Tackling inequalities in cancer care and outcomes: psychosocial mechanisms and targets for change

Abstract

Disparities in cancer incidence, outcomes, and prevalence persist globally among countries.1, 2 While incidence and mortality rates for most cancers are falling in many high-income Western countries, the opposite is true in numerous less-developed and economically transitioning regions.1 With regard to colorectal cancer, for example, there is a 10-fold variation in incidence rates worldwide, with high and rapidly increasing incidence rates in Eastern European countries and Japan, the lowest rates in Africa, Central and South America, and South Central Asia, and stabilizing or declining rates in the USA, New Zealand, and Canada.1 Five-year relative survival rates for colorectal cancer range from approximately 20% in Algeria compared with nearer 60% in the USA.2 Cancer incidence and mortality also vary within and across countries by sociodemographic factors such as gender, age, ethnicity, and socioeconomic status. For example, incidence rates of stomach cancer vary by gender, although much more so in some countries than others. Incidence rates per 100 000 are 65.8 in men compared with 24.2 in women in Japan, for instance, but just 3.3 in men relative to 2.0 in women in Egypt.1 Across countries, individuals with higher socioeconomic status have been found to have better cancer survival rates than those with lower socioeconomic status.3-5 In Sweden, for instance, individuals with university-level education have been found to have survival rates 40% higher than those of people with less than 9 years of school education.3 Disparities in cancer incidence and mortality are underpinned by geographical and sociodemographic inequalities at numerous points across the cancer continuum: in risk factors for cancer (e.g., smoking and Helicobacter pylori infection), in cancer screening participation, in cancer symptom awareness and timely diagnosis, in cancer treatment including psychosocial care, and in long-term morbidities among survivors (e.g., depressive symptoms and pain).1, 6-10 Some examples of these inequalities include the following: Research in Britain, stemming from the National Awareness and Early Diagnosis Initiative in England launched in 2008, has consistently found lower awareness of cancer warning signs and symptoms among minority ethnic groups and lower socioeconomic status groups.7 Some research studies have found age disparities in stage of cancer at diagnosis, although the nature of these disparities varies by cancer type. For some cancers (e.g., cervical), increasing age has been found to be associated with more advanced stage at diagnosis, whereas for other cancers (e.g., lung), the reverse has been found, with older patients less likely to be diagnosed at later stages.8, 11 While it is increasingly recognized that psychosocial care is a key component of high-quality, comprehensive cancer care, provision varies markedly geographically. Psychosocial care is generally more established and implemented in higher-income countries relative to less developed countries and is more likely to be provided in cancer centers and university hospitals, which are typically situated in urban areas than in smaller facilities based in the community.9, 12 Reducing disparities in cancer care and outcomes is an ongoing priority in many countries as well as internationally. For example, current cancer strategy in England (2015-2020) aims for “a reduction in the variability of access to optimal diagnosis and treatment and the resulting inequalities in outcomes” (www.cancerresearchuk.org/about-us/cancer-strategy-in-england), and achieving health equity and eliminating disparities is an overarching goal of the Healthy People 2020 initiative in the USA, which sets out a comprehensive agenda for improving the health of all Americans in numerous areas including cancer (www.cdc.gov/nchs/healthy_people/hp2020.htm). The Union for International Cancer Control's 2013 World Cancer Declaration calls upon government leaders and health policy-makers to reduce the global cancer burden and promote greater equity (www.uicc.org/world-cancer-declaration). This special issue of Psycho-Oncology focusing on cancer health disparities was called for and coordinated by members of the Executive Committee of the British Psychosocial Oncology Society in response to the fact that cancer disparities persist as a global problem and priority, and to highlight the contribution that psychosocial oncology research can make in helping to tackle these. Over 50 papers were submitted to the call for this issue (the most for any Psycho-Oncology special issue to date), and 148 reviewers based in 14 different countries have been involved in peer-reviewing the submissions. Sixteen of the submissions are included in this special issue: 3 literature reviews,13-15 9 research papers,16-24 and 3 clinical correspondence pieces,25-27 as well as a report from the International Federation of Psycho-Oncology Societies which documents and highlights the heterogeneity of the federated societies and the global between-country differences in the development and implementation of psycho-oncology care.28 Collectively, the papers in this special issue use a wide range of methodologies and analytic techniques with a diverse range of samples, to look at disparities in cancer care and outcomes at several points on the cancer continuum: from screening,16-18 to symptom attribution,19, 21 to diagnosis, treatment and care,20, 21, 25, 26 and into survivorship,22-24, 27 in relation to a number of different factors: chiefly education,16, 19, 22 ethnicity,14, 15, 18, 22, 24, 27 and cultural factors (such as religious