cross-Sectional Study

Background: Cancer remains a public health challenge in low-and-middle income countries (LMICs). Over 30% of those in need of palliative care (PC) worldwide suffer from cancer and 80% of patients with cancer in LMICs are diagnosed at advanced stages, by which time they can only benefit from PC services. The aim of this study was to determine factors influencing PC utilization among patients with cancer in KZN. Methods: A cross-sectional descriptive study was used among patients who had been diagnosed with different cancers and attending (IALCH), and Chatsworth Hospices. A systematic random sampling technique was used to select participants. Structured questionnaires were administered to a total of 394 patients. The questionnaire included two open-ended questions and the generated data was analysed and presented into two formats; firstly, the textual data was entered into REDcap (v.8.11.7) and then exported to IBM SPSS Statistics (v.25) and recoded for emerging themes which were later presented in a graph format. Secondly, the thematic analysis was presented in textual format and supported with verbatims quotes. Results: The mean (+SD) age of the 394 participants was 51.89 (+14.67) years and about 22.9% were within the age-group 41–50 years. Majority of participants were females (77.7%) and were unmarried (60.6%). Most of the participants’ racial classification was black (81.5%) and 64.5% had primary education. More than four-fifths (82.7%) were unemployed, 55.8% lived in urban areas and 24.6% had cervical cancer. The analysis of the two open-ended questions incorporated into the questionnaire revealed five themes as major from seeking PC services. Conclusion: Lack of knowledge and awareness about cancer symptoms and PC services could be addressed by better education of the public. Hence, utilization of PC could be maximized, thereby providing a better quality of life for the sufferers and their families until the end of life’s journey. The study also provides information that can be used to strengthen and/or to develop PC guidelines addressing PC issues.

3 questionnaire revealed five themes as major barriers preventing cancer patients from seeking PC services.
Conclusion: Lack of knowledge and awareness about cancer symptoms and PC services could be addressed by better education of the public. Hence, utilization of PC could be maximized, thereby providing a better quality of life for the sufferers and their families until the end of life's journey. The study also provides information that can be used to strengthen and/or to develop PC guidelines addressing PC issues. Keywords: Cancer, Palliative care, barriers, utilization, KwaZulu-Natal background Recent statistics revealed that cancers cumulatively rank as the leading cause of death worldwide, with the mortality rates outpacing that of HIV and AIDS, TB and malaria combined (1). The LMICs currently account for slightly more than half of all new cancer cases each year and these countries are projected to carry as high as over 80 percent of the global cancer burden by 2030 (2). However, only five percent of all global resources being spent on cancer are allocated for use in LMICs, therefore contributing to severe inequities in cancer management worldwide (1,3).
If the issue of resource allocation remains unaddressed, the rising rates of cancer will continue to impose a considerable burden on the health and socio-economic stability of sub-Saharan Africa (SSA) (4).
The increase in prevalence of cancer cases may be attributed to different factors including aging and an increase in the world population (5). An increase in the adoption of cancer-causing behaviours like smoking especially in the developing countries has amplified the prevalence of the disease (6). The result has been a double disease burden in developing countries as communicable diseases are still a 4 serious health problem that these countries are grappling with (7). In Nigeria, 60-70% of patients with cancer present for treatment late. This is mainly due to the belief that many patients have concerns regarding the origin of their conditions (8).
A lot of patients attribute their cancer to spiritual forces (9). Also, because many cancer patients die even after receiving conventional western medical treatment (10), other cancer patients tend to be discouraged from using western medicine, and therefore resort to complementary and alternative medicine (CAM) treatments, prayers or faith to obtain healing (11). Furthermore, poverty, lack of information and almost non-existent specialized care have also resulted in over 95% of cancer cases being diagnosed at the late or advanced stages of the disease (12,13). The increase in incidence of cancer has been exacerbated by increased incidence of HIV, most notably Kaposi's sarcoma (14,15). SSA health systems are mainly orientated towards care for acute illness and maternal and child health (16). For most of the cancer patients in Africa, the unavailability of (i) anticancer drugs, (ii) oncology specialists, (iii) radiotherapy centres and (iv) PC services means limited chances of survival and increased mortality rates (17). Although infectious diseases are still the leading cause of death in SSA, the emergence of non-communicable diseases (NCDs), notably cancer, presents a significant challenge for healthcare systems (18).
This contrast requires complex interventions that can be sustained throughout the continuum of care (19). PC offers a realistic public health approach to cancer care where most people with cancer present at healthcare facilities at the incurable disease stage, and where disease modifying treatment is not accessible to most of them (19,20). Paice et al., (20) highlighted that appropriate pain treatment and 5 access to analgesics are both essential in PC and should be prioritized as a public health matter, from a national and global perspective (21). If this is not done, an escalating proportion of the world's population will continue to live with pain unnecessarily and will furthermore die in pain (22,23). Therefore, the aim of this study was to determine factors influencing PC utilization among patients with cancer in KwaZulu-Natal (KZN), South Africa.

