Abstract
Background: Sweden’s national cancer strategy points out several areas of cancer care that need improvements. Among them the need for supportive care resources to be accessible through the entire cancer trajectory and the reduction of socioeconomic inequalities. The aim of this study was to compare the patient-reported access to supportive care in the Stockholm-Gotland region between patients diagnosed in 2014 and 2016. The aim was also to describe how socioeconomic and other demographic factors impact access to supportive care.
Material and methods: All patients with gynaecological, head and neck, haematological and upper gastrointestinal cancers diagnosed in the Stockholm-Gotland regions were identified through the Swedish Cancer Registries. Data were collected via a questionnaire on demographic, socioeconomic factors and patients’ perception (n = 1872) of access to supportive care. Data were summarized using descriptive statistics and logistic regression was used for relevant variables.
Results: Access to some supportive care resources, such as contact nurses (CNs) and individual written care plans (IWCPs) had significantly improved from 2014 to 2016. The proportion of patients that had received information about patient advocacy groups (PAGs) had also improved but remained on a relatively low level (29 and 35%, respectively). The proportion of patients being refereed to palliative care (PC) did not change between 2014 and 2016. In total, 10% of the patients reported to having received information on second medical opinion (SMO). Patients that had undergone multimodality cancer treatment were more likely to report access to supportive care, and those with lower education levels were more likely to have access to CNs and IWCPs.
Conclusion: Access to some of the supportive care resources have shown improvements in the Stockholm-Gotland region but further efforts are required, especially regarding access to PC, information about PAGs and SMOs.
Introduction
Sweden’s first national cancer strategy was introduced in 2009 [Citation1]. The strategy describes several areas in need of improvements, such as survival rates, socioeconomic inequalities and waiting times, availability and patient-experiences in general. It also suggests several supportive care strategies to help improve the experiences related to cancer. To achieve the goals of the strategy, six Regional Cancer Centres (RCCs) were established in Sweden during 2011. These centres are assigned to provide support and knowledge to the health care organizations responsible for cancer care, and to follow-up the results and outcomes.
The term Supportive care has been defined as ‘The prevention and management of the adverse effects of cancer and its treatment’ [Citation2]. This includes the management of both symptoms and side effects across the cancer pathway from diagnosis, through treatment, follow-up or end-of-life care. WHO have defined palliative care (PC) as aiming to increase both quality of life and death, alleviate symptoms and to include the patient’s family during the whole care trajectory [Citation3].
In a recent publication [Citation4] we evaluated the access to some of these supportive care strategies among patients with gynaecological, head & neck (H&N), haematological and upper gastrointestinal (GI) cancer diagnosed in 2014 in the Stockholm-Gotland region. The results showed that the supportive care strategies described in the Swedish National Cancer Strategy were only available for some of the patients and therefore making the impact difficult to evaluate. The results from this previous paper were used by the RCC Stockholm-Gotland team together with the clinical cancer care providers, to tailor and implement interventions aiming at improving the supportive care for cancer patients in the region. The interventions focussed mainly on increasing the access to contact nurses (CNs) and to use individual written care plans (IWPCPs). CNs are the main point of contact, a resource for education, information, support and coordination of the clinical pathway for patients and their families. IWCPs include contact information, appointments, individualized information regarding treatment/care, rehabilitation and general information on patients’ rights, resources etc. Actions were also taken to improve the collaboration between acute and PC teams and thereby access to PC. Further efforts were made to improve the information for patients regarding existing patient advocacy groups (PAGs).
Socioeconomic inequalities in both cancer survival and other outcomes have been reported frequently during the last decades, even in countries whose cancer care system provide care without cost at the point of delivery [Citation4–8]. A recent Swedish study [Citation9] showed that lower education, lower income and having single status at the time of pancreatic cancer diagnosis was associated with increased mortality. Another Swedish study based on nation-wide register data [Citation10] showed older adults ‘with higher education were more likely to die on hospitals, compared with those with lower education. It is assumed that, well-organized supportive cancer care could possibly reduce the differences related to socioeconomic factors, but this area requires further studies [Citation11].
In this study, we follow-up the results from the previous first study on cancer patients’ experiences of supportive care strategies among cancer patients in the Stockholm-Gotland region. We also evaluate the impact of socioeconomic factors on access to supportive care.
Aim
The aim of this study was 2-fold. The primary aim was to compare patient-reported access to supportive care in the Stockholm-Gotland region between patients diagnosed in 2014 and those diagnosed in 2016. The secondary aim was to describe how socioeconomic and other demographic factors impacted access to supportive care.
