Reducing the unmet supportive care needs of people with colorectal cancer

Abstract

The diagnosis and treatment of cancer can have a major impact on many aspects of health and well-being, with patients experiencing a variety of supportive care needs. In recognition of these needs, cancer policy internationally now acknowledges the importance of supportive care alongside clinical care. In order to reduce unmet supportive care needs among people with cancer, these unmet needs must first be quantified and then potential interventions to reduce unmet needs must be developed and tested. This thesis presents a series of studies which addresses the following aims: 1) To quantify and measure the unmet supportive care needs of people with colorectal cancer; 2) Ascertain the utility of a particular approach of supportive care service delivery, specifically nurse-delivered telephone supportive care; 3) To evaluate in a pilot study the novel nurse-delivered telephone supportive care intervention - CONNECT. Chapter One first provides an epidemiological perspective of colorectal cancer. Within Australia and NSW it is the second most common cancer, and cases in both jurisdictions have doubled in the last two decades. Within NSW, future modelling predicts that the number of new colorectal cancer cases in 2016 will represent a 19.5% increase from what was reported in 2007, and by 2036 will represent a 60.3% increase. Therefore the burden of this disease within the community will remain high. Chapter One next summarises the main treatment options for colorectal cancer, that is surgery, radiotherapy and chemotherapy. Clinical overviews of these treatment modalities are provided with the rationale for their use described. Advances in clinical management have improved post-operative mortality and survival outcomes of people treated for colorectal cancer. Therefore these survival improvements, coupled with increases in cases of colorectal cancer, will result in more people living with the outcomes of treatment. It is expected that in the future, disability due to colorectal cancer and its treatment will become a more important contributor to the burden of disease than mortality. The remainder of Chapter One then describes the range of adverse outcomes of colorectal cancer treatment. These include reductions in quality of life and specific morbidities such as pain, issues with wound healing, bowel, urinary and sexual dysfunction. The negative psychological consequences of treatment are also explored. Chapter One concludes by acknowledging that cancer treatment and treatment outcomes will result in a unique set of health care needs. Chapter Two introduces that health care need as a foundation to guide cancer service development and delivery. Despite the growing interest in health care need, there is little discourse of what is actually meant by the term ‘need’ and the underlying theoretical or conceptual basis of its measurement. Therefore a critical overview is presented in Chapter Two of the range of definitions and theories that seek to conceptualise what is meant by the term ‘need’. Need interpretations are described from psychological, humanitarian, social policy, academic and health care perspectives. The concept of unmet need is also introduced whereby unmet need is viewed as resource deficit. The findings of this review confirm that ‘need’ and ‘unmet need’ are conceptually different and distinct. Measurement of need is argued to be limited in scope as it only identifies potential problems. Need identification does not provide information about whether need has been met or whether current service interventions are successful at addressing and satisfying need. In contrast, ‘unmet need’ is considered a more patient relevant concept. It is indicative of where patient care is compromised and therefore is more relevant to health service development, as the focus of attention is on areas were service provision is inadequate, inaccessible or not in existence. Following this, Chapter Two concludes with a definition of the types of potential unmet needs that people with cancer could experience, in particular unmet supportive care needs. Supportive care is defined as an umbrella term that embraces the full range of needs, physical and psychosocial, that emerge for a patient due to the impact of cancer. Following this discussion of need and unmet need, Chapter Three describes the specific aims of this thesis. Chapter Four presents the results of a systematic review of the international literature in which the prevalence of unmet supportive care needs of people is described. The initial search identified 22,781 potentially relevant articles. However only 94 were found to be relevant to the research question and only 57 studies quantified unmet supportive care need. These studies were classified by specific time of the cancer illness, for example diagnosis, treatment and post-treatment. Prevalence of need was highly variable in all domains (ranging from 1% to 93%), across all time points and varied within and between studies. The most frequently reported unmet needs were those in the activities of daily living domain (1–73%), followed by psychological (12–85%), information (6–93%), psychosocial (1–89%) and physical (7–89%). Needs within the spiritual (14–51%), communication (2–57%) and sexuality (33–63%) domains were least frequently investigated. Compared to any other time points of the cancer illness, the treatment phase exhibited the largest variation in the prevalence of unmet need for each domain. However, unmet needs were more likely to be found in a larger proportion of people post-treatment. Trends and predictors were also highly variable across all time points. Tumour specific unmet needs were difficult to distinguish and notably, there is a significant lack of evidence pertaining to any specific unmet needs of people with colorectal cancer. The review also highlights the difficulties in gauging levels of unmet need given the varied study questionnaires used and differences in the classification of unmet need reporting methods. Further, a lack of standardised approach to reporting of the instruments precludes comparison of data from different sources. Accurate measurement of unmet need is essential to be able to monitor patients as they progress through their cancer illness and identify where additional support is required. It is also imperative that within supportive care trials unmet need is reliably assessed to determine the true efficacy of an intervention. Therefore, given the issues with unmet need measurement identified in Chapter Four, Chapter Five sought to provide a systematic overview of instruments that measure this concept. Eighteen instruments were identified and were