beliefs17 and English language proficiency25, 26) and the presence of comorbid health conditions13, 20, 22 but also residential location,15 age,22, 24 and relationship status.23 For example, Ruiz and colleagues24 examine differences in smoking, alcohol, and drug use on the basis of age and ethnicity among adolescent and young adult survivors of childhood cancer, who remain at risk for poor long-term health outcomes. Lower educational level, and more broadly lower socioeconomic status, is associated with lower uptake of cancer screening,6, 29 increased likelihood of being diagnosed with certain types of cancer,5, 30 a more advanced cancer stage at diagnosis,4, 8 and poorer morbidity and mortality outcomes.3-5, 10, 31 Three of the papers in this special issue examine education-based cancer disparities.16, 19, 22 Importantly, these papers elucidate psychosocial and behavioral factors and processes that, to some extent, mediate the link between low educational attainment and poorer preventive care and outcomes. Crucially, as these mediators are potentially modifiable, they offer a target for interventions to reduce cancer inequalities. The papers also usefully identify psychosocial factors which do not appear to underpin education-based disparities and are thus unpromising targets for intervention. For example, in their paper concerning inequalities in colorectal cancer screening participation in the UK, Smith et al.16 demonstrate a consistent and graded effect of education on perceived emotional and practical barriers to uptake. In particular, beliefs that one would worry more about colorectal cancer after screening and concerns about tempting fate were strongly negatively associated with education. Interestingly, no educational gradient was observed for perceived benefits of screening. Marcu and colleagues' paper19 focusing on cancer symptom attribution shows that a lower level of education is associated with a lower likelihood of considering cancer a possible cause for breast cancer symptoms. Moreover, they found that this is the case not only for a less well-known symptom (nipple rash) but also for a very familiar symptom (axillary lump) and that this association persisted even after adjustment for cancer avoidance. There is some evidence to show that, in comparison with majority ethnic groups, some minority ethnic groups are less likely to participate in cancer screening.32, 33 Also, some ethnic groups have a higher risk of being diagnosed with some types of cancer and higher mortality rates for some cancer types. For example, research in many countries consistently finds an elevated incidence of prostate cancer among men of African descent,34, 35 and there is strong evidence of a disproportionately high breast cancer mortality rate in African American women, even after accounting for socioeconomic status.36 Nine of the papers in this issue seek to enhance our understanding of ethnic and cultural cancer disparities.14, 15, 17, 18, 22, 24-27 For instance, Seymour-Smith and colleagues18 elucidate reasons for the low uptake of digital rectal examination screening for prostate cancer among UK African-Caribbean men. Interviews exploring the men's perceptions of digital rectal examination reveal salient fears of the examination stemming from cultural homophobic beliefs and in turn a fear of stigma associated with transgressing cultural norms. In a similar study, but focused on breast cancer screening, Padela et al.17 examine how Islamic religious beliefs inform American Muslim women's intentions to undergo mammography. This study highlights the influence of religious beliefs on screening intentions, but moreover, details the complexity of this influence; although some religion-related beliefs may be barriers to screening (e.g., modesty preservation necessitating gender concordant health care), others may in fact enhance mammography intention (e.g., notions about duty to God and stewardship of one's body). Both Seymour-Smith and Padela's papers provide insight into, and inform understanding of, belief systems influencing cancer screening participation among groups with low uptake. Such knowledge is crucial for developing culturally sensitive initiatives to reduce ethnic and cultural disparities. Perez and collegaues26 report one of the first studies to interview US-based medical interpreters to obtain their perceptions of potential mechanisms underlying disparities in the cancer care of patients with limited English proficiency. The study highlights underuse of interpreters as a principal factor compromising patient care due to factors such as privacy concerns and families' desires to interpret or withhold information, as well as lack of understanding of the US medical system among patients with limited English proficiency, which linguistic interpreter services alone cannot address. With a growing aging population in many countries, increasing numbers of people living with and beyond cancer will also have a diagnosis of 1 or more other chronic physical, mental, and/or cognitive health conditions. There is evidence to indicate that patients with cancer with comorbidities are more likely to be diagnosed at a later stage (depending on the type of cancer and the type and severity of comorbidity),37 are less likely to be offered and to receive curative treatment,37, 38 and have poorer survival.5, 37, 38 However, evidence in this area is lacking due to, among other things, under and unsystematic recording of comorbidity information in medical records and cancer registries and the fact that patients with comorbidities are often excluded from randomized controlled clinical trials.37 Three papers in this special issue examine the experiences of patients with