Material and methods
Design
This cross-sectional study is part of a larger project with the aim to evaluate a newly introduced regional cancer nursing role in Sweden, Coordinating CN. In this study, we analyzed and compared data from this project, on patients’ perception of availability to supportive care strategies between patients diagnosed 2014 and 2016 with gynaecological, H&N, haematological and upper GI cancers. These groups of often vulnerable patients, undergoing complex cancer treatment are often underrepresented in health care research. We also evaluated the impact on demographic and socioeconomic factors on supportive care resources. The study was approved by the Regional Ethical Review Board, Stockholm (Dnr 2016/5:4).
Participants
The patients were identified through The Swedish Cancer Registry and the cancer site specific quality registries. This Registry includes 99% of all cancer cases that are clinically reported and morphologically verified [Citation12]. The Swedish Cancer Registry was founded in 1958 and covers the whole population. It is compulsory for every health care provider to report newly detected cancer cases to the registry. The diagnostic specific registers contain detailed disease-related data and other quality parameters. The report rate may vary between the registers and over time but are generally over 90%. Data found in the above described registers were cross-referenced with the Swedish National Population Registry to identify and select the participants. This cross-sectional sampling method also minimized the risks of sending questionnaires to deceased patients.
Inclusion criteria
All patients diagnosed with a new primary gynaecological, haematological, H&N or upper GI cancer, diagnosed in 2014 and in 2016, living within the Stockholm Gotland area and ≥ age of 18 were invited to participate in the study.
Data collection
Data were collected via a study-specific questionnaire, designed by the research team and distributed via regular mail, in April–May 2015 and 2017, respectively, ensuring that the participants would be in the follow-up phase, post initial cancer treatment.
A cover letter explaining the purpose of the study, patient confidentiality was sent out (with a pre-stamped envelope) with the return of a completed questionnaire considered as informed consent. One written reminder was sent out three weeks following the initial postal date. The questionnaires could be completed in paper form or electronically via a secure internet link.
Study-specific questionnaire
The questionnaire (57 items), designed to evaluate the new CNN role, based on the role description, in the larger study included questions in the following areas; demographics (age, educational and occupational status, living situation, place of birth and type of treatment, 8 items), investigation and diagnosis (11 items), information exchange and patient participation (20 items), PC and care transitions (18 items). Education level was categorized as low (primary school, secondary school and college) or high (university or university college). Occupational status was categorized as working, retired or other (sick-leave, student or parental leave). The following categorization were used for place of birth (Sweden or abroad) and type of cancer treatment (single- or multi-modality treatment).
The following Likert scales were used for response options; 5-, 4- and 3-point scales. In this study, we analyzed and compared demographic data plus six other items concerning access to CN (two items ‘Did you have contact with a CN during the work-up process?’, and ‘Did you have contact with a CN during the cancer treatment?’), IWCP (one item ‘Have you had access to an IWPCP?’), information from the cancer care team on second medical opinion (SMO) (one item ‘Have you received information from your cancer care team on the possibility to ask for a so called SMO?’), PAG (one item ‘Have you received information about support from a PAG?’) and referral to PC (one item ‘Did you get refereed to PC during or after your cancer treatment?’) for patients diagnosed in 2014 and 2016, respectively.
Statistical methods
Patient characteristics are presented as descriptive data with frequencies and percentages, and the age variable is also presented with a mean and standard deviation, together with a minimum and maximum value.
Logistic regression was used in the analysis of the variables CN, IWCP, PAG and PC. The four-point scale was dichotomized into a two-point scale (very/a lot and a little/not at all). Both univariate and multivariate analyses were performed using clinical and demographic/socioeconomic variables.
Results from the logistic regression models are presented as ratios of the odds together with 95% CIs. All reported p-values from the regression models refer to Wald-tests, with a significance level of .05.
Results
Patient characteristics
Out of 2941 eligible patients, the results are based on data received from 1872 (response rate 64%) cancer patients, 46% (n = 869) of which were diagnosed in 2014 and 54% (n = 1003) in 2016. The majority of the responders were women (n = 1098, 59%). The mean age was 67 years and gynaecological and haematological cancers were the most frequently common cancer sites. Most of the patients were born in Sweden (81%, n = 1374), retired (69%, n = 1163) and were cohabiting (67%, n = 1125). More than half of the patients (59%, n = 976) reported that they had undergone single modality cancer treatment. Patient characteristics are presented in .
Table 1. Patient characteristics for responding patients.