classified as generic, need, cancer type or stage of illness specific. Based on the descriptions and psychometric properties of each instrument, the following recommendations are made for the most optimal instruments to measure unmet need in people with cancer from each category of instrument; the Supportive Care Needs Survey was considered to be the best generic instrument to measure unmet need, whereas the Prostate Cancer Needs Questionnaire and Needs Evaluation Questionnaire were the best available tumour and stage of disease specific assessments respectively. Chapter Six reports the findings from an empirical study to quantify colorectal cancer patients’ supportive care needs following their discharge after cancer surgery. As post-discharge needs are managed by a senior colorectal cancer nurse co-ordinator, a review of these occasions of service provides a measure of the level of unmet need. Nursing records for consecutive patients discharged from a quaternary referral colorectal cancer surgical unit in Sydney were reviewed. All non-admitted patient occasions of service (NAPOOS) were identified. Of 521 eligible patients, 219 (42%) received 988 postoperative NAPOOS. These were largely delivered over the telephone (72%) or during face-to-face contact at outpatient appointments (26%). Most NAPOOS were recorded within the first two weeks following discharge but some occurred beyond 6 months. Overall, 1369 specific unmet needs were identified during these NAPOOS. Ongoing support, reassurance and monitoring were required for 186 patients (85%). Physical needs relating to wound care and bowel function were prevalent for 15-20%, information needs for 20% and assistance organising follow-up appointments for 36% of the sample. Predictors of unmet need in this patient group were varied. Older patients (>65 years) were significantly less likely to record a NAPOOS (OR 0.97, 95%CI: 0.96-0.99) whilst people with rectal cancer were significantly more likely to report multiple NAPOOS (OR 2.80, 95%CI: 1.60-5.01) and physical needs (3.56, 95%CI: 2.03-.27). Rehabilitation needs were more likely to be expressed by emergency department admitted patients (AOR 2.23, 95%CI: 1.07-4.65). After adjusting for admission status, patients residing in capital cities were significantly less likely to experiences need in this domain (AOR 0.37, 95%CI: 0.20-0.70). Given that ad-hoc nurse-delivered supportive care by telephone is currently provided as described in Chapter Six, systematisation of this type of intervention holds promise to reduce unmet needs in this patient group. Chapter Seven presents a systematic review of studies which have evaluated nurse-delivered supportive care interventions. Critical appraisal of studies was conducted using standardised instruments namely the CONSORT statement for randomised trials and STROBE statement for non-randomised studies. Of 2,451 potentially relevant articles, only 13 studies were found to be relevant to the research question, comprising five randomised controlled trials and eight nonrandomised studies. All studies demonstrated that it is feasible to deliver supportive care over the telephone for people in follow-up for cancer. Furthermore, using a nurse to deliver such interventions was found to be acceptable to patients resulting in a positive experience. Some studies have highlighted that potentially, significant improvements can be made to patients’ quality of life, satisfaction with care and unmet information needs, but definitive conclusions could be recorded due to the variable methodological quality and reporting characteristics. Chapter Eight describes a nurse-led telephone intervention, the ‘CONNECT’ intervention and presents a randomised pilot study. CONNECT is delivered by a cancer nurse over the telephone in the six months following patients’ discharge from hospital after surgery. This intervention was designed to supplement exist follow-up strategies. The content of each CONNECT call is standardised to address supportive care domains of unmet need. To evaluate this intervention, patients with colorectal cancer at Royal Prince Alfred Hospital, Sydney were randomised to receive the telephone intervention (n=39) or usual follow-up care (n=36). Feasibility and acceptability of this service delivery approach were assessed as a primary outcome. Patient reported primary outcomes assessed included unmet supportive care needs, quality of life, psychological distress and health service utilisation. Outcomes assessments were assessed at baseline then at one, three and six months post-discharge using a series of postal or telephone questionnaires. The consent rate to this trial was 86%. The CONNECT intervention was shown to be feasible and successful in identifying patient unmet needs. Between 84% and 95% of intervention group patients completed each CONNECT call with the nurse, where a range of unmet needs were identified. Feedback from participants about the intervention was extremely positive. The reassurance, support and sense that someone was monitoring individual needs during each call were seen as benefits of this service. Statistically significant differences between study groups in terms of all patient reported outcomes eluded this evaluation. Scores and changes scores for unmet need, quality of life, psychological distress and health service utilisation were comparable between groups. However scores favoured the intervention group. For example, at three months total SCNS-SF34 scores for intervention group was 98.1 and 110.0 for the control group. At six month total CaSUN scores were 10.0 (intervention) and 14.0 (control). Trends over time for these outcomes were also not statistically significantly different between groups. However some clinically relevant findings from this evaluation were demonstrated. Total FACT-C scores at six months showed that intervention group participants reported higher quality of life compared to control group participants (106.0 versus 98.6). Whilst not statistically different, the findings were in the expected direction and the difference (7.4) between the groups is clinically relevant. The most promising development from utilisation data was the difference, favouring the intervention group, resulting in a ≥10% reduction in presentations to emergency departments and readmission rates to hospital. Given this trial was a pilot study, post-hoc power was ascertained. This found that despite an a priori sample size calculation, the trial was underpowered. This may have resulted in a false negative result. Chapter Nine provides an overview of the key findings from the series of studies undertaken in this thesis. The implications of study results are explored in relation to policy, clinical practice and research.