Study design and population
This was a cross-sectional study with a total of 394 cancer patients recruited during the period of November 2018 to March 2019 from the targeted healthcare facilities in KZN; namely: IALCH, Addington Hospital, Grey's Hospital, Msunduzi Hospice and Chatsworth Hospice. The criteria for patients to be considered for inclusion in this study were that they: (i) had to be aged 18 years and above; (ii) had to have a confirmed cancer diagnosis; and (iii) willing to participate in the study. The study sites were selected through purposive sampling (based on the oncology units offering PC services in the province). Study participants were recruited using the systematic random sampling technique until the calculated sample size was achieved.

Sample size
To assess the outcome of interest, assuming 95% confidence and an acceptable margin of error of 5%, and maximum variability, i.e. 50% (given unknown prevalence), a sample size of 384 participants was required. The sample was further increased to 394 to account for contingencies such as that of non-response or recording error. Increasing the sample size reduced the type I and type II errors and 6 known and unknown confounders' effects. The power (1-β) (the % chance of detecting difference) of the study was set at 80%.

Recruitment and data collection
Prior to the commencement of the study, nurses working in the oncology unit identified cancer patients in accordance with the eligibility criteria. All eligible participants who met the study criteria were enrolled into the study using a systematic random sampling technique. The systematic random sampling technique implies that the interval may be in terms of time, space or order. For instance, an element appearing after every 30 minutes, or present at a distance of two metres, or every 5th element present on a list, would be recruited (24). What are the main barriers that prevent the community members from seeking PC?
What do you think can be done by healthcare providers to make PC services in the healthcare facilities appealing?
The purpose of the study was explained, the study information sheet was shared, and the informed consent form was signed by the potential participants prior to administering questionnaires. The questionnaires were administered to each individual participant by the research team, using a private room, in order to ensure that their identity and confidential information were protected. The questionnaires took about 10-15 minutes to complete. None of the participants reported feeling discomfort or distress during the questionnaire administration.

Data analysis
The two open-ended questions from the questionnaire generated data that was 7 analysed and presented into two formats; firstly, the textual data was entered into REDcap (v.8.11.7) (25), exported to IBM SPSS Statistics (v.25) (26) and then recoded for emerging themes which were later presented in a graph format.
Secondly, the thematic analysis was presented in textual format and supported with verbatims quotes. Data was further checked for possible errors and missing values and any errors were corrected before analysis. For descriptive analysis, sociodemographic data was summarized into a table that included age, gender, marital status, ethnic group, level of education, employment status, geographical area and cancer diagnosis.

Participants' socio-demographic characteristics
The mean (+SD) age of the 394 participants was 51.89 (+14.67) years and about 22.9% were within the age group 41-50 years. The majority of participants were females (77.7%) and were unmarried (60.6%). Most of the participants' racial classification was black (81.5%) and 64.5% had primary education as their highest level of education. More than four-fifths, 82.7% were unemployed, 55.8% lived in urban areas and 24.6% had cervical cancer, the most common type of cancer among the participants (Table: 1).