We found no statistical difference between responders and non-responders regarding gender. However, the mean age was higher among the responders (67 years) compared with the non-responders (64 years), diff [95% CI] –2.93 [–4.04, –1.81], p < .0001.
Access to supportive care
We found that patient-reported access to both CNs and IWCPs had significantly (p < .0001) improved from 2014 to 2016 (). The proportion of patients that had received information about PAGs had also improved significantly from 29% (n = 218) for patients diagnosed in 2014, to 35% (n = 301) for patients diagnosed in 2016 (p = .0166). In the total sample, only 10% (n = 161) of the patients reported to receive information on SMO; however, the proportion of patients receiving information on SMO improved significantly between 2014 and 2016 (p = .0355). We found no statistically significant difference between 2014 and 2016 regarding referrals to PC ().
Table 2. Comparison of patient-reported access to supportive care between patients diagnosed 2014 and 2016.
Factors impacting access to CNs
In the total sample, we found no statistically significant association between patient-reported access to a CN with regards to age, gender, place of birth (Sweden/abroad), occupational status or living situation (cohabitant/living alone). However, patients with lower education levels were more likely to have access to a CN, compared with patients with higher education levels (p = .0377). Patients that had undergone multimodality cancer treatment were also more likely to have access to a CN compared with patients that had undergone single modality cancer treatment (76 versus 24%, p < .0001) (.
Table 3. Association between access to CN and patient characteristics for the total sample.
Factors impacting access to IWCPs
We found no statistically significant differences in the total sample between patients reporting to have or not having access to an IWCP regarding age, diagnosis, occupational status or living situation. However, men were more likely to have a IWCP (p = .0387), as were patients born outside of Sweden (p = .0016), patients with lower education levels (p = .0223) and patients that had undergone multimodality cancer treatment (p < .0001) (.
Table 4. Association between access to IWCPs and patient characteristics for the total sample.
Factors impacting receiving information concerning PAGs
Just over one third of the total sample (35%, n = 519), reported having received information concerning PAGs from the cancer care team. We found no significant difference between patients reported to receive/not receive this information regarding gender, place of birth, occupational status or education level. Among the four groups, patients with upper GI cancer were least likely (21%, n = 78) to report having this information. Patients that received information about PAGs were younger (p < .0001) and more likely to be cohabitant (p = .0144) and to have undergone multimodality treatment (p < .0001) (.
Table 5. Association between information concerning PAGs and patient characteristics for the total sample.
Factors impacting patients being referred to PC
Only 20% (319 of 1621) of the total sample reported being referred to PC at any time during their cancer trajectory. We found no statistically significant association between referral to PC and age, gender, place of birth, education level or living situation. Patients with gynaecological cancer were less likely to be referred to PC compared with the other patient groups (p < .0001). Working patients were less likely to be referred to PC compared with retired patients (p = .021) and patients that had undergone multimodality treatment were more likely to be referred compared with patients that had undergone single modality treatment (p < .0001) (.
Table 6. Association between reported referral to PC and patient characteristics for the total sample.
Discussion
This large study, using population-based data with a relatively high response rate compares patient-reported access to supportive care among certain groups of cancer patients diagnosed in 2014 and 2016, respectively. We found improvements in access between 2014 and 2016 for most of the supportive care resources studied in this paper, indicating positive results from the interventions that have been implemented during the last few years within the framework of the national cancer strategy. These efforts need further development, however, to reach all cancer patients in need of supportive care.
The proportion of patients being referred to PC remains on a relatively low level (both in 2014 and in 2016), which is a worrying result, given that these groups of patients often have severe symptoms after their cancer treatment and PC is well developed in the region. Several new PC services have opened the last few years. We can only speculate on the reasons. It may reflect a common notion among many health care providers and patients that PC is equal with end of life care, underestimating this important resource of supportive care through the entire cancer trajectory [Citation13]. It may also reflect the fact that most hospitals have no in-house PC services, which might challenge collaboration and referral between acute and PC teams [Citation14]. Another explanation, might be the specific groups of cancer patients included in our study. A recent study by de Oliveira Valentino et al. [Citation15] shows that both patients with upper GI cancer and haematologic malignancies were referred to PC at a later stage, compared to other cancer patients. In our study, patients with gynaecological cancer were less likely to be referred to PC, probably reflecting the relatively large proportion of patients in this group undergoing single modality treatment, associated with less complex cancer care trajectories and better prognosis compared with the other populations. This is supported by the fact that significantly more patients in the total sample had undergone multimodality cancer treatment were referred to PC. However, evidence suggests that patients with gynaecological cancer may be experiencing more distress and therefor needing more support compared with some other groups of cancer patients [Citation16]. PC could be an important resource to meet this need.