cancer with comorbidities.13, 20, 22 These include our Editor's Choice paper by Hopkinson and colleagues,13 which reports a novel and topical systematic review on what is currently known about the experience of cancer treatment and its outcomes in adults with dementia. Hopkinson et al. find that, compared with other patients with cancer, those with dementia are diagnosed at a later stage, receive less treatment, are more likely to experience treatment complications, and have poorer survival and that the reasons for these disparities remain to be determined. In their paper reporting on a national survey of US-based cancer survivors, Stein et al.22 show that among survivors experiencing ongoing pain, those with physical and/or mental health comorbidities had experienced more barriers to pain management at patient, provider, and system levels. Compared with survivors with no comorbidities, those with other health conditions had greater odds of reporting, for example, a belief that their pain could not be relieved, concerns about addiction and side effects, that their physician did not ask them about their pain, and that their health insurance would not cover pain treatment. Two of the clinical correspondence papers in this special issue report pilot study results of interventions to ameliorate cancer disparities. Lipson-Smith et al.25 examine a multi-faceted communication intervention to increase non-English speaking migrant patients' involvement in, and understanding and recall of, clinical consultations. They found the intervention proved culturally acceptable, although only the audio-recording of consultations component was perceived to be effective. Two other aspects of the intervention (information sheets and a question prompt list) were unused by many participants, suggesting that in future work the intervention should be simplified. Bava et al.27 appraise a parent-directed psychoeducational intervention to improve scholastic attainment among survivors of childhood leukemia; the intervention is part of the clinical services routinely provided at their US-based institution, which treats a high proportion of Latino, lower socioeconomic status patients. They report that the intervention is valued by families and shows preliminary feasibility and efficacy. Although just 2 of the papers in this special issue report intervention research, the findings of all the papers have, to varying degrees, implications for intervention development. Considering these implications collectively, 3 key themes emerge: That interventions to reduce cancer disparities (a) need to be targeted and socioculturally tailored to groups experiencing disparities; (b) might, amid other things, fruitfully look to boost social support and its salubrious effects among these groups; and (c) need also to focus on enhancing relevant aspects of health professionals' knowledge and skills. Several papers in this issue highlight that reducing the inequalities gap requires interventions that are targeted and socioculturally tailored to particular groups most in need. For example, Seymour-Smith et al.18 assert that initiatives to increase prostate cancer digital rectal examination screening among African-Caribbean men need to address the fear of homophobia in this population, and Padela et al.17 recommend that interventions to improve mammography rates among Muslim women incorporate religiously tailored messages. Interventions which are not carefully designed to address the specific psychosocial factors and processes contributing to the disadvantaged position of a particular group are unlikely to succeed in narrowing the inequalities gap. Smith et al.'s paper,16 for instance, found that while some perceived barriers to colorectal cancer screening were strongly graded by education level, perceptions of the benefits of screening were not. Therefore, initiatives focused on increasing the perceived benefits of screening may improve uptake overall but are unlikely to reduce the socioeconomic gap in participation rates. It is possible that ill-designed interventions could even exacerbate inequalities, if they were to improve health outcomes among already advantaged groups, but prove ineffective among those groups experiencing disparities.39 A large body of literature attests to the importance of social support to health and wellbeing,40 and some of the papers in this issue underscore the importance of social support to good outcomes across the cancer continuum. For example, Leung et al.23 show that breast cancer survivors without partners have poorer health-related quality of life than those with partners, and, moreover, demonstrate that this disparity is mediated by lower levels of social support among women without partners. Interventions to reduce inequalities might look to provide or enhance social support among groups experiencing disparities; this may be particularly important in communities or countries where stigma surrounds cancer and its treatment. In their meta-synthesis review examining minority ethnic patients' experiences of prostate cancer, Rivas et al.14 found that the literature indicates that there is a “silencing stigma” which discourages men from disclosing their diagnosis and thus receiving social support, yet also shows that men who do disclose report the value to wellbeing of social support from close networks. Indeed, supportive social networks may help buffer against stigma. In their paper reporting on interviews with breast cancer survivors in Uganda, Meacham et al.21 show that stigma impedes women from seeking and completing medical treatment, but more importantly, they also show that women who do engage with, and remain in cancer care, create “pockets” of social