Major barriers to palliative care utilization
Over a third (40%) of the participants reported lack of cancer knowledge and the pursuit of alternative treatment options, only a few reported stigma and discrimination (19%) and 4% reported long waiting periods for results of the 8 accurate diagnosis or to be attended to by the doctor and for the date of the next appointment. In addition, participants reported lack of screening and diagnostic equipment (1%) and 25% cited socio-economic barriers, which include transport issues and employment status as factors contributing to the underutilization of PC services in KZN (Fig: 1).
Furthermore, when participants were asked about what could be done to improve the uptake of PC services, the majority cited that the Department of Health (DoH) should allocate enough resources for cancer awareness campaigns in the communities (45%). A total of 27% recommended that the DoH should improve healthcare services and only a few suggested quality in-service training and continuous professional development (CPD) for healthcare providers (8%). The analysis reveals that the majority of participants believed that when the above mentioned is prioritized, the number of premature deaths would decrease and that CPD on PC would enhance healthcare providers' knowledge and practice (Fig: 2).

Qualitative Findings
The five major themes on factors that affect cancer patient in utilizing PC services were identified and are discussed below: (1) lack of cancer knowledge and pursuit of treatment options, (2) stigma and discrimination, (3) long waiting periods for referral, diagnosis, treatment and care, (4) lack of screening and diagnostic equipment and (5) 26).

Stigma and Discrimination
The findings revealed a prevailing expectation that people with cancer would experience stigma and discrimination. Therefore, they delayed in reporting symptoms because of expectations, based on previous experience, that they would be stigmatized as having multiple partners and blamed for their illness (Participant: 53

Socio-economic barriers
Socio-economic barriers were also reported to play a major role in the healthcare  (26,29).
Their study showed that among the elderly, utilization increased with age and this was due to the increased prevalence of chronic illnesses like cancer (29). Among other barriers to the utilization of PC, participants reported how fear of being stigmatized and discriminated against often prevents many patients from seeking needed PC services for any cancer diagnosis. Participants further mentioned that many people with cancer have been evicted from their homes by their families and rejected by their friends and colleagues. Our findings are consistent with the results of earlier studies, which found that a particular group sampled in Botswana, 40% of cancer patients had experienced stigma (30). Further to the scientific evidence, it was reported that doctors from SSA are usually reluctant to pass bad news to patients on any health matter, especially regarding cancer as this puts patients at risk of being stigmatized and discriminated against (31).
In addition, this current study demonstrated that socio-economic barriers to PC utilization were reported most often by the participants. Our findings were similar to other studies done across the sub-Saharan countries which found that socioeconomic inequalities contribute to and shape people's health seeking behaviour adversely (32). This negative effect of people's socio-economic realities on their health behaviour is not different in PC. Screening participation is higher amongst people with advanced socio-economic status. Such socio-economic status includes material resources (e.g. income), social status (e.g. occupation) and level of education (33). For example, recent US data shows a strong association between self-reported mammography participation and both education and income (34).
Similar associations with income and education were found in Great Britain (34).
Furthermore, lack of cancer knowledge and pursuit of treatment options was most pronounced among those living in the least affluent areas and those who had primary education as their highest level of education. The positive relationship established implies that as one moves from a low level of education to a higher level of education, there is a significant increase in utilization of PC progressively.
These results concur with the study findings of Hirai et al., who found that, despite the inequities present, the number of years of education did improve the prevalence rates of seeking healthcare, even among the poor (35). This is also because as people become more educated they are able to identify their healthcare needs and seek them (36). Education levels have been shown to often influence the occupation of an individual and this will directly translate to income earned and ability to afford the healthcare services (37

Availability of data and materials
Data from this study is the property of the Province of KwaZulu-Natal, Department of Health and University of KwaZulu-Natal and cannot be made publicly available. Patients' perspectives on how DoH can improve utilization of cancer palliative care (N=394)