Information on SMOs also remains on a very low level. It is of great concern that only 10% of the total number of patients received this information from their cancer care team. This is particularly important as it is principally a legal right for patients with life-threatening disease in Sweden [Citation17]. In a recent German study [Citation18] among patients with breast cancer, younger patients (<75-years old) and patients with higher education levels were more likely to be informed about SMO by their physician. In the same study, the authors found that patients that gained this information; also were more inclined to seek an SMO, indicating the importance that this information is being provided by the cancer care team. Since so few patients in our sample reported receiving information on SMO, analysis of impact of demographic and socioeconomic factors were not relevant.
Even if most patients diagnosed in 2016 seem to have had access to a CN for support, the results show differences between the patient groups, indicating that the CN role is not fully implemented in all cancer care settings. However, patients that may be particularly vulnerable and therefore in need of supportive care, due to higher symptom burden (undergone multi-modality cancer treatment) or socioeconomic factors (patients with lower education levels and patients born abroad) are also more likely to gain access to CNs.
A lot of effort has been put into implementing IWCP in the Stockholm-Gotland region during the last few years. Our results reflect this and show an increase in availability for patients diagnosed in 2016, but as in the case of CN, more effort is needed to implement IWCPs for all patients. We also found that males, patients that had undergone multimodality treatment (compared with single modality treatment), patients born outside of Sweden (compared with patients born in Sweden) and patients with lower education levels were more likely to have access to the IWCPs. We assume that these results also reflect a selection from the cancer care providers, offering support (IWCPs) to patients that they estimate might need it the most. The fact that male patients more often reported having an access to IWCPs are more difficult to explain and a concern. Previous research from our region [Citation3] and elsewhere [Citation19] indicates clear benefits regarding information provision and patient satisfaction with different types of IWCPs.
Our results show that both demographic and socioeconomic factors impact access to supportive care for patients undergoing cancer treatment in the Stockholm-Gotland region. Patients undergoing multimodality cancer treatment are more likely to have access to supportive care, which is not surprising considering the often, complex symptom burden associated. However, patients with single-modality treatment may also need supportive care. Priorities should be based on individual needs, rather that type of cancer treatment.
Patients with higher education levels didn’t report greater access to any of the supportive care resources included in the study. This is reassuring and in contrast to the results by others in pancreatic cancer [Citation9] and among patients with gynaecological cancer [Citation16]. It should be emphasized, however, that educational level is not always the same as income level. Other researchers have found that levels of income can result in inequalities in health care outcomes in Sweden [Citation20]. Low income among patients with gynaecological cancer for instance, has been described as a risk factor for psychosocial distress [Citation16]. If we had collected data on family income, our results might have been different. In Sweden, some groups of blue-collar workers have higher incomes than defined academics. In contrast, patients with lower levels of education more often reported access to both CNs and IWCPs. This might be partly contributed by increased awareness among cancer care providers, from RCC Stockholm-Gotland’s strategic work to reduce the socioeconomic inequalities related to cancer prevention, treatment and care. This strategy included health education campaigns, targeting socioeconomically disadvantaged communities in the region but also a tailored education programme on cancer specific inequalities for different groups of cancer care staff.
The living circumstances (cohabitant or living alone) didn’t seem to impact the access to CNs, IWCPs or PC referral. However, patients living alone were less likely to report having received information about PAGs than patients living together with someone. Patients that reported having received information about PAGs were also younger. These results are difficult to explain, as one can assume that older patients and patients living alone may need this support even more than patients with a stronger social and/or professional network. Patients with upper GI cancer were less likely to receive this information, which might be explained by the fact that the regional PAG for patients with pancreatic cancer were initiated only in recent years, while the other PAGs have been established much longer.
The issues around selection bias are always of importance to consider in studies based on questionnaire data. In this study, we only included four groups of cancer patients and only patients from one health care region, the Stockholm-Gotland region. The results might have been different if other groups or patients from other regions had been included. Also, the non-responders were younger than the responders, which may have impacted the results. However, we found few differences between the four patient groups, indicating that results can be generally interpreted. We also exclusively chose to collect data directly from patients rather than from the cancer care providers, to reflect the aim of the study. If we had collected data from the cancer care providers, we might have found differences between patient groups and different cancer care providers. However, access to supportive care services should be studied from the patients’ perspective rather than from the health care perspective.