support which are a key enabling factor. A number of the papers in this issue call for interventions to increase health professionals' awareness and understanding of relevant psychosocial factors and processes contributing to cancer disparities, so that they may provide care and services which help to mitigate these disparities. For example, Marcu et al.19 suggest that knowledge of sociodemographic variation in cancer symptom attribution would allow general practitioners to have a greater sense of patients who may normalize their symptoms and thus be at risk of late diagnosis, and Padela et al.17 recommend programs that sensitize health care providers to the importance of modesty and gender concordance for Muslim patients. Also, several papers recommend initiatives to enhance various different aspects of health professionals' communication skills. For example, Flynn et al.20 interviewed people with intellectual disabilities and found that they are often excluded in cancer consultations from conversations and decisions about their care and treatment, and thus call for interventions to enhance health professionals' patient-centered skills to enable optimal involvement of patients with intellectual disabilities in their own care. As Flynn et al. note, theoretically, such skills are generalizable and thus may serve to enhance consultations with patients with other comprehension or communication difficulties (such as dementia or limited proficiency in the local language). The work in this special issue highlights several areas and priorities for future research. The review papers all identify a number of research gaps. For instance, Youl and colleagues15 examined variations in psychosocial outcomes by residential location and indigenous status in women with breast cancer in Australia and found no research literature examining inequalities by indigenous status, only a few studies which directly compared outcomes according to residential location and a lack of studies focusing on women with advanced breast cancer. Hopkinson and colleagues' review13 highlights the dearth of research investigating the supportive care needs and preferences of patients with comorbid cancer and dementia; yet, such research is needed to inform guidelines for the care and management of these patients. Despite the vital role of family carers in supporting patients and the adverse impact caregiving can have on carers' own health and wellbeing, Hopkinson et al.'s review also highlights the absence of research considering carers and their support needs. Almost all of the papers in this special issue report research conducted in high-income, developed countries. There is a need for more research examining inequalities in less developed and economically transitioning countries, particularly as the burden of cancer shifts toward these countries, where incidence and mortality rates are increasing.1 With an aging population globally, in future years, many countries will see an increasing prevalence of comorbidity and multi-morbidity among people with cancer. More multi-morbidity research is needed to understand the influence of other co-occurring health conditions on cancer symptom awareness, help-seeking and diagnosis, and to inform guidelines for the management and long-term follow-up of patients with cancer with other chronic conditions. This will mean collaborative working outside the single-disease framework which is currently typical of most specialist health care services, research, and education. The papers in this issue examine disparities based on numerous different factors (e.g., age, socioeconomic status, and ethnicity); yet, many of these have overlap. Future research should focus on the needs and outcomes of those patients at the intersection of multiple categories of disadvantage who are likely to experience the greatest inequalities. A priority for future research is to develop, test, and implement interventions to tackle cancer inequalities. Interventions may have a greater chance of success if they are not only targeted and socioculturally tailored to particular groups experiencing disparities but also developed in collaboration with them. In their paper on mammography participation among Muslim women, Padela and collegaues17 recommend that developing successful interventions to improve uptake requires partnering with Islamic scholars and Imams in a two-way dialogue, and Seymour-Smith and colleagues18 have used the findings from their interview study to produce a mobile application (PROCEE, which provides prostate cancer information and evaluates risk) which was designed with key members of the African-Caribbean community. To summarize, the excess mortality attributable to socioeconomic and other inequalities is considerable and reducing disparities in cancer care and outcomes between and within countries should remain a high priority on political and research agendas. The research in this special issue shows the contribution that psychosocial oncology research can make towards understanding and tackling disparities across the cancer continuum. Psychosocial research can contribute by delineating theoretically modifiable psychological, social, and behavioral factors and processes underpinning cancer disparities and developing interventions to mitigate or change these and reduce disparities. However, much remains to be done to realize the potential of psychosocial research, and future work needs to focus on designing and implementing successful interventions which can effect change and reduce cancer inequalities. We would like to thank Maggie Watson and Jimmie Holland for their support and guidance, and the editorial team at Wiley for their input and assistance.