Conclusions
Patient-reported access to supportive care services has increased in the Stockholm-Gotland region, but needs to improve even more to meet the goals in the first national cancer strategy in Sweden. Socioeconomic and demographic factors impact access to supportive care services. The cancer care providers seem to prioritize access to supportive care to patients that have complex care needs, such as patients undergoing multimodality cancer treatment. However, more effort is required to increase access and individualize the supportive care resources. Information regarding PAGs and SMOs particularly need to be improved as does the number of patients being offered a refereed to PC. More research is required, including long term follow-up and also to involving other groups of cancer patients.
Disclosure statement
No potential conflict of interest was reported by the authors.
References
- Statens offentliga utredningar (SOU). En nationell cancerstrategi för framtiden 2009:11 [A National Strategy for the Future 2009;11] Stockholm: Social departement et; 2009 [cited 2018 Apr 30]. Available from: http://www.regeringen.se/rattsdokument/statens-offentliga-utredn-ingar/2009/02/sou-200911/
- Strategic Plan 2017: Multinational Association of Supportive Care in Cancer (MASCC). Available from: http://www.mascc.org/index.php?option=com_content&view=article&id=493:mascc-strategic-plan&catid=30:navigation
- World Health Organisation (WHO). Definition of Palliative Care, 2005; Available from: http://www.who.int/cancer/palliative/definition/en/
- Westman B, Kirkpatrick L, Ebrahim F, et al. Patient-reported experiences on supportive care strategies following the introduction of the first Swedish national cancer strategy and in accordance with the new patient act. Acta Oncol. 2018;57:382–392.
- Rachet B, Ellis L, Maringe C, et al. Socioeconomic inequalities in cancer survival in England after the NHS cancer plan. Br J Cancer. 2010;103:446–453.
- Woods LM, Rachet B, Coleman MP. Origins of socio-economic inequalities in cancer survival: a review. Ann Oncol. 2006; 17:5–19.
- Lundqvist A, Andersson E, Ahlberg I, et al. Socioeconomic inequalities in breast cancer incidence and mortality in Europe—a systematic review and meta-analysis. Eur J Public Health. 2016;26:804–813.
- Ramsay SE, Morris RW, Whincup PH, et al. Time trends in socioeconomic inequalities in cancer mortality: results from a 35 year prospective study in British men . BMC Cancer. 2014;14:474.
- Talbäck M, Ljung R, Feychting M, et al. Socioeconomic inequalities in mortality from pancreatic cancer – A Swedish national cohort study. Eur J Public Health. 2017;27(1): November.
- Kelfve S, Wastesson J, Fors S, et al. Is the level of education associated with transitions between care settings in older adults near the end of life? A nationwide retrospective cohort study. Palliat Med. 2018;32:366–375.
- Wagner EH, Ludman EJ, Aiello Bowles EJ, et al. Nurse navigators in early cancer care: a randomized controlled Trial. Jco. 2014;32:12–18.
- Barlow L, Westergren K, Holmberg L, et al. The completeness of the Swedish Cancer Register: a sample survey for year 1998. Acta Oncol. 2009;48:27–33.
- Haun MW, Estel S, Rücker G, et al. Early palliative care for adults with advanced cancer. Cochrane Database Syst Rev. 2017;6:CD011129.
- Ullgren H, Kirkpatrick L, Kilpeläinen S, et al. Working in silos? Head & Neck cancer patients during and after treatment with or without early palliative care referral. Eur J Oncol Nurs. 2017;26:56–62.
- de Oliveira Valentino TC, et al. Factors associated with palliative care referral among patients with advanced cancers: a retrospective analysis of a large Brazilian cohort. Support Care Cancer 2018;Jan 5. [Epub ahead of print].
- Bergerot CD, Clark KL, Obenchain R, et al. Breast and gynecological cancer patients’ risk factors associated with biopsychosocial problem-related distress. Psychooncology 2017; Dec 11. [Epub ahead of print].
- Patient Act. 2014;821. Available from: http://www.riksdagen.se/sv/dokument-lagar/dokument/svensk-forfattningssamling/patientlag-2014821_sfs-2014-821.
- Grosz SE, Hillen MA, Pfaff H, et al. Second opinion in medical encounters - A study among breast cancer patients. Patient Educ Couns. 2017;100:1990–1995.
- Brennan ME, Gormally JF, Butow P, et al. Survivorship care plans in cancer: a systematic review of care plan outcomes. Br J Cancer. 2014;111:1899–1908.
- Östergren O. Growing gaps: The importance of income and family for educational inequalities in mortality among Swedish men and women 1990-2009. Scand J Public Health. 2015;43:563–570.